Ha aft at ett Tab a8 At! Wh i WE LS a tie CMe ees Pee at H sy Vin’, ah Pe LETC RY a gin " Ba "a dh Se fs 2 rt, ‘ tanh aa wt 7 Ma 14 4e a ig rr iat 0d ait PCAN » Aoi ne ke Ws) nits Cre Speen ae M4 ay) ; f OR EN seat 4 aad ome ae i DEA Hr EAS poprins 8 as dio Na ah at 88 i ay ae eS Fath aA Te We MAINS OSD a eed ENN wat g bit SA aay di Fm Ra ge ge B66 on adh De Te BUA tiers pins si pi seat ARE Sher) ear aed nce WE SID vedere no aN Wirt sites 5 RSM gid FF vi. po SU di aNinn dost dian nt Sivas wlonaty tepans key Ue eee "Oa hs eel Swi ws wie ( OI ewe Nain “WmPnachihensaesyn poate Sash ey i SR Mm et are ph. dtd Wie Sing ol abuso ) EA ALE AT rai Taree Cl : Ui 4 wt Mk qym as : i i maw, Bind wla sa gabe ATA BA Mw 1 r = Ny ‘ ana ae ANE) nt Uns abe ta ay wh ate ey ar Peers ie Sythe Hai gh Cn as Sie ako TTS ye mA AT go Bane ete ale re iGo peta Be tf eat ite te fe UN was ae reat, LeREL ey Se isiit tps een SOS lee meee at IM ae Eimeria PG ow ere Misty ey tpt or Seite Be Sa RU a eige ye. eee ate Some. ime Sigtaisiemti = 4 areuny eu) 3 SUP eR aK, Sm Aart ys Fh ge i With Sey er De ee many f th % fis i H 1 i i i “fyegl An WATSONIA THE JOURNAL OF THE BOTANICAL SOCIETY OF THE BRITISH ISLES VOLUME 27 EDITED BY M. BRIGGS, J.R. EDMONDSON, M.J.Y. FOLEY, D.L. KELLY, D.R. McKEAN, M.S. PORTER, M.N. SANFORD 2008-2009 PRINTED FOR THE SOCIETY BY HENRY LING LIMITED, THE DORSET PRESS, DORCHESTER, DORSET DT1 1HD PUBLISHED AND SOLD BY THE BOTANICAL SOCIETY OF THE BRITISH ISLES, C/O DEPARTMENT OF BOTANY, THE NATURAL HISTORY MUSEUM, LONDON 5W7 SBD i ea i i { a { i a y i | 1 _— 7 aA a me See ei S's 's eo ss SoS Ss Sos es S'S S's) Sle S's a2 78 82 83 19 190 IDS 22) 226 Me Ma 249 250 US) 253 Zi) 311 34] 329) 313) S55) 335 335 355) 338 338 338 338 338 Jol) 360 361 > MZ OO, line 3 up, col. | table line 11 up table line 15 table line | line 15 ne ZS line 16 up line 13 up line 17 up line 14 up line 14 up line 12 up line 7 up line 18 table line 3 table line 3 up ines WH line 8 up DATES OF PUBLICATION Part 1, pp. 1-98, 11 February 2008 Part 2, pp. 99-206, 4 September 2008 Part 3, pp. 207-282, 16 February 2009 Part 4, pp. 283-392, 17 August 2009 table line 13 up For: line 2 For: line 11 For: line | For: ime col. 2 For: line 18, col. 1 For: nmmexon col 2 For: fig. caption For: line tS up, col. I For: lines2. col. I For: fig. caption For table line 3 up For line 8 up, col 2 For table line 5 up For ane 3) Wyo), Cole, For Errata caespitosa read: cespitosa kamschaticum read: kamtschaticum sylvatica read: silvatica M. x smithii read: M. smithii spirodella read: spirodela artemisifolia read: artemistifolia spp. read: subsp. proctoris read: proctoriana proctoris read: proctoriana proctoris read: proctoriana : proctoris read: proctoriana : rivinianna read: riviniana : Petastites japponicus read: Petasites japonicus : sylvatica read: sylvestris : erectus read: erecta bottom line, col. 2 — For: Hypochoeris read: Hypochaeris : S. vagensis read: S. x vagensis : Buddleia read: Buddleja : Buddleia read: Buddleja : Buddleia read: Buddleja : Hypocharis read: Hypochaeris : capraea read: caprea : hordaceus read: hordeaceus : Cystisus read: Cytisus : Rumexs read: Rumex : hordaceus read: hordeaceus : sivatica read: silvatica : capraea read: caprea : Elytrigia canina read: Elymus caninus : CARDUUS read: CIRSIUM : Carduus heterophyllus read: Cirsium heterophyllum For: idaeaus read: idaeus Insert: *64, before: *Mid-W. For: Sisyrhinchium read: Sisyrinchium ill a= ue | AC tin oe aes w m .; INDEX WATSONIA VOLUME 27 (2008-09) Prepared by C. R. Boon Abbott, P.P. - Plant atlas of mid-west Yorkshire (Bk Rev.) 195 Abies alba (v.c. H12) 74; grandis (v.c. 7) 381; procera (v.c. 90) 381 Abutilon theophrasti (v.c. 44) 76 Acacia dealbata (v.c. 10) 384; anserinifolia (v.c. 110) 383; caesiiglauca (v.c. 81) 384; inermis (v.c. 69) 266; novae-zelandiae (v.c. 52) 78; ovalifolia (v.c. 38) 177, (v.c. 62) 266 Acer 196; cappadocicum (v.c. 52) 267; (v.c. 69) Volemeplatanoides. 334, 335, 337; pseudoplatanus 47,164, 251, 252, 327, 329, 331, 334, 337; saccharinum (v.c. 62) 267, [(v.c. 62)] 384; saccharum 384 Achillea millefolium 164, 249, 334, 335, 337, 338; nobilis (v.c. 29) 85; ptarmica 132 Acorus 275 Actaea erythrocarpa (v.c. 81) 382; spicata 30, Se 133 Aegopodium podagraria 252, 334, 335, 337 Aesculus 196; carnea (v.c. 69) 81; hippocastaneum 164; indica [(v.c. 69)] 81, (v.c. 63) 179 Aethusa cynapium 334, 335, 337, subsp. agrestis (v.c. 10) 378, (v.c. 64) 378 xA gropogon littoralis (v.c. 63) 263 Agropyron littorale var. setigerum 369 Agrostis canina subsp. capillaris 32, subsp. montana 32; capillaris 150, 153-155, 163, 164, 248, 249, 252, 254, 255, 334, 335, 337, 338, 343, 350; capillaris x stolonifera (v.c. 48) 87, (v.c. 44) 263; curtisii (v.c. 45) 87; gigantea (v.c. 100 (Bute)) 186; lachnantha (v.c. 64) 386; xmurbeckii (v.c. 48) 87, (v.c. 44) 263; stolonifera 54, 123, 124, 167, 327, 329-331, 334, 337, 342, 347; stolonifera x Polypogon monspeliensis (v.c. 63) 263 Aira caryophyllea 164; praecox 155, 164; setacea Hudson (Poaceae). Neotypification of, 239-242, (illus.) 241 Ajuga xpseudopyramidalis (v.c. 106) 82, reptans X pyramidalis (v.c. 106) 82 Akeroyd, J.R. - Obit. of John (‘Jack') Gregory Hawkes (1915-2007) 199-200 Alchemilla 138, 315, 319, 320; acutiloba 315- Pl oOo I alpina 321; conjuncta (vic. 44) 266, 321; filicaulis subsp. filicaulis Sy Ieeesubsp.| vestia 320 Velabra 321, glaucescens 321; glomerulans 315-319, 321; micans 321; minima 321; mollis (v.c. G03) ss. SAI; 3334s 33), S74. Behe monticola 315-317, 319, 321; subcrenata (v.c. 67) 177, 315-317, 319, 321; tytthantha 321; venosa 321; vulgaris L. agg. and other hay-meadow species in Northern England since the 1950s. The decline of Lady's- mantles 315-321; wichurae (v.c. 106) 78, 315-319, 321; xanthochlora 138, 321 Alisma lanceolatum (v.c. 100 (Great Cumbrae)) 185, (v.c. 51) 263 Allen, D.E. - British species of Rubus L. (Rosaceae) in Maine and North Brittany, France 243-244 Allen, D.E. - Series reassignment of Rubus cerdicii 72 Alliaria petiolata 252, 328, 330, 334, 335, 337 Allium ampeloprasum var. ampeloprasum (v.c. H12) 88, var. babingtonii (v.c. 10) 387; carinatum (v.c. 10) 387; cepa (v.c. 64) 387; oleraceum 337, 338; paradoxum (v.c. 35) 88, (v.c. H12) 88, (v.c. 100 (Bute)) 186; porrum (v.c. 45) 271; roseum (v.c. 85) 186; schoenoprasum (v.c. 64) 264; subhirsutum (v.c. 41) 88, (v.c. 62) 271; triquetrum (v.c. 7) 186, (v.c. 48) 271, (v.c. 69) 271; ursinum 3, PO NGI SV SOE S85, S87) Alnus cordata x incana (v.c. 29) 173; glutinosa XTINCANG (Vics 29) IS eluninosa 125, 232: DO 334s 339, 301 xhybrida (vc. 29) N13: rubra (v.c. 59) 173, (v.c. 62) 264 Alopecurus xbrachystylus’ (v.c. 63) 263; borealis (v.c. 65) 186; bulbosus 167; myosuroides 186; pratensis 334, 337; pratensis X geniculatus (v.c. 63) 263 Alstromeria aurea (v.c. 70) 272 Althaea cannabina (v.c. 29) 174 Alyssum saxatile 334, 335, 337, 338 Amaranthus albus (v.c. 10) 382; bouchonii (v.c. 10) 382; caudatus (v.c. 38) 173; cruentus (v.c. 31) 382; hybridus (v.c. 100 (Bute)) 173, (v.c. 31) 265; retroflexus (v.c. 91) 382 Ambrosia artemisiifolia (v.c. 45) 85, 88, (v.c. 91) 85, (v.c. 67) 184, 190, (v.c. 90) 386 Amelanchier lamarckii (v.c. 41) 79 Ammi majus (v.c. H12) 81 Ammophila arenaria 150, 154, 155, 164 Amsinckia micrantha (v.c. 44) 82, (v.c. 46) 268 Anacamptis 247; morio x Orchis mascula 247 xAnacamptorchis 247; morioides (Brand) Stace, comb. nov. 247 393 394 Anagallis arvensis 164; minima 143, 144, (v.c. 85) 175 Anchusa azurea (v.c. 52) 268; vulgaris 164 Andromeda polifolia 132, 133 Anemanthele lessoniana (v.c. 91) 88, (v.c. H12) xo, Ovdes LU) Zl, (ics (OD) ZI Cres W7/) Dan Anemone apennina (v.c. 7) 382; blanda (v.c. 42) 74, (v.c. 46) 264; hupehensis x vitifolia (v.c. H12) 74; xhybrida (v.c. H12) 74; nemorosa 32, 252 Angelica archangelica (v.c. 13) 179; 252, 334, Sh GOT Anisantha diandra (v.c. 69) 271, (v.c. 90) 387; madritensis (v.c. 67) 186; sterilis 334, 335, 3875338 Anoda cristata (v.c. 29) 76, var. brachyantha 76 Antennaria dioica 249 Anthemis arvensis (v.c. 100 (Bute)) 184; austriaca (vec. 17); 2705 (Gc. 10) 386; punctata subsp. cupaniana (v.c. 62) 270; tinctoria (v.c. 46) 270 Anthericum 275 Anthoxanthum odoratum 123, 155, 164, 254, 33340 Sd), Sos Solon yo) Anthriscus sylvestris 252, 334, 335, 337, 338 Anthyllis vulneraria 150, 155, 164, 249 Antirrhinum majus 332, 334, 335, 337, 338 Apera spica-venti (v.c. 48) 263 Aponogeton distachyos (v.c. 35) 270 Aquilegia vulgaris (v.c. 110) 172, 334, 335, 33/88 Arabidopsis 275; thaliana 330, 334, 335, 337 Arabis caucasica 328, 330, 337; hirsuta 337, 338; petraea (v.c. 91) 377 Aralia elata (v.c. 58) 384 Araucaria araucana (v.c. 90) 381 Arctium minus 334, 335, 337 Arctostaphylos uva-ursi 138 Arenaria ciliata subsp. hibernica 249; norvegica 248, 249, subsp. anglica 248, 249, subsp. norvegica in Ireland. The re- discovery of, 248-250; serpyllifolia 155, 163, 164, 337, 338 Argentina egedii agg. 54 Armeria maritima 54,123, 249 Aronia melanocarpa (v.c. 70) 79 Arrhenatherum elatius 32, 164, 327, 329, 331, 334, 337 Artemisia abrotanum (v.c. 85) 184; vulgaris 8)04213)2)9)5 219) INDEX TO WATSONIA VOLUME 27 Arum 279; italicum (v.c. 67) 185, (v.c. 80) 386, subsp. italicum (v.c. 57) 86, (v.c. 31) 386; maculatum (v.c. 110) 86, 379; maculatum x italicum (v.c. 113(S)) 379 Aruncus dioicus (v.c. 62) 266 Asarina procumbens (v.c. 81) 385 Ashton, P.A., with M. Dean et al. - Description, ecology and establishment of Carex salina Wahlenb. (Saltmarsh Sedge) - a new British species 51-57 Asparagus officinalis 155, 164 Asperula cynanchica 248, 249 Asplenium 275, 332; adiantum-nigrum 334, 335, 337, 338; marinum 124; ruta-muraria 114, 326, 327, 329-332, 334, 337; trichomanes 114, 329; 330) 33253345 3855050 Aster lanceolatus 334, 337; novi-belgii (v.c. H12) 85, 334, 335, 337; tripolium 54 Astilbe xarendsii (v.c. 63) 176; japonica x rosea X chinensis (v.c. 63) 176 Athyrium filix-femina 334, 335, 337, 338 Atriplex xgustafssoniana (v.c. 13) 173; halimus (v.c. 59) 173, (v.c. 70) 264; patula 334, 337; portulacoides (v.c. 64) 260; prostrata x longipes (v.c. 13) 173 Aubretia deltoidea (v.c. H12) 77, 328, 330, 337 Avena fatua (v.c. 100 (Bute)) 186; sativa (v.c. 110) 386 Avenella flexuosa 239 Azara microphylla Hook. f. (v.c. 29) 76 Azolla filiculoides (v.c. 100 Arran)) 172 Bacopa ‘Snowflake’ 181 Bailey, J.P., Kay, Q.O.N., McAllister, H. & Rich, T.C.G. - Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles 69-72 Ballota nigra 334, 335, 337 Bartramia pomiformis 32 Bassia scoparia (v.c. 7) 382, (v.c. 8) 382 Bateman, R.M. & Sexton, R. - Is spur length of Platanthera species in the British Isles adaptively optimized or an evolutionary red herring? 1-21 Beesley, S. - County Antrim, scarce, rare and extinct vascular plant register (Bk Rev.) 194 Bellis perennis 154, 248, 249, 334, 335, 337, 342, 350 Berberis aggregata (v.c. 110) 172; darwinii (v.c. 91) 75, ~.c. 38) 172 eae (v.c. 62) 264, (v.c. 10) 382; julianae (v.c. 57) 74, (v.c. 38) 172, (v.c. 44) 264; INDEX TO WATSONIA VOLUME 27 thunbergii (v.c. 69) 74, (v.c. 91) 74, (v.c. 46) 264, (v.c. 62) 264, var. atropurpurea 74; verruculosa (v.c. 77) 75; wilsoniae (v.c. 44) 264; wilsoniae x aggregata (v.c. 70) 264 Bergenia xschmidtii (v.c. 46) 265; crassifolia x ciliata (v.c. 46) 265 Betula 47; ermanii (v.c. 51) 264; lutea (v.c. 63) 382; nana (v.c. 78) 75; pendula 164; pubescens 29, 32-34, 124, 125, 232, 334, Bo5soo/5 338, Subsps foriuosa (vic. 38) 173, (v.c. 110) 173, (v.c. 63) 376 Biscutella auriculata (v.c. 38) 175 Blackstock, T.H. & Jones, P.S. - Obit. of David Paul Stevens (1958-2007) 95-97 Blechnum spicant 124, 252 Blindow, I, with H. Schubert (eds) - Charophytes of the Baltic Sea (Bk Rev.) 192-193 Blysmus compressus 192, 194, (L.) Panz. ex Link on the Sefton Coast sand-dunes, Merseyside. Distribution, status and ecology of, 339-353, (map) 344; rufus 54, 351 Bolboschoenus maritimus 343 Bonner, I.R. - Anglesey (v.c. 52) county rare plant register. Flowering plants and ferns (Bk Rev.) 194 Book Reviews 189-198, 273-276, 389-392 Borago officinalis (v.c. 110) 180 Brachyglottis compacta x laxifolia (v.c. 93) 386; ‘Sunshine’ (v.c. 93) 386 Brachypodium pinnatum (v.c. 42) 263; sylvaticum 32, 47, 124, 252, 254, 334, 335, 337) Bradshaw, M.E. - The decline of Lady's- mantles (Alchemilla vulgaris L. agg.) and other hay-meadow species in Northern England since the 1950s 315-321 Bradshaw, Tony, FRS FLS (1926-2008) (Obit.) 282 Brassica napus subsp. oleifera (v.c. 44) 265; oleracea (v.c. 113(S)) 77 Briza maxima (v.c. 41) 87, (v.c. 8) 185, (v.c. 49) 270, (v.c. 69) 270; media 32, 33, (v.c. 100 (Arran)) 185, 249, 254; minor (v.c. 10) 380 Bromopsis erecta 114, 253; ramosa 334, 335, 331, 3308) Bromus arvensis (v.c. 10) 186, (v.c. 13) 186; commutatus (v.c. 85) 186; hordeaceus 164, 334, 335, 337, 338, subsp. thominei (v.c. 10) 380; racemosus (v.c. 45) 87, (v.c. 81) 380; secalinus 186, (v.c. 39) 186 395 Brummitt, R.K., with V.H. Heywood et al. - Flowering plant families of the world (Bk Rev.) 196-197 Brunnera macrophylla (v.c. 57) 82 Bryum 391 Buddleja davidii 114, 115, 327, 329, 331, 334, Bow Bull SA aie. ihe babingtonianus (Rosaceae) 245-246 Bunias orientalis (v.c. 63) 175 Bupleurum longifolium (v.c. rotundifolium 391 Butomus umbellatus (v.c. 44) 86 Buxus sempervirens (v.c. 91) 378 Rubus Watson Status ol: WEGCaAR® 38) 179; Calamagrostis canescens 91; deschampsioides 54; epigejos (v.c. 106) 87; stricta 91 Calendula arvensis (v.c. 48) 85; officinalis (v.c. 48) 85, 334, 337 Calla palustris 380 Calliergonella cuspidata 232 Callitriche 246; brutia 246, subsp. hamulata (Kitz. ex W. D. J. Koch) Stace, comb. nov. 246, var. hamulata 246, (v.c. 51) 262; hamulata 246; intermedia subsp. hamulata 246; platycarpa (v.c. 79) 378; truncata WECRSZ)LO2s(VEG. 15) 378 Calluna 29; vulgaris 33, 124, 155, 249 Calypogeia muelleriana 32 Calystegia xlucana (v.c. H12) 82; pulchra (v.c. 113(S)) 385; sepium 334, 335, 337, subsp. roseata (v.c. 31) 262; sepium x sylvaticum Wecs HZ tS2 sil vatiea 334-335" 55) 5358; S49) (ec. 79) S85) SUDSpaaUsiuncia (Vic: 110) 180, (v.c. 44) 268 Camelina sativa (v.c. 69) 260 Campanula alliarjifolia (v.c. 63) 385; carpatica (v.c. 29) 83; cochlearifolia (v.c. 59) 181; glomerata 254; medium (v.c. 46) 269; DGG Ween 29) p SIE 34S O5h Oo: DeTSiCiolian \(Wece 448) 7 2695" 337, 338: portenschlagiana (v.c. H12) 83, (v.c. 85) SCC SS) AA OOMN SSA SDS OO8, (v.c. 81) 385; poscharskyana (v.c. 48) 269, (v.c. 70) 269; rapunculus 334, 335, 337; rotundifolia 32, 124, 155, 249, 325, 334, 3395 SIT Campylium stellatum 32 Canna 275 Cannon, John Francis Michael (1930-2008) (Obit.) 279-281 Capsella bursa-pastoris 334, 335, 337, 338 396 Cardamine 252; amara 252; bulbifera (L.) Crantz (Brassicaceae) in Mid-west Yorkshire (v.c. 64). Water-borne dispersal of, 250-253, (map) 251, f. bulbifera 250, f. ptarmicifolia 250; corymbosa (v.c. 41) 77, ne SD) 71, Wes DT) WIS (ee. SY) ITS? flexuosa 2525334, 335, 337, 338; hirsuta 329%, 3305) 33453375 umpatiens 730, Si: pratensis 342, 357, 360; raphanifolia (v.c. 48) 77, 250, 252 Cardaminopsis 275 Carduus nutans (v.c. 110) 379; tenuiflorus (v.c. 29) 83 Carex 5S¥; sect. Phacoeystis 52> (key) 53» 94: acuta 185; acuta x nigra [(v.c. 69)] 185; acutiformis 350; acutiformis X acuta (v.c. 17) 263; acutiformis x riparia (v.c. 29) 87; aquatilis 55; arenaria 150, 154, 155, 163, 164, 342; atrata 192; buchananii (v.c. 29) 87 “canescens, “I9iIky Vcapillanis 192: caryophyllea 249; comans (v.c. 17) 386; xcsomadensis (v.c. 29) 87; curta 191; xdecolorans (v.c. 46) 87; diandra (v.c. 43) 263: digitata’ (WiC. 457), 87s. distans: x viridula subsp. viridula (v.c. 44) 87; disticha 343, 347, 350; echinata 167; xelytroides [(v.c. 69)] 185; ericetorum 133; flacca 124, 248, 249, 254, 255, 342, BAGS 345, 3502 flava? SIRE Ose vale oedocarpa 167; xfulva (v.c. 38) 185; hirta 342; hostiana x viridula (v.c. 38) 185; lasiocarpa’ 234, D36,.(wecs +538)" 580; leporina 191; limosa 234, 235; xluteola (v.c. 44) 87; lyngbyei x nigra 54; lyngbyei x subspathacea 54; mackenziei 54; maritima (v.c. 110) 86, (v.c. 110) 380; montana 95; muricata subsp. lamprocarpa 191, subsp. muricata 30, 195, subsp. pairae 191; muricata x divulsa (v.c. 10) 380: nigra. S295, 07, IOI 343. era x bigelowii (v.c. 46) 87; otrubae x divulsa subsp. divulsa (v.c. 10) 185; otrubae x remota (v.c. 59) 185; ovalis 191; paleacea 51-55; panicea 167, 350; paniculata (v.c. 48) 263; parallela 191; pendula 252, 334, 335, 337, (v.c. 90) 380; xpseudoaxillaris (vic. 59) 1853 pulicaris 249 srectara2y 53) 55, (v.c. 106) 87; remota 334, 335, 337; riparia 167; riparia Xx vesicaria (v.c. 29) 872 rostrata. 52 (cs 29) 185. 23255236: salina Wahlenb. (Saltmarsh Sedge) - a new British species. Description, ecology and establishment of 51-57, 191; xsooi (v.c. 29) I SeMeKeUUS MCC, 17). AOS subspathacea 51-55, 191; trinervis 191; vacillans 52, 53, 55; vaginata (v.c. 65) 185; viridula subsp. brachyrrhyncha (v.c. 47) 263, (v.c. 10) 380, subsp. viridula 343, 347 INDEX TO WATSONIA VOLUME 27 Carpobrotus edulis (v.c. 113 (S)) 173 Carrichtera annua (v.c. 81) 383 Cassinia fulvida 86; leptopylla (v.c. H12) 86 Catapodium rigidum 334, 335, 337, 338 Cayouette, J., with M. Dean er al. - Description, ecology and establishment of Carex salina Wahlenb. (Saltmarsh Sedge) - a new British species 51-57 Ceastium tomentosum 328 Cedrus deodara (v.c. 29) 74 Centaurea cyanus (v.c. 113(S)) 379; jacea (v.c. 38) 182, (v.c. 63) 269; xmoncktonii (v.c. 35) 262, (v.c. 14) 379; nigra 164, 254, 255; 334, 335, 337, 3385 3483 lenamex Jacea (v.c. 35) 262, (v.c. 14) 379 Centaurium erythraea 155, 163, 164, 343 Centranthus ruber 65, (v.c. 91) 83, 326-330, 3325 3845337 Cephalanthera longifolia 357, 360 Cephalaria gigantea (v.c. 81) 386 Cerastium arctium 367; arvense (v.c. 52) 75; fontanum 164, 334, 335, 337, 338, 342; glomeratum 334, 335, 337, 338; pumilum 283, (v.c. 8) 376; semidecandrum 342; tomentosum (v.c. 113(S)) 174, 330, 334, 3355. 937) Ceratocapnos calviculata (v.c. 29) 75 Ceratochloa carinata (v.c. 7) 186, (v.c. 38) 271, (v.c. 45) 271; cathartica (v.c. 85) 186 Ceratophyllum demersum 129 Cercis siliquastrum (v.c. 29) 178, (v.c. 17) 267 Cerinthe major (v.c. 17) 268; minor (v.c. 46) 268 Ceterach 275; officinarum (v.c. 91) 74, (v.c. 67) 1725332 Chaenorhinum origanifolium origanifolium (v.c. 35) 269 Chaerophyllum temulentum 334, 335, 337, 338 Chamaecyparis lawsoniana (v.c. H12) 74 Chamerion angustifolium 164, 327, 329-332, 334 33i7/ subsp. Chara aspersa var. subinermis 193; baltica 193; contraria 193; intermedia 193; virgata 129; vulgaris 129, 193 Charles, C.A., with T.C.G. Rich et al. - The diversity of Sorbus L. (Rosaceae) in the Lower Wye Valley 301-313 Chater, A.O. - Rev. of The flora of Rum. An Atlantic island reserve (by D.A. Pearman, C.D. Preston, G.P. Rothero & K.J. Walker) 197-198 Chelidonium majus (v.c. 100 (Arran)) 173, BBs) SOT) INDEX TO WATSONIA VOLUME 27 Chenopodium 246; xbontei 246; carinatum x cristatum 246; xchristii 246; ficifolium (Wey ESO) 60-5 slaucumaW.c.4° 7x 173; polyspermum (vy.c. 100 (Bute)) 173; pumilio x carinatum 246; quinoa (v.c. 8) WSs CuEGs65) 17S: rubrum (v:c. 91) 75; urbicum (v.c. 77) 264 Chiapella, J.O. - Neotypification of Aira setacea Hudson (Poaceae) 239-242 Chiastophyllum oppositifolium (v.c. 60) 383 Chionodoxa forbesii (v.c. 90) 387; luciliae (v.c. 63) 387; sardensis (v.c. 59) 186 Cicendia filiformis 143 Cichorum intybus (v.c. 91) 379 Cinclidium stygium 235 Circaea lutetiana 252 Cirsium 256: acaule 255, 256: arvense 88,164, DAD ASD. 931. 330; 342: xcelakov- skianum (v.c. 49) 84, 256; dissectum 167, 350; xforsteri 256; heterophyllum 31, 32, Bib OomNO. Soi. 500°x 337; setacea 239, (map) 240, (llus.) 241, 242, (v.c. 110) 380 Deutzia scabra (v.c. 17) 265 Dianthus carthusianorum (v.c. 90) 382; caryophyllus 334, 337; gratianopolitanus 283, 299 Dicentra formosa (v.c. 90) 382; spectabilis (v.c. 63) 173, (v.c. 64) 382 Dicranum bonjeanii 32; majus 32; scoparium INDEX TO WATSONIA VOLUME 27 Digitalis purpurea 123, 330, 334, 335, 337, 359, 360 Digitaria ciliaris (v.c. 46) 88 Diphasiastrum complanatum (v.c. 81) 376 Diplophyllum albicans 32 Dipsacus fullonum (v.c. 100 (Bute)) 182, 334, 335, 337, 338; laciniatus (v.c. 29) 83, (v.c. 67) 83 Doronicum columnae_ (v.c. 13) 386; pardalianches 250, 252; hirsutum (v.c. H12) 79 Draba 367; muralis (v.c. 90) 377 Dransneldes I — Rev of Flora of Monmouthshire (by T.G. Evans) 189-190 Drosera 146; xobovata (v.c. 67) 174; rotundifolia x anglica (v.c. 67) 174 Dryas 248; octopetala 249 Dryopteris affinis subsp. affinis (v.c. 63) 172, subsp. borreri (v.c. 63) 172, subsp. cambrensis (v.c. 63) 172, subsp. paleaceolobata (v.c. 110) 376; carthusiana x dilatata (v.c. 49) 74; xcomplexa (v.c. 64) 260, (v.c. 81) 376, nothosubsp. complexa (v.c. 46) 74; xdeweveri (v.c. 49) 74; filix-mas 32, 326- 331, 334, 337; filix-mas x affinis (v.c. 64) 260, (v.c. 81) 376; filix-mas x _ affinis subsp. affinis (v.c. 46) 74 Duchesnea indica (v.c. 69) 266 xDupoa labradorica 54 Duvigneaud, J., with J. Lambinon & L. Delvosalle - Nouvelle flore de la Belgique, du Grand-Duché de Luxembourg, du nord de la France et des régions voisines (Ptéridophytes et Spermatophytes. 5th ed. (Bk Rev.) 275 Dysphania 246; xbontei (Aellen) Stace, comb. nov. 246; carinata x cristata 246; xchristii (Aellen) Stace, comb. nov. 246; pumilio x carinata 246 Echinochloa colona (v.c. 46) 88; crus-galli 76; frumentacea (v.c. 44) 88 Echinops bannaticus (v.c. 85) 182, (v.c. 90) 386; exaltatus (v.c. 85) 182, (v.c. 77) 269; sphaerocephalus 182 Echium vulgare 164 Edmondson, J. - Obit. of Tony Bradshaw FRS FLS (1926-2008) 282 Edmondson, J. - Rev. of Mabberley's plant- book: a portable dictionary of plants, their classification and uses (by D.J. Mabberley) 390-391 Egeria densa |(v.c. 80)] 86 399 Eichhornia crassipes (v.c. 57) 88, (v.c. 60) 186 Elaeagnus macrophylla (v.c. 63) 384; multiflora (v.c. 41) 80; umbellata (v.c. 57) 80, (v.c. 44) 267 Eleocharis austriaca 191; mamillata 191, subsp. austriaca 191; multicaulis 233; palustris 54, 191, 342, 350; quinqueflora 342, 347; uniglumis 54, 350 Eleogiton fluitans (v.c. 29) 86, (v.c. 91) 86, 191 JD WS: COPM|IS BA, - 224 Bo 4 Sb Serer pycnanthus var. setigerus 369 Elytrigia 369; xacuta subsp. obtusiuscula (v.c. 44) 264; atherica Kerguélen f. setigera (Dumort.) Stace, comb. nov. 369; atherica x juncea (v.c. 44) 264; juncea 51; repens SSS OO Atos O45 509, La aristata 369 Epilobium 65, 66, 252; xbrevipilum (v.c. 38) 261; brunnescens (v.c. 7) 384; ciliatum BYV.ON O07 329. 98.05.3534. 330: ciliatum x palustre (v.c. 59) 179; xdacicum (v.c. 44) 80, (v.c. 52) 179, (v.c. 38) 261; xerroneum (v.c. 10) 178; xfloridulum (v.c. 52) 80, @Ace=52) I 9s xfossicolay(vECy SD) 19: xhaussknechtianum (v.c. 63) 179; hirsutum ©5, 334; 335, 337, 343, 346; hirsutum xX montanum (v.c. 10) 178; hirsutum ~x_ palustre (v.c. 44) 261; hirsutum X parviflorum (v.c. 44) 80, (v.c. AG) O05 (Veco SS) ee ZOE whinsutum x tetragonum 65, (v.c. 38) 261; xinterjectum (v.c. 44) 80, (v.c. 38) 261; xlamotteanum (v.c. 44) 261; lamyi 65, 66; lanceolatum (v.c. 50) 80; lanceolatum x obscurum (v.c. 44) 261: xlimosum (v.c. 44) 80, (v.c. 58) 378; mentiens (v.c. 38) 262: montanum 326, 327, 329-331, 334, 337; montanum x ciliatum (vic. 44) 80, Gc. 38) 261: montanum xX roseum (v.c. 63) 378; montanum xX tetragonum (v.c. 63) 179; xmutabile (v.c. 63) 378; xnutantiflorum (v.c. 63) 378: obscurum 343; obscurum xX ciliatum (v.c. 91) 80; xpalatinum (v.c. 10) 7D SanWeCs 5S) @ 2Olespalusine. 343: parviflorum 334, 337; parviflorum x Cillatuinniua (VECs 52) SOs AWcamoZ)ily 179: parviflorum x montanum (v.c. 44) 80, (v.c. 58) 378; parviflorum x obscurum (v.c. 44) SOM (WECHIS2) IO (vic 38) ZO: parviflorum x palustre (v.c. 44) 80; parviflorum x tetragonum (v.c. 10) 179, (v.c. 38) 261; xrivulare (v.c. 44) 80; RONgiin < Cilio Ones WS) Sihse XMISCMMVODSCUUMD AVEC. 23S)e) 262); xsubhirsutum (v.c. 44) 80, (v.c. 46) 80, (v.c. 38) 261; tetragonum 65, 66, subsp. tetragonum 65, 66, fettragonum subsp. 400 tournefortii naturalised in Britain 65-67, (v.c. 29) 179; tetragonum x ciliatum (v.c. 38) 262; tetragonum x obscurum (v.c. 38) 262; xvicinum (v.c. 91) 80; xwaterfallii (v.c. 44) 261 Epimedium alpinum (v.c. 46) 75 Epipactis atrorubens 137; atrorubens x helleborine (v.c. 69) 89; dunensis 164, 274; helleborine 137; palustris (illus.) 135, 136, 342, 347; phyllanthes (v.c. 64) 380; sancta 274; xschmalhausenii (v.c. 69) 89 Equisetum 367; arvense 155, 164, 334, 335, 337, 338, 343; arvense x palustre (v.c. 49) 73, (v.c. 10) 172; xbowmanii (v.c. 106) 73; xdycei (v.c. 100 (Arran)) 172, (v.c. 100 (Bute)) 172, (v.c. 110) 172, (v.c. 17) 376; fluviatile 233, 343, 345; fluviatile x palustre (v.c. 100 (Arran)) 172, (v.c. 100 (Bute)) 172, (v.c. 110) 172, (v.c. 17) 376; xlitorale 347; palustre 342; pratense 32; xrothmaleri (v.c. 49) 73, (v.c. 10) 172; sylvaticum (v.c. 14) 376; sylvaticum x telmateia (v.c. 106) 73; variegatum 342, 347 Eragrostis minor (v.c. 29) 88 Erica ciliaris 276; cinerea 124 Erigeron acer 65, 164; glaucus (v.c. H12) 85, (v.c. 38) 184, (v.c. 63) 386; karvinskianus (v.c. 85) 184, 337, 338, (v.c. 60) 386; speciosus (v.c. 62) 85 Erinus alpinus (v.c. 38) 181, 328, 330, 332, 337 Eriocaulon aquaticum 233 Eriophorum angustifolium 231, 343; brachyantherum 192; gracile Koch ex Roth (Cyperaceae), Slender Cotton-grass, in Ireland. The _ distribution § and conservation of, 229-238, (map) 232, (map) 233; latifolium 192; opacum 192; vaginatum 95, 137, 139 Erodium xXanaristatum 164; cicutarium 164; lebelii 163, 164; maritimum (v.c. 10) 378 Erophila glabrescens (v.c. 63) 175, (v.c. 48) 265; majuscula (v.c. 29) 77; verna 337, 338 Eruca vesicaria (v.c. 10) 383, subsp. sativa (Aes Se AGS) Eryngium campestre (v.c. 29) 179 Erysimum cheiranthoides (v.c. 51) 260, 332, 3843350574638 Erythronium dens-canis (v.c. 62) 272 Escallonia xlangleyensis (v.c. 38) macrantha x virgata (v.c. 38) 175 Eschscholzia californica (v.c. 44) 75, (v.c. 62) TUS OTEM 804) Wd5 Wes 1) SS Euonymus europaeus 133; fortunei (v.c. 64) 267; japonicus (v.c. 63) 384 W/ae INDEX TO WATSONIA VOLUME 27 Eupatorium cannabinum 124 Euphorbia amygdaloides 326, 334, 335, 337, 338, subsp. robbiae (v.c. 67) 80, (v.c. 7) 179; (wec. 38) 207; | characias ssubsp: characias (v.c. 41) 80, (v.c. 62) 267, subsp. veneta (v.c. 62) 267; cyparissias (v.c. 10) 384; dulcis (v.c. 44) 267; esula 337, 3383 helioscopia’ 334) 335smesIE maculata (v.c. 29) 80; mellifera (v.c. 41) 80, (v.c. 113(S)) 384; myrsinites (v.c. 62) 80; oblongata (v.c. 41) 80, (v.c. 57) 80, (v:c.' 7) 179; W.c. 38) 2672 palasonsa@wze 69)) 80; peplus 334; 33855337 heeSo- portlandica 164; xpseudovirgata (v.c. 51) 80, 261; serrulata (v.c. 39) 179, 189, (v.c. 44) 262; thymifolia (v.c. 7) 384; waldensteinii (v.c. 39) 179; waldsteinii x esula (v.c. 51) 80 Euphrasia 198; arctica subsp. arctica (v.c. 70) 83; arctica x foulaensis (v.c. 110) 181; arctica xX nemorosa (v.c. 100 (Great Cumbrae)) 181; confusa x micrantha (v.c. 69) 83; heslop-harrisonii (v.c. 110) 378; marshallii x scottica (v.c. 110) 378; nemorosa 342; nemorosa Xx confusa (v.c. 17) 262, (v.c. 48) 262; nemorosa x confusa xX micrantha (v.c. 62) 83; nemorosa xX micrantha (v.c. 81) 378; ostenfeldii (v.c. 110) 83; rostkoviana (v.c. 81) 378 Eurhynchium striatum 30, 32 Evans, T.G. - Flora of Monmouthshire (Bk Rev.) 189-190 Evans, T.G. - Monmouthshire county rare plant register (Bk Rev.) 194 Fagopyrum.tataricum (v.c. 62) 75 Fagus 305; sylvatica 33, 48, 79, 125, 252; 307, BU S34, 3354 39)7/5 BIC Fallopia baldschuanica 334, 335, 337; xbohemica (v.c. 70) 75, (v.c. 91) 75, (v.c. H12) 75, (v.c. 62) 265; convolvulus 334, 337; dumetorum 203; japonica 334, 335, 337; japonica x sachalinensis (v.c. 70) 75, (v.c. 91) 75, (v.c. H12) 75, (v.c. 62) 265 Farrell, L. - Obit. of Terence (Terry) Charles Ernest Wells (1935-2008) 277-279 Fay, M.F., with R.S. Cowan ef al. - Genetic variation in Irish Whitebeam, Sorbus hibernica E.F. Warb. (Rosaceae) and its relationship to a Sorbus from the Menai Strait, North Wales 99-108 Festuca 247; arundinacea xX gigantea [(v.c. 70)| 185; arundinacea x Lolium perenne (v.c. 10) 380; brevipila (v.c. 10) 185; xbraunit 247; xfleischeri [(v.c. 70)] 185; INDEX TO WATSONIA VOLUME 27 40] gigantea (v.c. 110) 380; gigantea x Lolium multiflorum 247; heterophylla (v.c. 13) 386; longifolia (v.c. 10) 185; ovina 63, 114, 123, 150, 154, 155, 164, 248, 249, 334, 335, 337, subsp. hirtula (v.c. 62) 263, subsp fenuifolia 32; pratensis 334, 335, 337; pratensis x Lolium multiflorum 247; pratensis x Lolium perenne (v.c. 90) 380; RopGeoo 4. 123) 291255 1552 1645 1167, Nera 4 592595 192)/5) 3292331, 03346 337, 342, 347, 350, subsp. arctica (v.c. 110) 380, subsp. litoralis (v.c. 63) 380; vivipara 32 xFestulolium braunti 247; holmbergii (v.c. 10) 380; loliaceum (v.c. 90) 380; nilssonii 247 Ficaria 246; verna 246, subsp. chrysocephala (P. D. Sell) Stace, comb. nov. 246, subsp. fertilis (Lawralrée ex Laegaard) Stace, comb. nov. 246, subsp. ficariiformis 246, subsp. verna 246 Ficus carica 334, 337 Filago arvensis 203; minima (v.c. 10) 184 Filipendula camtschatica 266; xpurpurea (v.c. 62) 266, ulmaria 252, 334, 335, 337, 343, 350; vulgaris 133 Fitz-Roberts, John: a little-known seventeenth century botanist 131-141 Foeniculum vulgare (v.c. 67) 179, 334, 335, 337 Foley, M.J.Y. - John Fitz-Roberts: a little-known seventeenth century botanist 131-141 Foley, M.J.Y. - Some localised early plant records from North-west England: then and now 355-364 Foley, M.J.Y. - The Svalbard plant collection in the Arctic Herbarium at the University of Lancaster (LANC) 365-368 Foley, M.J.Y. - Rev. of Orchids of Europe, North Africa and the Middle East. 3" ed. (by P. Delforge) 274 Foley, M.J.Y., with A.C. Jermy et al. - Sedges of the British Isles (Bk Rev.) 190-192 Forsythia suspensa (v.c. 29) 82, (v.c. 63) 385 Fragaria xananassa 334, 337; moschata (v.c. 10) 177; vesca 164, 334, 335, 337, 338 Frankenia laevis (v.c. 46) 260 Fraxinus angustifolia (v.c. 31) 268; excelsior Ppa) 252. 307, sl, 327-330, 334, 337; ornus (v.c. 81) 385 Fuchsia magellanica (v.c. 62) 80, (v.c. 70) 267 Fumaria. Fumitories of Britain and Ireland (by R.J. Murphy) (Bk Rev.) 389-390; bastardii var. hibernica 389; capreolata 389, subsp. babingtonii (v.c. 48) 75, 389; densiflora (v.c. 13) 173, 390; muralis 389, subsp. boraei 389, subsp. boraei var. britannica 389, subsp. muralis 389, subsp. neglecta 389; occidentalis 389, 390; officinalis subsp. officinalis var. officinalis 389, subsp. wirtgenil (v.c. 7) 173; parviflora (v.c. 44) 260, 389, var. symei 389; purpurea 389; reuteri 389; vaillantii 389, 390 Gagea 275; bohemica 275 Galanthus nivalis x elwesii (v.c. 62) 88; woronowii (v.c. 63) 387 Galeopsis tetrahit s.1. 392 GANinsSo ga parviflora (vee 452) 86; quadriradiata (v.c. 100 (Great Cumbrae)) 184, (v.c. 45) 270 Galium aparine 251, 252, 326, 327, 329-332; boreale 31, 32; constrictum 143, 145; debile 145; fleurotii 283; palustre 334, BOOS OAD WSAXAHIG 52. 1234325954. 337; spurium (v.c. 10) 182; verum 155, 164, 254, 255 Gastridium ventricosum (v.c. 10) 380 Gaultheria mucronata (v.c. 90) 383, shallon G75 003) IS) Genista lydia (v.c. 77) 79; pilosa (v.c. 63) 178; radiata 334, 337; rigida 325 Gentiana verna 248, 249 Gentianella amarella 132, 134; anglica (v.c. 8) 180; germanica (v.c. 8) 180, (v.c. 8) 378; germanica xX amarella (v.c. 8) 180; xpamplinii (v.c. 8) 180 Geranium 138, 252; xcantabrigiense (v.c. 46) 81; clarkei 'Kashmir Purple’ (v.c. 85) 179; dalmaticum Xx macrorrhizum (v.c. 46) 81; endressit (v.c. 110) 384; endressii x versicolor (v.c. 62) 81; himalayense (v.c. 110) 384; lucidum 328, 330, 334, 335, 337; macrorrhizum (v.c. 62) 81; molle 164; xoxonianum (v.c. 62) 81; pratense 138; purpureum 283; robertianum 114, 252, 327-331; 334, 337; rotundifolium Wee 39) WIS Wec. GTA Wer Sl) 262; SWMGTCTITB YD, Sls 37, ko, M5 Sila) Os 318, 319; versicolor (v.c. H12) 81, (v.c. 7) 179, (v.c. 29) 179, (v.c. 93) 384 Geum rivale 350; urbanum 251, 252, 326, 327, 329-331, 334, 337 Gilia achilleifolia (v.c. 62) 268 Ginkgo biloba (v.c. 63) 172 Gladiolus communis subsp. byzantinus (v.c. H12) 89, (v.c. 63) 272; illyricus 145 Glaux maritima 54, 342 Glechoma hederacea 334, 337, 343 Glyceria grandis (v.c. 63) 386; maxima 129 Gnaphalium luteoalbum (v.c. 55) 379; uliginosum 143, 145 402 Gratiola officinalis (v.c. 14) 385 Griselinia littoralis (v.c. H12) 80 Guizotia abyssinica (v.c. 17) 270 Gurney, M. - Viper's-grass Scorzonera humilis L. at Wareham Meadows, Dorset 167-170 Gymnadenia 247; borealis 247, (v.c. 110) 381; borealis x Dactylorchis maculata 247; conopsea 8, 247, subsp. borealis (v.c. 79) 89, subsp. densiflora (v.c. 110) 187; conopsea x Dactylorhiza fuchsti (v.c. 78) 264 Gymnocarpium dryopteris (v.c. 17) 260 Gypsophila viscosa (v.c. 85) 174 Habenaria xevansti 247 Hall, Peter Charles (1917-2008) (Obit.) 204-206 Hamamelis 275 Harris, S.A., with T.C.G. Rich & S.J. Hiscock - Five new Sorbus (Rosaceae) taxa from the _ Avon Gorge, England 217-228 Hawkes, John (‘Jack') Gregory (1915-2007) (Obit.) 199-200 Hebe albicans (v.c. 63) 385; brachysiphon (v.c. 62) 83; cupressoides [(v.c. 63)] 378; dieffenbachii (v.c. 38) 181; elliptica x speciosa (v.c. 62) 83, (v.c. 64) 269; xfranciscana (v.c. 62) 83, (v.c. 64) 269, MIEWISIL, SOA WES ODN OIE WAC ROMS OO: odora (v.c. 70) 269; salicifolia (v.c. 62) 83, (v.c. 44) 269; salicifolia x elliptica Wc sDi3ss Hedera algeriensis (v.c. 46) 81; colchica (v.c. 62) 81, (v.c. 90) 384; helix 47, 252, 326, 3275 329-33ie 3345 339 378 Helianthemum 275; canum 63, 134, subsp. incana 260, subsp. levigatum [(v.c. 64)] 260, [(v.c. 69)] 260; nummularium 134; oelandicum 133, 134, 138, 249, subsp. canum 134, 248 Helianthus xlaetiflorus (v.c. 52) 86, (v.c. 38) 184; tuberosus (v.c. 31) 270; tuberosus x pauciflorus (v.c. 52) 86, (v.c. 38) 184 Helichrysum bracteatum (v.c. 46) 85, (v.c. 63) 270 Helictotrichon pratense 32 Helleborus argutifolius (v.c. 62) 74, (v.c. 39) 172, =e. (63) 1725 Ace) aS (S)) esse foetidus (v.c. 45) 260; orientalis (v.c. 62) 7A, (v.c. 42) 264, (v.c. 80) 381 Hemerocallis fulva (v.c. 62) 271 Heracleum mantegazzianum 334, 337; sphondyinum (252) 33459 3953. or: sphondylium x mantegazzianum (v.c. 90) 378 INDEX TO WATSONIA VOLUME 27 Herminium monorchis 277 Herniaria glabra (v.c. 64) 376 Hesperis matronalis 334, 335, 337 Heuchera sanguinea (v.c. 62) 78 Heywood, V.H., Brummitt, R.K., Culham, A. & Seberg, O. - Flowering plant families of the world (Bk Rev.) 196-197 Hieracium 62,'63,, 109, 112, HAD 3=Isa9s: 198, 275; sect. Alpina 119, British alpine hawkweeds (by D.J. Tennant & T.C.G. Rich) (Bk Reva) 27/38 4 -aeesecie Cerinthoidea 119; sect. Hieracium 109; sect. Oreadea 61; sect. Stelligera 61, 109; sect. Subalpina 119; sect. Vulgata 61; argenteum 124, 125; argillaceum (v.c. 45) 84, (v.c. 47) 84, (v.c. 50) 84; arranense and H. sannoxense (Asteraceae), two endemic hawkweeds from the Isle of Arran, Scotland. The status of, 119-125, (illus.) 120, (map) 122; aterrimum (v.c. 29) 84; backhousei 273; basalticola 112; boswelliit 124; britannicoides 61, 62: britannicum (v.c. 50) 84; caesiomurorum 123; caesium 59, 109, var. decolor 59, 61; caledonicum 123; cambricum 61; cardiophyllum (v.c. 29) 84; carneddorum 62; cinderella (v.c. 47) 262; coniops 124, 125; consociatum (v.c. 50) 84, (v.c. 44) 262, (v.c. 47) 262: Gyelicum mss: decolor 59, 61; deganwyense 124; diaphanum (v.c. 47) 84; eucallum 124; eustomon (v.c. 110) 84; exotericum (v.c. 29) 84: firmiramum (v.c. 29) 84; gentile (v.c. 29) 84; glandulidens 124; hanburyi 273; holophyllum 61, 62; holosericeum 273; insigne f. insigne 273; koehleri (v.c. 29) 84; leyi 59, 62; maculatum s.s (v.c. 67) 184; murorum 109, subsp. pachyphyllum 109, var. pachyphyllum 109; neosparsum (v.c. 29) 84; notabile 273; onychodontum (v.c. 29) 84; pachyphylloides, Carbon- iferous Hawkweed and H._ vagicola, Tutshill Hawkweed (Asteraceae). Conservation of Britain's diversity: status of the two Wye Valley endemics, 109-118, (illus.) 110, (map) 113; pachyphyllum 109; pictorum (v.c. 77) 184; pseudoleyi (Asteraceae), Purple-flushed Hawkweed. Conservation of Britain's biodiversity 59- 64, (illus.) 60, (map.) 61; guadridentatum (v.c. 29). 85; rubicundiformemsO2- sabaudum 334, 335, 337; sannoxense (Asteraceae), two endemic hawkweeds from the Isle of Arran, Scotland. The status of HAH. arranense and, 119-125, (illus.) 121, (map) 122; saxifragum subsp. pseudoleyi 61; schmidtii 109; scotostictum (v.c. 13) 184, (v.c. 47) 270; seriflorum INDEX TO WATSONIA VOLUME 27 (v.c. 29) 85; smithii 61; stenolepiforme 283; subbritannicum 109; subcrocatum (v.c. 47) 84; subplanifolium 109; sylvularum (v.c. 29) 85; uisticola (v.c. 110) 84; vagense 61, 62; vagicola, Tutshill Hawkweed (Asteraceae). Conservation of Britain's diversity: status of the two Wye Valley endemics H. pachyphylloides, Carboniferous Hawkweed and, 109-118, (illus.) 111, (map) 113; vulgatum 124, 125, Be 923 34.337, 338 Hipkin, C.R., with T. Humphrey et al. - Vascu- lar plants from Swansea University Herb- arium (UCSA) now incorporated into the Welsh National Herbarium (NMW) 257 Hippophae rhamnoides 346, 347, 351 Hirschfeldia incana (v.c. 67) 175 Hiscock, S.J., with L. Houston ef al. - An account of the whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species 283- 300 Hiscock, S.J., with T.C.G. Rich & S.A. Harris - Five new Sorbus (Rosaceae) taxa from the Avon Gorge, England 217-228 Holcus lanatus 123, 164, 254, 327, 329, 330, Sool ns 4 45475 mollis 32. 334, 335, 337 Holodiscus discolor (v.c. 91) 78, (v.c. 29) 176, (v.c. 62) 266 Hordelymus europaeus (v.c. 39) 186, (v.c. 69) 264 Hordeum jubatum 54; murinum 334, 335, 337; vulgare (v.c. 62) 271 Hottonia palustris 129 Houston, L., Robertson, A., Jones, K., Smith, S.C.€., Hiscock, S.J. & Rich, T.C.G. - An account of the whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species 283- 300 Houston, L., Robertson, A. & Rich, T.C.G. - The distribution, population size and growth of the rare English endemic Sorbus bristoliensis A.J. Wilmott, Bristol Whitebeam (Rosaceae) 37-49 Houston, L., with T.C.G. Rich et al. - The diversity of Sorbus L. (Rosaceae) in the Lower Wye Valley 301-313 Flummaphreya I Naylor, C., Rich, 1.€.G. & Hipkin, C.R. - Vascular plants from Swansea University Herbarium (UCSA) now incorporated into the Welsh National Herbarium (NMW) 257 Huperzia selago subsp. arctica (v.c. 110) 171 Hutcheon, Key Yi with) IM; Deans e7) al - Description, ecology and establishment of Carex salina Wahlenb. (Saltmarsh Sedge) - anew British species 51-57 Hyacinthoides hispanica (v.c. 113 (S)) 186; non-scripta 32, 252 Ayacinthus orientalis (v.c. 62) 88, (v.c. H12) 88 Hydrangea macrophylla (v.c. 49) 265, (v.c. 52) 265 Hydrocotyle 146, 167, 196; ranunculoides (v.c. 52) 81, (v.c. 62) 81; vulgaris 143, 342, 347 Hylocomium splendens 32 Hyoscyamus albus (v.c. 113(S)) 384; niger (v.c. 52) 262 Hypericum androsaemum 334, 335, 337; androsaemum xX hircinum (v.c. 31) 265, (v.c. 52) 265, (v.c. 69) 265, (v.c. 63) 382; COLIC Ada S045 seas Vets xdesetangsii (v.c. 91) 76; elodes 143; forrestii (v.c. 60) 382, (v.c. 63) 382; Pd COtewn@v7C1 163)" 7 4en(v C1 44) 265: humifusum x linariifolium (v.c. 49) 76; xinodorum (v.c. 31) 265, (v.c. 52) 265, (v.c. 69) 265, (v.c. 63) 382; linariifolium (v.c. 43) 260; olympicum (v.c. 38) 174, (v.c. 62) 265; perforatum 334, 335, 337; perforatum xX maculatum (v.c. 91) 76; pulchrum 32; tetrapterum 359, 360; undulatum x tetrapterum (v.c. 2) 76 Hypnum jutlandicum 32 Hypochaeris glabra L. on the Sefton Coast, Merseyside in 2007. Population explosion of, 159-166, (map) 162; maculata 62, 277; LOGGCHA NLS SO S4ee 1552, 164) 254, 334, 335,300 Iberis sempervirens (v.c. 44) 77, (v.c. 62) 77, 334, 335, 337; umbellata (v.c. 91) 77 Ilex xaltaclerensis (v.c. 62) 267; aquifolium 334, 335, 337, 338; aquifolium x perado (v.c. 62) 267 Illecebrum verticillatum L. (Caryophyllaceaea) in Great Britain. The status of Coral- necklace, 143-148 Impatiens balfourii (v.c. 17) 268; glandulifera D2 8345 S85 uo sulianit (veCua4) Sl. (v.c. H12) 81 Inula conyza 132; hookeri (v.c. 60) 184 Tochroma australe (v.c. 10) 384 Tris germanica (v.c. 67) 89, 334, 337; orientalis (v.c. 41) 89; pseudacorus 343 Isoetes xhickeyi (v.c. 70) 73; lacustris x echinospora (v.c. 70) 73 Isolepis fluitans 191; setacea 143 404 Jasminum nudiflorum (v.c. 17) 268; officinale (v.c. 39) 181, (v.c. 62) 268 Jebb, M. - Reinstatement of the name Dactylorhiza_ kerryensis (Wilmott) P.F. Hunt & Summerh. for the Western Marsh Orchid, and a new varietal combination 371-372 Jermy, A.C. - Obit. of John Francis Michael Cannon (1930-2008) 279-281 Jermy, A.C., Simpson, D.A., Foley, M.J.Y. & Porter, M.S. - Sedges of the British Isles (Bk Rev.) 190-192 Jermy, A.C., with M. Dean et al. - Description, ecology and establishment of Carex salina Wahlenb. (Saltmarsh Sedge) - a new British species 51-57 Jones, K., with L. Houston et al. - An account of the whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species 283-300 Jones, P.S., with T.H. Blackstock - Obit. of David Paul Stevens (1958-2007) 95-97 Juncus acutiflorus 52, 167; anthelatus (v.c. 63) DIO articulatus V4A3 Ol, 2323425349: articulatus x acutiformis (v.c. 113(S)) 379; balticus 343, 346, 347, 350; bulbosus 143, 146; capitatus 147; compressus (v.c. 13) 185; xdiffusus (v.c. 49) 263, (v.c. 90) 379; effusus 167, 252; effusus x conglomeratus (v.c. 100 (Bute)) 185, (v.c. 44) 263, (v.c. 64) 379; filiformis (v.c. 110) 185; foliosus (v.c. 100 (Arran)) 185; gerardii 54, 343; inflexus 342; inflexus x effusus (v.c. 49) 263, (v.c. 90) 379; xkern-reichgeltii (v.c. 100 (Bute)) 185, (v.c. 44) 263, (v.c. 64) 379; subnodulosus 234; xsurrejanus (v.c. 113(S)) 379 Juniperus chinensis (v.c. 41) 74, (v.c. 44) 74 Kay, Q:O°N with “PR Bailey=erncala. - Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles 69-72 Kerria japonica (v.c. 42) 266, (v.c. 44) 266 Kindbergia praelonga 30 Kitchener, G.D. & Leslie, A.C. - Epilobium tetragonum subsp. tournefortii naturalised in Britain 65-67 Kniphofia uvaria (v.c. 41) 88 Kobresia 191 Koeleria macrantha 123, 254 Koelreuteria paniculata (v.c. 29) 179 Laburnum anagyroides 334, 335, 337, 338; anagyroides x alpinum (v.c. 44) . 267; xwatereri (v.c. 44) 267 : INDEX TO WATSONIA VOLUME 27 Lactuca sativa (v.c. 63) 269; serriola 76, (v.c. H12) 84; virosa (v.c. 67) 182 Lagarosiphon major (v.c. 80) 86, (v.c. H12) 86 Lambignon, J., Delvosalle, L. & Duvigneaud, J. - Nouvelle flore de la Belgique, du Grand- Duché de Luxembourg, du nord de la France et des _ régions voisines (Ptéridophytes et Spermatophytes). 5th ed. (Bk Rev.) 275 Lamiastrum galeobdolon subsp. argentatum (v.c. 51) 82, (v.c. 31) 385, (v.c. 90) 385 Lamium album 334, 335, 337, 338; confertum (v.c. 63) 378; hybridum (v.c. 43) 262; purpureum 334, 335, 337 Lamprothamnium sonderi 193 Landoltia punctata 127 Lansdown, R.V. - Red Duckweed (Lemna turionifera Landolt) new to Britain 127- Lappula squarrosa (v.c. 17) 268 Lapsana communis 327, 329, 331 Larix decidua x kaempferi \neecr rd: xmarschlinsii (v.c. 7) 172 Laser trilobum (v.c. 63) 180 Lathraea_ clandestina (v.c. 46) 83, 252; squamaria 200, 359, 361 Lathyrus grandiflorus (v.c. 62) 79, (v.c. 100 (Great Cumbrae)) 178; Jatifolius (v.c. 100 (Arran)) 178; linifolius 32; odoratus (v.c. 62) 79, 334, 337; pratensis 342 Lavandula angustifolia (v.c. 62) 82; stoechas (v.c. 63) 268 Lavatera 369; arborea (v.c. 67) 76, (v.c. 62) 260; xclementii (v.c. 69) 265, 369; cretica 369, (v.c. 10) 383; olbia x thuringiaca (v.c. 69) 265, 369 Leach, S.J. - Obit. of John Ounsted (1919- 2007) 200-204 Leach, S.J. - Rev. of Fumitories of Britain and Treland (by R.J. Murphy) 389-390 Leaney, R.M., with C.L. O'Reilly - A new nothospecies in Symphytum L. (Boraginaceae) 372-374 Leiocolea rutheana 236 Lemna 127, 128, (key) 128; sect. Lemna 128; gibba 127-129; minor 128,129; minuta (v.c. 70) 86, (.c) H12) 86, IWS IVoeGre 49) 270, ~Wic. 62) 270) Gree SO) Sse: trisulca 127, 129, (v.c. 100 (Great Cumbrae)) 185; turionifera Landolt new to Britain. Red Duckweed, 127-130; valdiviana 129 Leontodon autumnalis 334, 335, 337, 342; hispidus 254, 315, 334, 335, 337; saxatilis [55 l645345 INDEX TO WATSONIA VOLUME 27 Lepidium bonariense (v.c. 29) 175; latifolium 359, 361; sativum (v.c. 46) 265 Leslie, A.C., with G.D. Kitcener - Epilobium tetragonum subsp. tournefortii naturalised in Britain 65-67 Leucanthemella serotina (v.c. 62) 85 Leucanthemum paludosum (v.c. 62) 85; vulgare 313, SIS Leucojum aestivum (v.c. 60) 186, (v.c. 81) 387, subsp. aestivum (v.c. 63) 387, subsp. pulchellum (v.c. 46) 88; vernum (v.c. 67) 186 Levisticum officinale (v.c. 17) 384 Leycesteria formosa 334, 335, 337, 338 Ligusticum scoticum (v.c. 70) 262 Ligustrum lucidum (v.c. 67) 181; vulgare (v.c. 110) 82, 114, 334, 335, 337 Limnanthes douglasii (v.c. 62) 81 Linaria xdominii (v.c. 38) 181; maroccana WAC=O4d)) 385, (vic. IN3(S)) 385; pelisseriana (v.c. 7) 385; purpurea 328, 330, 334, 335, 337; purpurea x repens (v.c. 38) 181 Linnaea borealis (v.c. 106) 83, (v.c. 91) 379 Linum catharticum 254; usitatissimum (v.c. 51) | Listera ovata \3 Littorella uniflora 143 Lloydia serotina 200 Lobelia dortmanna 137; erinus (v.c. H12) 83, (v.c. 45) 269 Lolium 247, 318; xboucheanum (v.c. H12) 87, (v.c. 31) 263, (v.c. 62) 263; multiflorum 247, 334, 335, 337; perenne 247, 254, 327, 331, 334, 335, 337, 338, 342, 346; perenne x multiflorum (v.c. H12) 87, (v.c. 31) 263, (v.c. 62) 263 Lonicera caprifolium x etrusca (v.c. 69) 83, (v.c. 70) 269; fragrantissima x standishii (v.c. 70) 269; henryi (v.c. 38) 182; xitalica (v.c. 69) 83, (v.c. 70) 269; japonica (v.c. 48) 269; periclymenum 334, 335, 337, 338; pileata (v.c. 41) 83, (v.c. 44) 83, (v.c. 38) 182; xpurpusii (v.c. 70) 269; tatarica (v.c. 69) 83, (v.c. 70) 83, (v.c. 39) 182, (v.c. 64) 385; xylosteum (v.c. 29) 83, (v.c. 78) 269 Lophozia ventricosa 32 Lotus angustissimus (v.c. 10) 178, (v.c. 13) 178; corniculatus 63, 155, 163, 164, 249, Daa 255, 342; pedunculatus 342; subbiflorus (v.c. 13) 178 Ludwigia grandiflora (v.c. 10) 179, (v.c. 60) 179; palustris 145 Lunaria rediviva (v.c. 8) 383 405 100 178; Lupinus arboreus (v.c. nootkatensis 334, 337 Luzula xborreri (v.c. 10) 380; campestris 164; confusa 367; forsteri x pilosa (v.c. 10) 380; luzuloides (v.c. 60) 386; nivea (v.c. 62) 86, (v.c. 70) 270; sylvatica 32, 47, 124, Ds) Lychnis chalcedonica (v.c. 70) 265; flos-cuculi 350; flos-jovis (v.c. 64) 382 Lycium barbarum 334, 335, 337 Lycopersicon esculentum (v.c. 70) 82, (v.c. 48) (Bute)) Lycopodium 367; alpinum var. decipiens 376; complanatum 376 Lycopus europaeus 334, 335, 337, 343, (v.c. 81) 378 Lysichiton americanus 190, 250, 252 Lysimachia ciliata (v.c. 60) 175, (v.c. 42) 265; punctata (v.c. 31) 265; thyrsiflora (v.c. 13) 175; verticillaris (v.c. 63) 175; vulgaris 384593395337 Lythrum portula 143, 145; (v.c. 91) 80, 342 Mabberley, D.J. - Mabberley's plant-book: a portable dictionary of plants, — their classification and uses (Bk Rev.) 390-391 Magnolia 275 Mahonia aquifolium 334, 335, 337 Maianthemum bifolium 359, 361 Malcolmia maritima (v.c. 46) 76 Malus xdomestica 284; niedzwetskyana x atrosanguinea (v.c. 63) 178; xpurpurea (v.c. 63) 178 Malva 369; alcea (v.c. 58) 382, (v.c. 90) 382; x clementii (Cheek) Stace, comb. nov. 369; moschata (v.c. 100 (Bute)) 174, 369; olbia 369; olbia x thuringiaca 369; sylvestris 334, 335, 337, 338; thuringiaca 369 Matricaria discoidea 334, 337 Matteuccia struthiopteris (v.c. 62) 74, (v.c. 91) 1 Matthiola longipetala (v.c. 62) 76 McAllister, H., with J.P. Bailey ef al. - Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles 69-72 McCarthy, W., with J. Sawtschuk & T.C.G. Rich - Conservation of Britain's biodiversity: Hieracium pseudoleyi (Asteraceae), Purple-flushed Hawkweed 59-64 McCosh, D.J., with T.C.G. Rich - The status of Hieracium arranense and H. sannoxense (Asteraceae), two endemic hawkweeds from the Isle of Arran, Scotland 119-125 406 Meconopsis cambrica 326, 334, 335, 337, 338 Medicago arabica (v.c. 85) 178; lupulina 334, B85 29) 5 DIC Melampodium montanum (v.c. H12) 86 Melampyrum 257, (key) 24; aestivale 24; pratense 23, 24, (illus.) 25, 26-29, 32-34, f. purpureum 26, var. hians 24, (illus.) 25, DO" 2859295 Bile 324 var montanum: 28: silvaticum subsp. destivale 24, 31, subsp. subsilvaticum 24, var. subsilvaticum 24; sylvaticum L.; Scrophulariaceae in England. Small Cow-wheat, 23-36, (illus.) 25, subsp. aestivale 24, var. pallidiflora 23 Melica nutans 32, (v.c. 46) 87, 134; uniflora 114 Melilotus indicus (v.c. H12) 79; officinalis (v.c. 91) 79 Melissa officinalis (v.c. 31) 268, 334, 335, 337 Mentha aquatica 233, 334, 335, 337, 342, 347; arvensis X spicata (v.c. 31) 262; xgracilis (v.c. 31) 262; pulegium 145, 201; requienii GCs) yl Ole OORT (ca SO) O08; xrotundifolia (v.c. 81) 385; spicata 334, 335, 337; spicata x longifolia (v.c. 10) Sie suaveolens “I90= (vic. 150) 2e2: suaveolens x longifolia (v.c. 81) 385; xvillosonervata (v.c. 10) 181 Menyanthes trifoliata 233, 235 Mercurialis perennis 32, (v.c. 110) 179, 252, 334, 337/ Meum athamanticum 133 Mibora minima (v.c. 60) 186 Milium effusum 325, 334, 335, 337 Millegrana 146 Mimulus 247; cupreus x smithii (v.c. 44) 83; xhybridus cv. 'Mailibou Ivory’ (v.c. 44) 83; luteus 247, var. nummularius 247; nummularius (Clos) Stace, comb. nov. 247 Minuartia hybrida (v.c. 8) 174, (v.c. 8) 376; verna 248, 249 Miscanthus sinensis (v.c. 63) 271 Mnium hornum 32 Molinia 167; caerulea 167, 168, 233, subsp. arundinacea (v.c. 31) 380 Montia fontana 54 Morus nigra (v.c. 44) 264 Muehlenbeckia complexa (v.c. 113(S)) 382 Murphy, R.J. - Fumitories of Britain and Ireland (Bk Rev.) 389-390 Muscari armeniacum (v.c. 50) 271; latifolium (v.c. 63) 271, (v.c. 64) 387 Mycelis muralis 328, 330, 334, 335, 337 INDEX TO WATSONIA VOLUME 27 Myosotis ramosissima (v.c. 110) 180; secunda 143; sylvatica 334, 335, 337, 338 Myosoton aquaticum (v.c. 67) 174 Myriophyllum 392; alterniflorum 263; aquaticum (v.c. 48) 80, (v.c. 49) 80, (v.c. 31) 267; verticillatum 129 Myrrhis odorata 334, 335, 337, 338 Narcissus 252; cyclamineus (v.c. 41) 88; tazetta (v.c. H12) 88 Nardus stricta 124 Narthecium ossifragum 359, 361 Nectaroscordum siculum (v.c. 57) 88 Nemesia denticulata (v.c. 29) 83 Neottia nidus-avis 137; ovata 13 Nepeta cataria (v.c. 49) 262; racemosa (v.c. 2 Nicandra physalodes (v.c. 85) 180 Nicotiana alata (v.c. 62) 82; alata x forgetiana (v.c. 38) 180; xsanderae (v.c. 38) 180 Nigella damascena (v.c. 113 (S)) 172, (v.c. 31) 381, (v.c. 46) 264 Nolana paradoxa (v.c. 29) 81 Nonea lutea (v.c. 62) 82, (v.c. 69) 82, (v.c. 39) 180 Notes 65-72, 167-170, 243-257, 369-374 Nothofagus alpina (v.c. 69) 264; xdodecaphleps (v.c. 63) 382; obliqua (v.c. 7) 173; obliqua x alpina (v.c. 63) 382 Nothoscordum borbonicum (v.c. 63) 387 Nuphar lutea x pumila (v.c. 58) xspenneriana (v.c. 58) 376 Nymphaea alba 138; marliacea (v.c. 110) 172, (WAC) Son Nymphoides peltata 129 3/6: Obituaries 91-97, 199-206, 277-282 Oenanthe crocata (v.c. 29) 81, 334, 335, 337; fluviatilis (v.c. 57) 81; pimpinelloides (v.c. 29) 81, (v.c. H12) 81 Oenothera xfallax 155, 164; glazioviana 164 Olearia macrodonta (v.c. 70) 85, (v.c. 113(S)) 386; paniculata (v.c. 93) 386; solandri (v.c. 41) 85 Ononis repens 155, 164 Onopordum acanthium (v.c. 41) 84 Ophioglossum azoricum (v.c. 93) 376, (v.c. 110) 376 Ophrys 13, 274; apifera 277; fusca 274; insectifera 132; sphegodes 274 xOrchigymnadenia evansti 247 INDEX TO WATSONIA VOLUME 27 Orchis 275; xevansii 247; kerryensis 247, 371; majalis var. occidentalis 371; mascula x morio 247; morio 247, 277; xmorioides 247; occidentalis 371, subsp. kerryensis 371; purpurea (v.c. 14) 381; sancta 274 O'Reilly, C. - Rev. of Charophytes of the Baltic Sea (ed by H. Schubert & I. Blindow) 192- Is) O'Reilly, C.L. & Leaney, R.M. - A new MOUNOSpPeEcles in Symphytum “VL: (Boraginaceae) 372-374 Origanum mayjorana (v.c. 17) 268; vulgare 337, 338 Ornithopus perpusillus 150, 155, 164 Orobanche hederae (v.c. 67) 181, (v.c. 14) 378 Osmunda regalis (v.c. 31) 260 Oswald, P.H. - Rev. of The wild flowers of the Isle of Purbeck, Brownsea and Sandbanks (by E.A. Pratt) 276 Ounsted, John (1919-2007) (Obit.) 200-204 Oxalis acetosella 32, 252; articulata (v.c. 44) 81; corniculata 334, 335, 337, (v.c. 110) 384; debilis (v.c. 69) 81 Oxyria digyna 132, 134 Paeonia officinalis (v.c. 51) 76, (v.c. 51) 265 Paludella squarrosa 236 Panicum miliaceum (v.c. 46) 271 Papaver rhoeas (v.c. 100 (Bute)) 173, (v.c. 100 (Great Cumbrae)) 173, 334, 335, 337, 338; somniferum 334, 335, 337 Parentucellia viscosa (v.c. 67) 181 Parietaria judaica 332, 334, 335, 337, 338 Paris quadrifolia 29, 361, 362 Parmelia perlata 138 Parmotrema perlatum 138 Parnassia palustris 31, 54, 343, 347, 359, 360 Parr, S., with K.J. Walker & S. Ward - The re- discovery of Arenaria norvegica subsp. norvegica in Ireland 248-250 Parthenocissus inserta (v.c. 46) 81 Pastinaca sativa 343 Pearman, D.A. - The status of Coral-necklace Illecebrum verticillatum LL. (Caryoph- yllaceaea) in Great Britain 143-148 Pearman, D.A. - Rev. of Anglesey (v.c. 52) county rare plant register. Flowering plants and ferns (by I.R. Bonner) 194 Pearman, D.A. - Rev. of County Antrim, scarce, rare and extinct vascular plant register (by S. Beesley) 194 Pearman, D.A. - Rev. of Monmouthshire county rare plant register (by T.G. Evans) 194 407 Pearman, D.A. - Rev. of Wiltshire rare plant register. The rare and threatened vascular plants of North and South Wiltshire (by S. Pilkington) 194 Pearman, D.A., Preston, C.D., Rothero, G.P. & Walker, K.J. - The flora of Rum. An Atlantic island reserve (Bk Rev.) 197-198 Pedicuiaris palustris (v.c. 31) 262 Pelargonium xhortorum (v.c. 62) 81 Peltaria alliacea (v.c. 81) 383 Pentaglottis sempervirens 334, 335, 337 Persicaria affinis (v.c. 64) 382; capitata (v.c. 62) ie Save Ca OS) alas Mvec.) 70) 265% hydropiper 350." vivipara 30,31) 137: weyrichii (v.c. 63) 382 Petasites fragrans 190; hybridus 252, 334, 335, 337; japonicus (v.c. 96) 184, 250 Petrorhagia nanteuilli (v.c. 38) 174; prolifera (v.c. 62) 75; saxifraga (v.c. 13) 382 Petunia xhybrida (v.c. 44) 82, (v.c. 62) 82, (v.c. 64) 268, (v.c. 69) 268; axillaris x integrifolia (v.c. 44) 82, (v.c. 62) 82, (v.c. 64) 268, (v.c. 69) 268 Peucedanum palustre (v.c. 85) 180 Phacelia tanacetifolia (v.c. 51) 82, (v.c. 62) 82 Phalaris arundinacea 52, 334, 335, 337; canar- iensis (v.c. 91) 87; paradoxa (v.c. 59) 186 Philadelphus. coronarius xX microphyllus x pubescens (v.c. 38) 175; inodorus (v.c. 80) 77; xvirginalis (v.c. 38) 175 Phippsia algida 366 Phleum arenarium 164, (v.c. 63) 263; pratense 254 Photinia davidiana (v.c. 45) 79, (v.c. 91) 79 Phragmites 52; australis 167, 233, 236, 343 Phyllitis 275; scolopendrium 252, 326-331, 83433875 359 Phyllostachys bambusoides (v.c. 52) 271; nigra (v.c. 44) 271; viridiglaucescens (v.c. 52) 271 Physalis alkekengi (v.c. 62) 82, (v.c. 69) 82 Physocarpus opulifolius (v.c. 50) 78 Picea glauca (v.c. 63) 381; pungens (v.c. 63) 381 Picris hieracioides (v.c. 69) 262 Pieris japonica (v.c. 63) 383 Pilkington, S. - Wiltshire rare plant register. The rare and threatened vascular plants of North and South Wiltshire (Bk Rev.) 194 Pilosella 275: aurantiaca 334-5, 337, 338; officinarum 123, 155, 164, 334-5, 337, 338 Pilularia globulifera (v.c. 69) 74, (v.c. 8) 172, (v.c. 10) 376 408 Pimpinella major (v.c. 70) 262 Pinguicula grandiflora (v.c. 60) 181; vulgaris 357, 359-362 Pinus nigra 165, subsp. nigra (v.c. 44) 264; radiata (v.c. 44) 74 Pistia stratiotes (v.c. 57) 86, (v.c. 60) 185 Pisum sativum (v.c. 62) 80, 334, 337 Plagiochila asplenioides 32 Plagiothecium undulatum 32 Plant records 73-89, 171-187, 259-272, 375- 387 Plantago coronopus 154, 155, 165; lanceolata Sy IOS), BAS, AAO), M4 SWI, BAO. SSl. 3345, 397, 33045425 MAO S54 5555 So. 338, 342; maritima 54, 123, 124, 249; media (v.c. 48) 82, 334, 337 Platanthera species in the British Isles adaptively optimized or an evolutionary red herring? Is spur length of, 1-21; sect. Lacera 1; sect. Limnorchis 1; sect. Piperia 1; sect. Platanthera 1; aquilonis 1; bifolia 1, 2, 4-9, (illus.) 9-14, 17-21, subsp. graciliflora 8, var. robusta 8; chlorantha i 2 ine) 3. 254." Mls yO, Wyezile chlorantha x_ bifolia 10; dilatata 1; xhybrida 10; lacera 1; leucophaea | Pleurozium schreberi 32 Poa 350; angustifolia 338; annua 165, 327, 329-331, 334, 337, 343, f. purpurea (v.c. 46) 87; chaixii (v.c. 62) 87, (v.c. 48) 271; COMPVESSA 329.360, 3345 3395) Soi Ec 81) 380; humilis 165, 343; infirma (v.c. 31) 263, (v.c. 13) 380; infirma x annua (v.c. 10) 186; nemoralis 31, 32, 328, 330, 334, 335, 337, 380; pratensis 328-331, 334, 335, 337, 342; pratensis agg. 327; trivialis 252, 334, 335, 337 Poland, J. & Clement, E.J. - The vegetative key to the British flora (Bk Rev.) 391-392 Polemonium caeruleum (v.c. 41) 82 Polygala serpyllifolia 32; vulgaris 165, 337 Polygonatum multiflorum 132; odoratum 132, 358, 360 Polygonum aviculare 334, 335, 337; rurivagum (v.c. 110) 174 Polypodium 124; xfont-queri 30; vulgare 155, 2 Polypogon fugax (v.c. 8) 387; monspeliensis (v.c. 8) 380; viridis (v.c. 62) 87, (v.c. H12) Si WEC. SD) OTIC ZCs 8) S86 Polystichum xbicknellii (v.c. 64) 260; munitum (v.c. 58) 381; setiferum (v.c. 31) 260, 334, 335, 337; setiferum x aculeatum (v.c. 64) 260 INDEX TO WATSONIA VOLUME 27 Polytrichum formosum 32; juniperinum 150 Pontederia cordata [(v.c. 58)] 380 Populus 346; alba 165; balsamifera (v.c. 7) 174; nigra subsp. betulifolia (v.c. 35) 76; trichocarpa (v.c. 67) 174 Porter, M.S., with A.C. Jermy et al. - Sedges of the British Isles (Bk Rev.) 190-192 Portulaca oleracea (v.c. 13) 174 Posidonia 281 Potamogeton acutifolius 129; compressus (v.c. 57) 86; xXgessnacensis (v.c. 46) 86; gramineus 263; lucens 129; natans 129; natans X gramineus (v.c. 46) 263; natans x polygonifolius (v.c. 46) 86; obtusifolius (v.c. 62) 86; pectinatus 129; polygonifolius 143, 234; xsparganiifolius (v.c. 46) 263; xsudermanicus 127, 129; trichoides 129 Potentilla 275; anglica x reptans (v.c. 90) 378; anserina 334, 337, 342; argentea 132; atrosanguinea (v.c. 63) 177; crantzii 30, 31, (v.c. 91) 78; erecta 254, subsp. strictissima (v.c. 63) 177; erecta x anglica (v.c. 90) 3773, xmixta, Cees 90) reese montana (v.c. 46) 78; nepalensis (v.c. 62) 78; palustris 167, 233, 236; recta (v.c. 42) 266; reptans 254, 342; sterilis 337, 338; xsuberecta (v.c. 90) 377; tabernaemontani (v.c. 64) 260 Poterium 369; sanguisorba’ L. — subsp. balearicum (Bourg. ex Nyman) Stace, comb. nov. 369, sanguisorba subsp. muricatum 369 Pratt, E.A. - The wild flowers of the Isle of Purbeck, Brownsea and Sandbanks (Bk Rev.) 276 Preston, C.D., with D.A. Pearman et al. - The flora of Rum. An Atlantic island reserve (Bk Rev.) 197-198 Primula farinosa 358-362; xpolyantha (v.c. 48) 77; veris 254, (v.c. 110) 377: vuleaniss3Z 123; vulgaris: x veris (v.c. 48) iiees so. 360, 362 Proctor, M.C.F., with T.C.G. Rich - Some new British and Irish Sorbus L. taxa (Rosaceae) 207-216 Prunella vulgaris 165, 249, 254, 334, 335, 337, 338, 342 Prunus avium 334, 335, 337, 338; cerasus (v.c. 63) 177; domestica subsp. instititia (v.c. 91) 384; xfruticans (v.c. 62) 79, (v.c. 63) 177; laurocerasus 338; padus 334, 337, 361, 362; persica (v.c. 69) 266; serrula (v.c. 63) 177; spinosa 334; 3355 SoVienooe: spinosa Xx domestica (v.c. 62) 79, (v.c. 63) 7 INDEX TO WATSONIA VOLUME 27 Pseudofumaria alba (v.c. 62) 75; lutea (v.c. EMD) 595544 985533i41338 Pseudorchis albida 30, 31, (illus.) 135, 137 Pseudosasa japonica (v.c. 63) 270, (v.c. 8) 386 Pseudoscleropodium purum 32 Pteridium aquilinum 123, 252, 334, 337 Pteris multifida (v.c. 29) 74, (v.c. 17) 381 Pterocarya fraxinifolia (v.c. 62) 75 Puccinellia 54; distans 185, (v.c. 8) 380; xhybrida [(v.c. 67)] 185; maritima 52, 54; maritina xdistans [(v.c. 67)] 185; phryganodes 54; retroflexa 54 Pulicaria dysenterica 343; vulgaris 145 Pulmonaria officinalis 134; rubra (v.c. 59) 180 Pulsatilla vulgaris 277 Puschkinia scilloides (v.c. 64) 272 Pyracantha rogersiana (v.c. 7) 178, (v.c. 69) 267 Pyrola minor 161, (v.c. 13) 175; rotundifolia 30, 235, subsp. maritima (v.c. 49) 260 Pyrus aria yy intermedia 37; latifolia 37; pyraster (v.c. 44) 261 Quercus 47, 165; ilex 47, (v.c. 100 (Bute)) 173; spemacana4, 125,334, 337: robur 335; rubra (v.c. 46) 75 Radiola linoides 143, 144, 146 Ranunculus 146, 246; aconitifolius 250, 252; OCHS BX, 334, BOD, SOU, S202 (notelo (v.c. 100 (Bute)) 172; bulbosus 165, 254, Soipeoo osc. 9) 376; circinatus: 129: cymbalaria 54; ficaria 246, 251, 252, 275, SMOS PROM Dilifer | 246. subsp. chrysocephalus 246; subsp. fertilis 246, 275; subsp. ficariiformis (v.c. 67) 74, (v.c. 31) 264; flammula 143, 167, 197, 342; parviflorus (v.c. 85) 172, 276; pedatifidus Soi -srenens 252,327, 329, 331, 334, 337, 338, 342, 350; sulphureus 366 Raphanus raphanistrum subsp. maritimus (v.c. 63) 175, (v.c. 110) 377 Reseda luteola 334, 335, 337, 338 Rhagadiolus stellatus (v.c. 38) 184 Rhamnus cathartica 283 Rheum xhybridum (vy.c. 49) 265; palmatum (v.c. 69) 75; palmatum x rhaponticum (v.c. 49) 265 Rhinanthus minor 315, 342; minor agg. 54 Rhizomnium punctatum 32 Rhododendron luteum (v.c. 91) 77, (v.c. 90) 383; ponticum 334, 335, 337 Rhus typhina (v.c. 31) 384 409 Rhytidiadelphus squarrosus 32; triquetrus 30, 3) Ribes divaricatum (v.c. 46) 77; nigrum 252, 337, 338; odoratum (v.c. 41) 77, (v.c. 67) 77; rubrum 334, 335, 337; sanguineum (v.c. 100 (Bute)) 175, 337, 338, (v.c. 31) 383; uva-crispa 334, 335, 337, 338 Riccia fluitans 129 Rich, T.C.G. - Validation of names for new Avon Gorge Sorbus (Rosaceae) taxa 370 Rich, T.C.G. & McCosh, D.J. - The status of Hieracium arranense and H. sannoxense (Asteraceae), two endemic hawkweeds from the Isle of Arran, Scotland 119-125 Rich, T.C.G. & Proctor, M.C.F. - Some new British and Irish Sorbus L. taxa (Rosaceae) 207-216 Rich, T.C.G., Harris, S.A. & Hiscock, S.J. - Five new Sorbus (Rosaceae) taxa from the Avon Gorge, England 217-228 Rich, T.C.G., Houston, L., Charles, C.A. & Tillotson, A.C. - The diversity of Sorbus L. (Rosaceae) in the Lower Wye Valley 301-313 Rich, T.C.G., with L. Houston & A. Robertson - The distribution, population size and growth of the rare English endemic Sorbus bristoliensis A.J. Wilmott, Bristol Whitebeam (Rosaceae) 37-49 Rich, T.C.G., with J. Sawtschuk - Conservation of Britain's diversity: status of the two Wye Valley endemics AHieracium pachyphylloides, Carboniferous Hawkweed and H. vagicola, Tutshill Hawkweed (Asteraceae) 109-118 Rich, T.C.G., with D.J. Tennant - British alpine hawkweeds (Bk Rev.) 273-274 Rich, T.C.G., with J. Sawtschuk & W. McCarthy - Conservation of Britain's biodiversity: Hieracium pseudoleyi (Asteraceae), Purple-flushed Hawkweed 59-64 Richt= iGiGas swith ke Balleyeven al. = Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles 69-72 Rich, T.C.G., with R.S. Cowan et al. - Genetic variation in Irish Whitebeam, Sorbus hibernica E.F. Warb. (Rosaceae) and its relationship to a Sorbus from the Menai Strait, North Wales 99-108 Rich, T.C.G., with L. Houston et al. - An account of the whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species 283- 300 410 Rich, T.C.G., with T. Humphrey ef al. - Vascular plants from Swansea University Herbarium (UCSA) now incorporated into the Welsh National Herbarium (NMW) ATI Robertson, A., with L. Houston & T.C.G. Rich - The distribution, population size and growth of the rare English endemic Sorbus bristoliensis A.J. Wilmott, Bristol Whitebeam (Rosaceae) 37-49 Robertson, A., with L. Houston ef al. - An account of the whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species 283- 300 Rodgersia podophylla (v.c. 52) 265 Rorippa islandica (v.c. 52) 77, (v.c. 69) 77, (v.c. 70) 260 Rosa 114, 124; sect. Caninae subsect. Caninae 325; Albertine (ic. 113)(S)))) 177. xavrayensis (v.c. 37) 261; xbigeneris (v.c. 37) 261; caesia 261, subsp. caesia (v.c. 46) 261, [(v.c. 37)] 261, subsp. vosagiaca (v.c. 105) 261; caesia subsp. vosagiaca x rubiginosa (v.c. 108) 261; caesia subsp. vosagiaca X Ssherardii (v.c. 105) 261; caesia subsp. vosagiaca X tomentosa (v.c. S37) 20levcanina IZA5 2525 26s 3845335; 337, 338; canina agg. 165; canina x caesia subsp. vosagiaca (v.c. 105) 261; canina x OUTS NOK Es OS) MITE (Gies. SMe Zols Dorothy, VRerkinsy A@eCan lS (S)) eS; xdumalis. (vic. 105) 261 ‘Relicite Perpétue' (v.c. 113(S)) 78; ferruginea (v.c. 62) 79; majalis (v.c. 37) 266; micrantha 261; xmolletorum (v.c. 63) 177, [(v.c. 37)] 261; mollis (v.c. 36) 261; xobovata (v.c. 108)" 26k 73))n © 266: spinosissima xX mollis (v.c. 78) 260; tomentosa 261; tomentosa x rubiginosa (v.c. 37) 261; virginiana (v.c. 44) 79, (v.c. 52) 79, (v.c. 81) 384 Rosmarinus officinalis (v.c. 7) 181, (v.c. 63) 181 Rothero, G.P., with D.A. Pearman et al. - The flora of Rum. An Atlantic island reserve (Bk Rev.) 197-198 Rubus 125, 195, 198, L. (Rosaceae) in Maine and North Brittany, France. British species of, 243-244, 245, 252; ser. Rhamnifolii 72; ser. Vestiti 72; adscitus 243; aequalidens 243, 244; aghadergensis (v.c. 100) 177; ahenifolius (v.c. 11) 177; albionis (v.c. 71) INDEX TO WATSONIA VOLUME 27 377; althaefolius 245; andevagensis 243; anglofuscus (v.c. 11) 377; armeniacus (v.c. 31) 266; armipotens (v.c. 59) 176; asperidens 243; averyanus (v.c. 12) 176; babingtonianus W.C.R. Watson (Rosa- ceae). The status of, 245-246; babingtonii @c, 10) 177 barton (cea: bertramii 243; botryeros 244; caesius 155, 165, 342, 347; cambrensis (v.c. 11) 176, 377; cardiophyllus 243; cerdicii. Series reassignment of, 72; chamaemorus 358, 360-362; cockburnianus (v.c. 38) 176; conjungens (v.c. 71) 377; criniger (v.c. 10) Vi: dasyphyllus \(v.c.929) aioe distractiformis (v.c. 69) 78; echinatus 243; elegantispinosus 243; errabundus (v.c. 91) 377; flexuosus 243; fruticosus agg. 47, 243, 252; 325-329, 330 33leees4eoore furvicolor (v.c. 91) 78; halsteadensis (v.c. 29) 177; hastiformis 243; hindii (v.c. 9) I77: idaeus 334, 337 233 Sh oOo Ore idaeus X caesius (v.c. 10) 176; lanaticaulis 244: lasiodermis (v.c. 10) 377; leightonii (v.c. 29) 177; leuchostachys (v.c. 10) 177, 244, [(v.c. 10)] 377; loganobaccus (v.c. 57) 78, (v.c. 10) 383; melanocladus (v.c. 10) 177; moylei (v.c. 46) 78; nemoralis 72; neomalacus 244; newbouldii (v.c. 29) 177; parviflorus (v.c. 59) 176; peninsulae 243, 244; phoenicolasius (v.c. 49) 78, (v.c. 39) 176; pistoris (v.c. 71) 3773 plicaris 243; proiectus (v.c. 12) 177; xpseudoidaeus (v.c. 10) 176; pyramidalis 243; questieri 243: raduloides (v.c.. 29) M77 Gace) 377; rubritinctus 243-\saxailis sole oD (v.c. 91) 377; scaber 244; scabripes (v.c. 10) 377; septentrionalis (v.c. 12) 176; spectabilis .(v.c. 52) 78; (v.c. 38)" 176; surrejanus (v.c. 71) 377; tamarensis 244; trichodes 244; tricolor (v.c. 38) 176; venetorum 244; vestitus [(v.c. 77)] 78, 244, (v.c. 90) 377, (v.c. 110) 377; wirralensis Qen IUD STS Ones 7/1) S77! Rudbeckia hirta (v.c. 63) 270 Rumex x abortivus (v.c. 44) 76, (v.c. 44) 260; acetosa 123, 155, 335, 338sua@eetosella 154, 165, 334, 337; conglomeratus x obtusifolius (v.c. 44) 76, (v.c. 44) 260; crispus 334, 335;° 337) senispusmuex conglomeratus (v.c. 44) 75, (v.c. H12) 76, (v.c. 29) 174; crispus X sanguineus (V.c. 44) 76, (v.c. H12) 76; dentatus (v.c. 90) 382; xdufftii (v.c. 44) 76, (v.c. H12) 76; xhybridus (v.c. 90) 376; longifolius 334, 335, 337; longifolius x obtusifolius (v.c. 90) 376; obtusifolius 327, 33334655) 337, 338; palustris” 26\— jpulehemmole xsagorskii (v.c. 44) 76, (v.c. H12) 76, INDEX TO WATSONIA VOLUME 27 sanguineus 334, 335, 337; sanguineus x obtusifolius (v.c. 44) 76, (v.c. H12) 76; scutatus (v.c. 77) 265; xschulzei (v.c. 44) TO WECe 2) MOA Wci 29) 74: Ruscus aculeatus (v.c. 80) 387 Ruta graveolens (v.c. 62) 81, (v.c. 17) 268 Saginamapetala, 65, 334, 335, 337, 338; procumbens 334, 335, 337, 338, 343 Sagittaria sagittifolia (v.c. 44) 86, 129 Salicornia 54 Salix. New willow hybrid 67-69, 232, 346; QGUIMOMA(VeC. 52)) 10; Gurita ©8, 69, 125; caprea 68, 69, 334, 335, 337, 338, subsp. sphacelata (v.c. 67) 175; caprea x cinerea 69, (v.c. 44) 260; caprea x cinerea x repens 67; cinerea 68, 69, 125, 335, 342, 351, subsp. cinerea (v.c. 17) 377, (v.c. 58) 377, subsp. oleifolia 68; cinerea x myrsinifolia (v.c. 67) 175; xehrhartiana (v.c. 63) 383; fragilis x alba (v.c. 78) 260; xfruticosa (v.c. 85) 174; xlaschiana 69; xmollissima (v.c. 58) 376; pentandra x alba (v.c. 63) 383; xpermixta J. Webb, hybr. nov. 67, (illus.) 68; xreichardtii 68, 69, (v.c. 44) 260; repens 68, 69, 155, 165, 342, 345, 347, 349-351; xrubens (v.c. 78) 260; xstipularis (v.c. 37) 76; xstrepida (v.c. 67) 175; xsubsericea 69; triandra x viminalis (v.c. 58) 376; viminalis 334, 335, 337; viminalis x aurita (v.c. 85) 174; viminalis X caprea X aurita (v.c. 57) 76 Salvia officinalis (v.c. 63) 268; pratensis (v.c. 7) 378; reflexa (v.c. 10) 181; verbenaca (v.c. 81) 385 Sambucus ebulus (v.c. 78) 262; nigra 327, 329- ISL, Dos Ss Samolus valerandi (v.c. 17) 377 Sanguisorba 369; minor 63, (v.c. 100 (Bute)) 177, 254, 255, 369, subsp. balearica 369, subsp. muricata (v.c. 51) 78, 369, subsp. polygama 369; officinalis 32, 315 Sanicula europaea 124, 252 Sasa palmata (v.c. 67) 87, (v.c. H12) 87, (v.c. 52) 270 Sawtschuk, J. & Rich, T.G.R. - Conservation of Britain's diversity: status of the two Wye Valley endemics Hieracium pachy- phylloides, Carboniferous Hawkweed and H. vagicola, Tutshill Hawkweed (Asteraceae) 109-118 Sawtschuk, J., McCarthy, W. & Rich, T.C.G. - Conservation of Britain's biodiversity: Hieracium pseudoleyi (Asteraceae), Purple-flushed Hawkweed 59-64 41] Saxifraga aizoides 137, 138; xarendsii (v.c. 44) 78; cymbalaria (v.c. 60) 176; xgeum (v.c. 80) 383; granulata 95; oppositifolia 138, xpolita (v.c. 106) 78; spathularis (v.c. 90) 383; spathularis xhirsuta (v.c. 106) 78; tridactylites 248, 249; umbrosa (v.c. 67) 176; umbrosa x hirsuta (v.c. 80) 383; xurbium 334, 337 Scabiosa atropurpurea columbaria 63, 114 Scandix pecten-veneris (v.c. 51) 81 xSchedolium 247; brauniti (K. Richt.) Stace, comb. nov. 247; nilssonii (Cugnac & A. Camus) Stace, comb. nov. 247 Schedonorus 247; arundinaceus 247; giganteus 247; giganteus x Lolium multiflorum 247; pratensis 247; pratensis x Lolium multiflorum 247 Schizostylis coccinea (v.c. 110) 187, (v.c. 44) DA2A (WeCa59) 307 Schoenoplectus lacustris 236, (v.c. 35) 263 Schoenus 381; nigricans 234, 235 Schubert, H. & Blindow, I. (eds) - Charophytes of the Baltic Sea (Bk Rev.) 192-193 Scilla liliohyacinthus (v.c. 13) 387; siberica (v.c. 62) 88 Scorpidium revolvens 235 Scorpiurus muricatus (v.c. 13) 178 Scorzonera humilis L. at Wareham Meadows, Dorset. Viper's-grass, 167-170, (map) 169 Scrophularia aquatica 132; scorodonia (v.c. 90) 385 Seberg, O., with V.H. Heywood ef al. - Flowering plant families of the world (Bk Rev.) 196-197 Sedum aerée 772.150; 155." 163; 165; 197, 329, 300) 334, 335, 3372 album 334°.335, 337, 338: anelicum Woecs DMI 12324: confusum (v.c. H12) 77; dasyphyllum (v.c. 69) 77; forsterianum (v.c. 91) 77, 283; ‘Herbstfreude’ (v.c. 62) 176, (v.c. 63) 176; hispanicum (v.c. 63) 383; kamtschaticum (v.c. 100 (Bute)) 176, var. ellacombianum (v.c. 45) 78; nicaeense (v.c. 29) 176; pracaltum(v.c. Al) W7, (wc. 39). 176: reflexum 334, 335, 337; rosea 358, 360, 362; rupestre 65; sexangulare (v.c. 29) 77, (v.c. 13) 176; spathulifolium (v.c. 63) 176; spectabile (v.c. 49) 77, (v.c. 90) 383; spectabile x telephium (v.c. 62) 176, (v.c. 63) 176; spurium 337, 338; villosum 137 Selaginella kraussiana (v.c. 38) 171, (v.c. 100 (Great Cumbrae)) 171, 195, (v.c. 35) 264 Sell, P.D. - Rev. of British alpine hawkweeds (by D.J. Tennant & T.C.G. Rich) 273-274 GEC wr nOS) e269: 412 Sempervivum tectorum 334, 335, 337, 338 Senecio xalbescens (v.c. 62) 184; cinerea 334, 335, 337; cineraria x jacobaea (v.c. 62) 184; eboracensis 195; erucifolius (v.c. 69) 263; inaequidens (v.c. 29) 85, (v.c. 67) 85, @c. $5) 184 (2eo 7) S86) Cre7 88386; (ECO, N73), N54 WSD5 103, OD; 320; 3277, 329=33)\, 3342 339 342 acobacaex aquaticus (v.c. 78) 263; xostenfeldii (v.c. 78) 263; smithii (v.c. 91) 85; squalidus 326, 327, 329) 334 337 svernnalis -(v.c159) 184; viscosus 334, 335, 337, 338; vulgaris OSs S265 3275 3308 SS ISS 4e335e981 Serratula tinctoria 132, 254 Sesleria caerulea 133, 134, (illus.) 135, 137, 249 Setaria italica (v.c. 45) 88; pumila (v.c. 47) 88, (v.c. 62) 186 Sexton, R., with R.M. Bateman - Is spur length of Platanthera species in the British Isles adaptively optimized or an evolutionary red herring? 1-21 Sherardia arvensis 337, 338 Shimwell, D.W. - Studies in the floristic diver- sity of Durham walls, 1958-2008 323-338 Sidalcea malviflora (v.c. 81) 383 Silene dioica 252, 337, 338; gallica (v.c. 70) 260; vulgaris 334, 335, 337 Silybum marianum (v.c. 113 (S)) 182 Simpson, D.A., with A.C. Jermy ef al. - Sedges of the British Isles (Bk Rev.) 190-192 Sinapis arvensis 334, 337 Sison amomum 391 Sisymbrium altissimum (v.c. 46) 265; irio (v.c. 13), 175; W.c. 58) 3833 officinale 3273 329; 331, 334, 337; polyceratium (v.c. 81) 383; runcinatum (v.c. 81) 383; strictissimum (v.c. 38) 175 Sisyrinchium californicum (v.c. 46) 89, (v.c. 62) 186, (v.c. 64) 387; striatum (v.c. H12) 89, (v.c. 38) 187, (v.c. 62) 187 Skeffington, M.S., with J.P. Conaghan - The distribution and conservation. of Eriophorum gracile Koch ex _ Roth (Cyperaceae), Slender Cotton-grass, in Ireland 229-238 Smith, P.H. - Corynephorus canescens (L.) P. Beauv. (Grey Hair-grass) on the Sefton Coast, Merseyside (v.c. 59) 149-157 Smith, P.H. - Distribution, status and ecology of Blysmus compressus (L.) Panz. ex Link on the Sefton Coast sand-dunes, Merseyside 339-353 Smith, P.H. - Population explosion of Hypochaeris glabra L. on the Sefton Coast, Merseyside in 2007 159-166 INDEX TO WATSONIA VOLUME 27 Smith, R.J., with R.S. Cowan et al. - Genetic variation in Irish Whitebeam, Sorbus hibernica E.F. Warb. (Rosaceae) and its relationship to a Sorbus from the Menai Strait, North Wales 99-108 Smith, S.C.C., with L. Houston eft al. - An account of the whitebeams (Sorbus L.., Rosaceae) of Cheddar Gorge, England, with description of three new species 283- 300 Smyrnium olusatrum (v.c. 90) 378 Solanum . dulcamara 165, 3345 535550 oor- sarachoides (v.c. 90) 384; sisymbriifolium (v.c. 35) 268 Soleirolia soleirolii (v.c. 110) 75 Solidago canadensis 334, 335, 337; virgaurea 124 Sonchus arvensis 343; asper 165, 327, 329, 331, 334,337: oleraceus (oon lOs oie 329-3318 35455o"1 Sorbus 37-39, 47, 48, L. (Rosaceae) in the British Isles. Chromosome numbers in, 69- 72, from the Menai Strait, North Wales. Genetic variation in Irish Whitebeam, Sorbus hibernica E.F. Warb. (Rosaceae) and its relationship to a, 99-108, 189, L. taxa (Rosaceae). Some new British and Irish, 207-216, (Rosaceae) taxa from the Avon Gorge, England. Five new, 217-228, L., (Rosaceae) of CheddanaeGonge: England, with description of three new species. An account of the whitebeams, 283-300, (map) 285, L. (Rosaceae) in the Lower Wye Valley. The diversity of, 301- 313; Gap) 302 (Rosaceae) taxa. Validation of names for new Avon Gorge, 370, subgen. Aria 291, 293; subgen. Sorbus 227, admonitor M. C. F. Proctor, sp. nov. 207, 208, (illus.) 208; anglica 47, 70, 71,207, Gllus:)) 2095 2lO miei so- 284, 286-288, 291, (map) 293, 294, 297, 308, 309, 311, 312: ania@4a7s 10S ie @wece H12) 79, 99, 101-106, 215, 220, 222, 224- 227, 283, 284; 286-289) 29i 29S 298, 304, 305, 308, 309, 3115 subspsa7za 70, var. longifolia; aria x aucuparia 311; aria x bristoliensis 220), 3703eaniam< eminens 224, 370; aria x latifolia 70; aria x porrigentiformis 218, 220, 224, 294, 370: aria X torminalis 71, 311; arranensis 70, Gllus.)) 212) 213, 214: anenponaw2>: 32-34, 41; Cllus.) 212, 213, 2145 (Gilus?) 227, 283, 308, 309;* 315 3346 SS5smoore 338, subsp. aucuparia 70; aucuparia xX aria 213; aucuparia x Teyana 211; aucuparia X scalaris 226, 370; xavonensis T. Rich, hybr. nov. 370, 218, (illus.) 219, INDEX TO WATSONIA VOLUME 27 220, (map) 221, 294; bristoliensis A.J. Wilmott, Bristol Whitebeam (Rosaceae). The distribution, population size and growth of the rare English endemic 37-49, (map) 40, 70, 106, 217, 220, 222; “Bristol porrigentiformis’ 99, (illus.) 100, 101-105, 224; ‘Bristol pseudoporrigentiformis’ 224; cambrensis M. C. F. Proctor, sp. nov. 208, 209, 214; cheddarensis L. Houston & Ashley Robertson, sp. nov. 288, 284, 286, 287, 289, (illus.) 289, 291, 294, (map) 295, 297, 298; confusa 70; croceocarpa 38, 39, 70; cuneifolia T. C. G. Rich, sp. nov. 209, (illus.) 209, 210; decipiens 38, 70, 227; devoniensis 70, 207, 208, (illus.) 208; domestica 71, 305; eminens 70, 71, 99, (illus.) 100, 101-106, NOD DONA25 226, 283, 284; 286; 287, 291, 293, 294, (map) 295, 297, 303, 304, 308, 309, 311, 312; eminentiformis T. C. G. Rich, sp. nov. 210, 303, 308, 309, 311, 312; eminentoides L. Houston, sp. nov. 290, 284, 286-288, (illus.) 290, 291, 293, 294, (map) 296, 297, 298; graeca 304; hibernica 70, 71, E.F. Warb. (Rosaceae) and its relationship to a Sorbus from the Menai Strait, North Wales. Genetic variation in Irish Whitebeam, 99-108, (illus.) 100; xhoustoniae T. Rich, hybr. nov. 370, 220, (map) 221, dillus.) 222; intermedia 214, 227, 288, 291, 294, 308, OSmmolene 34. 335, 337; Joseph Rock' (v.c. 44) 266; lancastriensis 70; latifolia 38, 39, 70, 71; leighensis T. Rich, sp. nov. 370, 218, 220, (map) 221, (illus.) 222-224; leptophylla 71, 209, 215; leyana iilenClitass) 202, 213, 2142 xliljeforsim 2.13: margaretae M. C. F. Proctor, sp. nov. 210, (illus.) 211, 293; ‘Menai Strait’ 102- 105; minima 71, 214; xmotleyi T. C. G. Rich, hybr. nov. 211, (illus.) 212, 213; parviloba T. C. G. Rich, sp. nov. 306, Giluss\nees 07, 307-3095 5 39 1.312; porrigentiformis 71, 99, (illus.) 100, 101- MOO ZOOP Os 214. 2154 219% 220, 224, 283, 287, 288, 291, 294, (map) 296, 297, 298, 303, 304, 308-312; xproctoriana T. Rich, hybr. nov. 370, (map) 221, 226, 227, (illus.) 227; xproctoris nom.nud. (nap) I> 226, 227, ‘allus:)) 227: pseudofennica 71, 124, 213; xrobertsonii T. Rich, hybr. nov. 370, (map) 221, 224, Tyas) 225, 226,-294: rupicola 70, NOG ZO Glhiss) Qi 214 2155, 287, 293, 308-312; rupicoloides L. Houston & T. C. G. Rich, sp. nov. 291, 284, 286-288, (illus.) 292, 292-294, (map) 297, 297-299; saxicola T. C. G. Rich, sp. nov. 307, 303, 50803509) (allus.)) 310,31 312: sealaris 413 227, (illus.) 227; scannelliana T. C. G. Rich, sp. nov. 213, (illus.) 212, 214; stenophylla M. C. F. Proctor, sp. nov. 214, 209, 215; stirtoniana T. Rich, sp. nove. 215; subeuneata “Ji. 208: 32 mm) populations that predominate in southern England (Fig. 3). Only the ranges given by Harrap & Harrap (2005) match our own observations for both Platanthera species, though we have ‘not yet located any of the short-spurred (mean <23 mm) populations of P. chlorantha reported by all authors except Godfery (1933). Moving on to mainland Europe, Webb (1980, echoed by Davies et al. 1983) gave lower and upper limits to P. chlorantha that were less than those for P. bifolia, apparently having given undue emphasis to the extra- ordinary reversal in this character reported along the Baltic coast by Nilsson (1978, 1985). The ranges fail utterly to accommodate British populations. The remaining authors (Baumann & Kiinkele 1988; Buttler 1991; Bournérias & 8 R. M. BATEMAN AND R. SEXTON Prat 2005; Delforge 2006) succeeded in encompassing variation in both species, but only by offering ranges of spur length that are so broad that they lose any potential for being diagnostic of either species (and even then, Bournérias & Prat failed to accommodate the lower end of the spectrum of spur length recorded by us in both British and Continental P. bifolia). Perhaps the most interesting decision was that taken by Buttler (1991) and Delforge (2006) to separate two putative ecotypes of P. bifolia, partly on the basis of ecological preference and partly on the basis of vegetative vigour and spur length. Delforge (2006) offered slightly more liberal spur-length ranges than Buttler (1991), allowing the northwestern, later flowering, calcifugic ecotype (subsp. graciliflora Bisse, ? = var. robusta Seemen) to extend from 13-23 mm and the southern and eastern calcicolous ecotype to cover the range 20-41 mm. However, calcicolous populations from the UK approximate both the upper boundary of the smaller ecotype and the lower boundary of the larger ecotype (cf. Appendix | and Table 3). In Continental Europe, the four calcicolous populations of P. bifolia measured by Bateman & Rudall in southeastern France occupy the lower part of the range of the calcicolous ecotype, whereas the calcicolous population measured by Sexton in_ the Dolomitic Alps of northeastern Italy shows a typical “‘calcifugic” mean spur length of 17 mm. At best, the delimitation of these infraspecific taxa clearly requires further refinement. The above discussion may appear to have become enmeshed in unnecessarily fine details. However, our aim is to address a_ broader underlying theme. Specifically, spur length is in fact diagnostic of the two northern European species of Platanthera, but only if the latitude of populations is taken into account (Fig. 3). When latitude is not taken into account, the majority of the published ranges of spur length either fail to accommodate a_ substantial proportion of Platanthera populations, if they are too narrow, or fail to suggest any diagnostic potential, if they are too broad. We suspect that similar patterns are commonly reflected in quantitative characters presented in diagnoses of herbarium-based floras. There is no subs- titute for access to geographically and ecologically extensive field data. MUTATION Visual inspection of the data matrix suggested that frequencies of both presumed hybrid- ization and presumed mutation radically reducing spur length were surprisingly low. With regard to putative mutation, of 1876 plants measured, only four plants (0-:21%), apportioned equally between P. chlorantha and P. bifolia, yielded abnormally short spurs (defined as those less than 60% of the mean length for the population). This result contrasts with, for example, data for Gymnadenia conopsea and its close relatives, where the frequency of abbreviated spurs is substantially greater and has been inferred as a likely driver of speciation within the genus (Bateman 2005). Pseudopeloric individuals of P. chlorantha, possessing only simplified sepal-like labella bearing greatly reduced (essentially vestigial) spurs, have previously been reported from Keltneyburn, near Perth (averaging approx- imately 3% of the population: cf. McKean 1982; Bateman 1985) and Kennishead, Glasgow (averaging approximately 6% of the population: Dickson 1990). During the present study in 2007, Sexton identified 21 widely distributed pseudopeloric individuals out of 802 P. chlorantha flowering at Bomains Meadow, Falkirk (1.e. 2°6%). Recent assertions that these pseudopeloric plants could constitute incipient species (Rudall & Bateman 2002; Bateman & Rudall 2006b) have arguably been weakened by Sexton’s observation that the Bomains mutants verge on being male-sterile. We suspect that they are epimutants, reflecting hypermethy]l- lation of .the DNA bases rather than non- synonymous mutation of the bases themselves; a project is underway to test this hypothesis (Bateman, Rudall, Kidner & James, unpublished). HYBRIDIZATION Only five (0-27%) of the measured platantheras were accused by us of being hybrids between the two study species, identified primarily on the basis of their intermediate pollinaria positions. All five plants were reported by Bateman & Rudall: four from Bix Bottom, Oxfordshire (Fig. 2), plus one less certain identification from St Anne’s Chapel, Cornwall (regrettably, no data were obtainable from its gynostemium). Disappointingly, our published request for data from additional mixed SPUR LENGTH IN PLATANTHERA 9 FIGURE 2. Flowers of Platanthera bifolia (A left), P. chlorantha (C right) and a putative hybrid (B centre) from the Warburg Reserve, Bix Bottom, Oxfordshire. Scale bar = 5 mm. Photos: Richard Bateman. populations (Bateman & Sexton 2007) yielded only one new dataset (gathered near Perth, by Sexton), and no putative hybrids were reported from this locality. On the other hand, the Bull’s Cross locality studied by Tony Hughes contained convincing hybrids, though they were not reliably separated from their co- occurring parents. The hybrids recorded from southern England by Bateman & Rudall (Fig. 2), Tony Hughes (Hughes 2007) and Sarah Whild (Whild & Lockton 2007) during the present survey have some significance, as most authors of floras have been uncharacteristically reluctant to accept the occurrence of hybridization between the two species of Platanthera, especially within the British Isles. For example, Godfery (1933) reported an occurrence of a putative hybrid from Sligachan, Skye, but then qualified this observation by noting that the “hybrid” had co-occurred with a pseudopeloric mutant (sensu Bateman 1985). This sceptical view- point was adopted by most subsequent authors, who argued that evidence of hybridization in British platantheras was inconclusive and that most if not all putative hybrids were actually aberrant individuals of one of the parents (cf. Hunt 1975; Sell & Murrell 1996; Stace 1997, in press; Foley & Clarke 2005; Harrap & Harrap 2005). Further questionable arguments frequently deployed against the existence of hybrids included supposed differences in habitat preference, peak flowering period and, most bizarrely, the presumed reduced fitness of the hybrids caused by suboptimal presentation of the viscidia to pollinators. (This is actually an argument against introgression, rather than against hybridization per se, since the key assumption is that primary hybrids could only act as pollen recipients, not as pollen donors [Nilsson 1983]. However, recent field obser- vations question this assumption, as moths have been photographed removing pollinia from hybrids: Claessens ef al. in press.) In adopting this sceptical viewpoint, these authors were echoing Charles Darwin, who took considerable exception to the decision of Bentham & Hooker (e.g. 1886) to treat the two British platantheras as mere varieties of a single species — a decision taken on _ the grounds that “intermediate” forms occurred between them (cf. Summerhayes 1951). Darwin was driven to uncharacteristically waspish exaggeration in his opposition, stating that “the two forms differ in a large number of characters, not to mention general aspect and the stations inhabited”, and that “these two forms certainly differ from one another more than do most species belonging to the same genus’. Having noted that loose waxy scales on the heads of visiting moths most likely confine placement of pollinia to the proboscis (P. bifolia) or eyes (P. chlorantha), Darwin (1877, pp. 73-4), apparently prompted in part by Muller (1865), concluded that there is no “doubt that the Larger and Lesser Butterfly Orchids are distinct species, masked by close external similarity” (thereby somewhat undermining his own earlier statement regarding their appreciable morphological distinctness). Given the continuing uncertainties over the nature of morphological intermediates between these species, the surprising failure thus far of molecular systematists to identify genetic markers that reliably separate the two species is particularly unfortunate — indeed, it once again raises questions regarding the now generally accepted status of these taxa as separate species (Bateman 2005; Bateman ef al. in prep.). 10 R. M. BATEMAN AND R. SEXTON Bs e bifolia (shade) = 1000 seta — northwest OPE a enoRet) © ae Scatland m@ chlorantha (shade) S SOU MeL \ oO chlorantha (open) D Scotland Cig ic , Pallieae \ @ hybrids (shade) = ‘ : g enianee a ae indeterminate (shade) = 400 \ oO D { - N = | © 200 southern = = is > A . Z England co 3 O =e hvu ® O = O tp -200 GS Austrian/ K i eG Italian Alps : ag 100 S southeast | : B -600 France ® 0 5 10 15 20 25 30 35 40 mean spur length (mm) FIGURE 3. Mean spur length (mm) regressed against latitude for populations of Platanthera bifolia (ett, n = 38, r = 0-56) and P. chlorantha (right, n = 43, r° = 0-42). Shade and open habitats are also distinguished. The dashed line connects mean values for both parents and the putative hybrids at the Bix Bottom locality; the two juxtaposed arrowed populations of P. bifolia show evidence of introgression. The Alpine population of three flowering plants labelled “indeterminate” was reported as P. bifolia but has a mean spur length more characteristic of P. chlorantha; this, together with the hybrid group, was omitted from the regression analysis. Returning to the present study, the putative hybrids from Bix had spur lengths intermediate between those of the parental species co- existing at the two relevant localities (Appendix |, Fig. 3). Indeed, we suspect that hybridization in the Bix and Bull’s Cross populations may be more extensive than we originally believed, since spur-length measure- ments for supposedly pure plants of P. bifolia in these two populations are the longest of any in southern England (Fig. 3), suggesting the genetic > influence” “or co-occurmne 2 chlorantha. Similar observations of spur intermediacy in hybrids were made on larger numbers of putative hybrids of P. chlorantha x bifolia (= P. x hybrida Bruegger) along the Baltic coast of Sweden by Nilsson (1985), though here P. bifolia typically has spurs that are substantially longer than those of P. chlorantha — an extraordinary character reversal relative to the more typical condition (Bateman 2005). Indeed, intermediacy of spur dimensions has characterized all of the many quantitative case-studies of hybridization among spurred European orchids known to us (reviewed by Bateman & Haggar in prep.). This reliable intermediacy renders extra- ordinary the report of extensive hybrid-ization in two mixed populations of Platanthera in the Netherlands, where the putative hybrids show longer average spur lengths (32 mm) than co- existing populations of both P. chlorantha (26 mm) and P. bifolia (23 mm) (Claessens & Kleynen 2006; Claessens et al. in press). It seems likely that the evolutionary processes operating in these ambiguous Dutch populations are more complex than simple introgression. SPUR LENGTH IN PLATANTHERA 1 GEOGRAPHICAL AND ECOLOGICAL TRENDS Prior to the present survey, we wrote that “as expected, spurs of P. chlorantha were typically considerably longer than those of P. bifolia (averaging 34 mm vs 19 mm) ... Data from Scotland (very limited in the case of P. bifolia) suggest that P. bifolia retains its English dimensions there, whereas the spurs of P. chlorantha are significantly shorter” (Bateman & Sexton 2007, p. 61). The addition of substantial datasets from southeast France, the Alps, Cumbria and Scotland requires substantial revision of that preliminary view- point. Regression of mean spur length against latitude (Fig. 3) revealed strong correlation coefficients, especially for P. bifolia (1° = 0:56, vs 0:42 for P. chlorantha), and suggested similar trends in both species; each shows a reduction in mean spur length of approximately 2:2% per 100 km of increasing latitude. Figure 3 also demonstrates that spur length dist- inguishes the two species only if latitude is taken into account, as populations of P. bifolia from the Vercors in southeast France show similar mean spur lengths to populations of P. chlorantha occurring approximately 1300 km to the north in southern Scotland (c. 26 mm). Those occurring a little further north of the Vercors in the Alps also fit into the general trend, though the spread of populations appears relatively wide (even discounting the excep- tional mean length of 35 mm obtained in Pettneu, western Austria, which is rendered suspect by both the unusually small sample size of three plants and uncertain species assignment). The obvious temptation is to interpret these latitudinal trends as reflecting adaptation of the spur to proboscis length in the respective pollinators of these two orchid species. However, data gathered by Bateman and co- workers from all major features of the plants of both species suggest that, at least in southern England, most features of P. bifolia are on average only two-thirds of the dimensions of equivalent features in P. chlorantha (Bateman 2005; Bateman ef al. in prep.). A similar ratio was evident among floral dimensions recorded in south-central Scotland by Sexton (unpublished). If this ratio is repeated across most of the (largely coincident) ranges of the two species, the latitudinal gradient in spur length observed in this study (Fig. 3) could simply represent an allometric ratio across most organs of the plant that is merely reflected in spur length. In this case, the overall size differences could, for example, ultimately reflect the greater relative degree of insolation impacting on the (usually paired) leaves of plants at lower latitudes (though note that, during the summer months when Platanthera leaves are fully expanded, day length at least is longer in higher latitudes). Spur length would then be inferred to be under at most relatively weak and/or sporadic selection pressure (a conclusion reached, on the basis of more sophisticated data, for the North American P. lacera by Little et al. 2005). Furthermore, re-examination of the present data (Fig. 3) suggests that application of more complex algorithms for fitting curves to the spur-length data might ultimately achieve a more exact fit. For example, the spur-length data for P. bifolia could legitimately be inter- preted as showing a constant size of approx- imately 18 mm for most of the latitudinal range, deviating only at the two extreme ends of the latitudinal gradient (i.e. upwards in southeast France and downwards in northwest Scotland). Similarly, data for P. chlorantha could be considered to be constant for most of its latitudinal range, plateauing at approximate- ly 28 mm (a similar figure also characterizes this species in the Low Countries: Claessens & Kleynen 2006), but to have deviated upward to values of 32-37 mm over much of southern England (a mean spur length of 32 mm was also reported for a co-latitudinal population of P. chlorantha from Poland by Stpiczynska 2003a). Positive deviations in the spur length of P. bifolia (mean values 27—34 mm) charac- terize much of Scandinavia, falling to 18-21 mm in the birch forests of the north (Nilsson 1983, 1985; Tollsten & Berstrom 1993; Maad 2000). Moreover, on the Baltic island of Oland there has been divergence between short- spurred populations of P. bifolia in grassland (means 19-23 mm) and those occupying decid- uous woodland (means 28-40 mm) (Nilsson 1983; J. Maad pers. comm. 2007). This observation suggests that, beyond latitude, the other extrinsic factors that could potentially have influenced spur length are those relating to habitat preference, particularly soils (pH and moisture content) and degree of shade experienced by the orchids. Our present data for soils are inadequate, though there exists much anecdotal evidence that P. bifolia is more tolerant than P. chlorantha of wet acidic soils. In contrast, we were able to amalgamate the habitat descriptions listed in Appendix 1 into two broad categories of 2 exposure to light: shaded (woodland and scrub) and open (grassland, heathland and moorland). Among our sampled populations, shaded populations do not extend northward beyond southern England for P. bifolia and beyond northern England for P. chlorantha. In the case of P. bifolia, within specific latitudinal zones, all populations occurring in shade have longer spurs than those occurring in the open (all occur on the positive side of the regression line in’ Fig. 3)> Ghepattemmis lessscleanmion se chlorantha, for which in addition we presently lack data from Continental Europe. In southern England, shaded populations appear to span a wider range of spur lengths than exposed populations (only shaded populations span the upper end of the range of means, at 34-37 mm). However, no significant difference between shaded and open populations was observed in northern England, and no shaded populations were recorded in Scotland. None- theless, potential correlations between average spur length and environmental parameters evidently merit more detailed scrutiny. Lastly, we note that although some of our study populations of both .Platanthera species encompassed several hundred flowering plants, the majority were small, many _ typically yielding less than ten flowering plants in any one year. Where such small effective pop- ulation sizes pertain, deviation from previous distributions of spur length can be readily achieved via genetic drift, whereas the potential for long-term directional or disruptive selection is greatly reduced (e.g. Tremblay et al. 2005). Genetic drift merits consideration in any study of spur-length variation in Platanthera, though its random effects might have been expected to obscure the apparent correlation between habitat preference and mean spur length. POLLINATOR SPECIFICITY REVISITED As noted earlier, past assertions of strong selection pressure on spur length have focused on morphological and ecological data for the two Platanthera species in Scandinavia (Nilsson 1978, 1983, 1985; Maad 2000; Maad & Nilsson 2004), where mean spur length of P. bifolia in particular appears to show greater regional variation than is evident in our measurements taken elsewhere in western Europe. The Swedish data suggest that strong pollinator specificity among different species of sphingid moth provides a meaningful level 2 R. M. BATEMAN AND R. SEXTON of reproductive isolation separating the two species and different geographic/ habitat races within P. bifolia; indeed, the observed mechan- ism has become a model system of plant— pollinator co-evolution (e.g. Hapeman & Inoue IQ): The classical model states that P. bifolia places its pollinaria anywhere along the length of the moth’s proboscis, whereas P. chlorantha requires the moth to ram its head into the column in order to attach the viscidia to the surface of its eyes. This contrast in the mechan- ism of pollinium placement would suggest that spur length should be more critical to the re- productive success of P. chlorantha than to that of P. bifolia. Setting aside sites that yielded very small sample sizes and those showing evidence of introgression, coefficients of var- iation for spur length within our study populations range from 5—18% for P. bifolia (typically approximately 11%) and 6—14% for P. chlorantha (typically approximately 9%: Appendix 1). Thus, if there is a contrast in sel- ective pressure on spur length between the two species it does not appear to be a strong one. Darwin (1877) and subsequent commentators (e.g. Summerhayes 1951; Nilsson 1983) argued adamantly that only long-tongued moths effect pollination in European Platanthera. However, Darwin also persistently rejected observations of other natural historians, notably Miiller (1865), that many European orchids did not actually reward their pollinators, whereas subsequent studies have proven Miiller right; many orchid species succeed by deceiving rather than rewarding pollinators (e.g. van der Cingel 1995;° Neiland & Wilcock 1998; Cozzolino & Widmer 2005). Many authors have noted that species offering genuine rewards to pollinators generally achieve sub- stantially higher frequencies of pollination than do co-existing food-deceptive species (Neiland & Wilcock 1998; Cozzolino & Widmer 2005). However, fewer authors have noted that this comparatively successful pollination rate has not translated into greater numbers of rewarding than non-rewarding species or to greater average size or number of populations per rewarding species. And even fewer authors have noted that there is no obligation on pollinators to accept any reward offered. Surely, an insect that is demonstrably willing to be duped into pollinating an unrewarding orchid will not balk at the prospect of SPUR LENGTH IN PLATANTHERA 13 pollinating a species that offers a reward that the insect cannot satisfactorily exploit? Having thus hypothesized that at least some occasional pollinators of Platanthera would show behaviour patterns that are unlikely to be influenced by spur length, we were briefly excited when our 2007 spur-length survey prompted Tony Hughes (2007) to capture daytime images in Gloucestershire of a distinctive Oedermera beetle transporting pollinia across a putatively hybrid Platanthera inflorescence at Bull’s Cross. Unfortunately, more careful scrutiny suggests that the pollinia in question were obtained from adjacent plants of Neottia (Listera) ovata. Feasibility studies using moth traps at P. chlorantha sites in south-central Scotland showed that moths with pollinia attached to their eyes were caught fairly readily (Sexton & McQueen 2005). Moreover, John Knowler (pers. comm. 2006) reported that catches included the Beautiful Golden Y (Autographa pulchrina), Silver Y (A. gamma) and Gold Spangle (A. bractea); all three moths are prominent pollinators of P. chlorantha in Sweden (Nilsson 1978). Nonetheless, recent combined population genetic and morphological studies of gene flow in other groups of European orchids also counsel caution when assuming strong polli- nator specificity. Even in the notionally most pollinator-specific genus, Ophrys, natural hybrids occur among all major species groups, demonstrating both their interfertility and the fact that pollinators frequently transfer pollinia between species (cf. Paulus & Gack 1990; Devey et al. in press; S. Cozzolino pers. comm. 2007). In addition, the ability to identify pollinia through DNA sequencing (e.g. Soliva & Widmer 2003) means that we can assess pollinator error rates, which even in Ophrys are proving to be relatively high. Also, artificial crosses among Ophrys species have demon- strated fertility in Fls, F2s and backcrossed progeny (S. Malmgren pers. comm. 2006; S. Cozzolino pers. comm. 2007). Current (albeit limited) evidence suggests that P. bifolia and P. chlorantha may similarly introgress when- ever they co-exist in substantial numbers in the same or immediately adjacent habitats. For the present, the jury is out regarding the degree of reproductive isolation enjoyed by the two European Platanthera species. Clearly, it is desirable to gather data on the range and frequency of success of pollinators from other regions of Europe to test the long-standing hypotheses of strong specificity. CONCLUSIONS We have confidently identified latitudinal and habitat influences on mean spur length in Platanthera bifolia and P. chlorantha. Once these factors have been compensated for, P. chlorantha maintained spurs approximately 50% longer than those of P. bifolia throughout our latitudinal transect, though previous studies (Nilsson 1983, 1985; Maad & Nilsson 2004; Boberg et al. 2007; J. Maad pers. comm. 2007) suggest that this apparently linear relationship breaks down in parts of Scandinavia. We recognize that detection of further patterns of regional highs and lows would once again encourage consideration of potential local adaptation of spur length to optimally fit the morphology of co-evolving pollinating moths, and that our observations do not necessarily reject previous selectionist interpretations. However, we do suggest that, in future, potential non-adaptive causes of variation in spur length, relating to allometric relationships with other structures of the plant and/or to vigour of growth influenced by habitat and latitude, are considered alongside more conventional co-evolutionary explanations. The new data have allowed us to formulate more precise questions regarding the evolution of spur length in the study species, even if they have not provided unequivocal answers. Adding further spur-length data, preferably targeted to fill current geographical lacunae in our trans-European sampling, should help to clarify patterns of variation. However, the above hypothesis of a non-adaptive allometric relationship between spur length and other parts of the plant can only be tested effectively by expanding the study to encompass measure- ments of other characters from the same plants that are presently yielding spur-length data. Top priority for such supplementary characters would be metric measurements of the column/ pollinia, labellum and = (typically paired) expanded leaves. Data on the ontogeny of the spurs of Platanthera, and on the underlying developmental genetic mechanisms, would also be welcome. Similarly, data on_ pollinator specificity are needed across the full geograph- ical ranges of both species before meaningful interpretations of the possible relationship between adaptation and speciation can be made. More generally, we hope that we have successfully demonstrated that field-oriented projects recruiting inexperienced researchers 14 R. M. BATEMAN AND R. SEXTON can be successfully extended beyond the conventional limit of simply recording the presence of a particular species, into the realm of quantitatively measuring features of potentially high biological significance. Returning to butterfly-orchids, we note that it is B.S.B.L.-inspired field surveys that identified Platanthera bifolia as one of the most rapidly declining species of flowering plants in the British Isles (Preston et al. 2002; Braithwaite et al. 2005), and led to a subsequent resurvey of its localities in Scotland that relied largely on ACKNOWLEDGMENTS We warmly thank the following members of the Hardy Orchid Society for donating data to the project (number of datasets is given in parentheses): Peter Daltry (1), Ian Denholm (2), Jonathan Fenton (5), Alan Gendle (11), Lois and Nigel Harbron (2), Tony Hughes (7), Christine and David Hughes (15), David Johnston (2), Harold and Jane Lambert (1), Sarah Longrigg (7), Kathy Stott and Dave Pearce (5), Jonathan Tyler (1) and Iain Wright “amateur” botanists (Farrell 2006, Lavery 2007). (1). Several of these contributors showed It would therefore be appropriate if inexpert- considerable inventiveness in designing their ienced researchers also contributed further own supplementary analytical approaches. scientific data with the potential to explain Salvatore Cozzolino, Jean Claessens and those declines. Such an explanation should be Barbara Gravendeel kindly shared _ their facilitated by the co-occurrence of a very closely related species, P. chlorantha, that does not appear to be experiencing a rapid decline and therefore constitutes an ideal comparative yardstick. Comparison of rare and/or rapidly declining species with closely related species that are common and/or rapidly expanding is presently an under-exploited approach to exploring the British and Irish flora. observations on orchid pollination. Richard Bateman thanks the Botanical Research Fund and Botanical Society of the British Isles for partially supporting his fieldwork through small grants, and is grateful to Johanne Maad, Paula Rudall and an anonymous referee for critically appraising the manuscript. REFERENCES BATEMAN, R. M. (1985). Peloria and pseudopeloria in British orchids. Watsonia 15: 357-359. BATEMAN, R. M. (2005). Circumscribing and interpreting closely related orchid species: Platanthera, Dactylorhiza and the crucial role of mutation. Journal of the Hardy Orchid Society 2(4): 104-111. BATEMAN, R. M. & DIMICHELE, W. A. (2002). Generating and filtering major phenotypic novelties: neoGoldschmidtian saltation revisited. In Q. C. B. CRONK, R. M. BATEMAN & J. A. HAWKINS (eds.), Developmental genetics and plant evolution, pp. 109-159. Taylor & Francis, London. BATEMAN, R. M. & HAGGAR, J. (in prep.). Infrageneric and intergeneric hybridization of Dactylorhiza Necker ex Nevski in the British Isles. Botanical Journal of the Linnean Society. BATEMAN, R. M., HOLLINGSWORTH, P. M., PRESTON, J., LUO, Y.-B., PRIDGEON., A. M. & CHASE, M. W. (2003). Molecular phylogenetics and evolution of Orchidinae and selected Habenariinae (Orchidaceae). The Botanical Journal of Linnean Society 142: 1-40. BATEMAN, R. M., JAMES, K. E. & RUDALL, P. J. (in prep.). Morphological versus molecular divergence between two closely related species of Platanthera (Orchidaceae) in southern England: recent evolution with a strong allometric overprint. Annals of Botany. BATEMAN, R. M. & RUDALL, P. J. (2006a). Evolutionary and morphometric implications of morphological variation among flowers within an inflorescence: a case-study using European orchids. Annals of Botany 98: 975-993. BATEMAN, R. M. & RUDALL, P. J. (2006b). The Good, the Bad, and the Ugly: using naturally occurring terata to distinguish the possible from the impossible in orchid floral evolution. Aliso 22: 481-496. BATEMAN, R. M. & SEXTON, R. (2007). Survey of the spurs of European butterfly-orchids. Journal of the Hardy Orchid Society 4(2): 60-63. BAUMANN, H. & KUNKELE, S. (1988). Die wildwachsenden Orchideen Europas. Kosmos, Stuttgart. BENTHAM, G. & HOOKER, J. D. (1886). Handbook of the British flora (Sth edn). Reeve, London. BOBERG, E., ALEXANDERSSON, R., JONSSON, M., MAAD, J., AGREN, J. & NILSSON, L. A. (2007). Pollinator shifts and evolution of spur length in the moth-pollinated orchid Platanthera bifolia. Abstracts of the Sixth Biennial Conference of the Systematics Association (Edinburgh): 69. Box, M. S., BATEMAN, R. M., GLOVER, B. J. & RUDALL, P. J. (in press). Floral ontogenetic evidence of repeated speciation via paedomorphosis in subtribe Orchidinae (Orchidaceae). Botanical Journal of the Linnean Society. SPUR LENGTH IN PLATANTHERA 15 BOURNERIAS, M. & PRAT, D., eds. (2005). Les Orchidées de France, Belgique et Luxembourg (2nd edn). Mezée. BRAITHWAITE, M. E., ELLIS, R. W. & PRESTON, C. D. (2006). Change in the British flora 1987-2004. Botanical Society of the British Isles, London. BRZSOKO, E. (2003). The dynamics of the island population of Platanthera bifolia in the Bierbrza National Park (NE Poland). Annali Botanici Fennici 40: 243-253. BUTTLER, K.-P. (1991). Field guide to orchids of Britain and Europe. Crowood, Swindon, Wilts. CLAESSENS, J. & KLEYNEN, J. (2006). Anmehungen zur Hybridbildung bei Platanthera bifolia und P. chlorantha. Journal Europdischer Orchideen 38: 3-28. CLAESSENS, J. GRAVENDEEL, B. & KLEYNEN, J. (in press). Cucullia umbratica L. als Bestéuber von Platanthera x hybrida Bruegg. in Siid-Limburg (Niederlands). Journal Europdischer Orchideen. CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. (1962). Flora of the British Isles (2nd edn). Oxford University Press, Oxford. COZZOLINO, S. & WIDMER, A. (2005). Orchid diversity: an evolutionary consequence of deception? Trends in Ecology and Evolution 20: 487-494. DARWIN, C. (1877). The various contrivances by which orchids are fertilised by insects (2nd edn). Murray, London. DAVIES, P., DAVIES, J. & HUXLEY, A. (1983). Wild orchids of Britain and Europe. Chatto & Windus, London. DELFORGE, P. (2006). Orchids of Europe, North Africa and the Middle East. A. & C. Black, London. DEVEY, D. S., BATEMAN, R. M., FAY, M. F. & HAWKINS, J. A. (in press). Friends or relatives? Phylogenetics and species delimitation in the controversial European orchid genus Ophrys. Annals of Botany. DICKSON, J. H. (1990). Aberrant Greater Butterfly Orchids (Platanthera chlorantha) in the Glasgow area. Glasgow Naturalist 21: 599-602. ETTLINGER, M. T. (1997). Notes on British and Irish orchids. Published by the author, Dorking, Surrey. FARRELL, L. (2006). Lesser Butterfly Orchid — Platanthera bifolia. SNH-BSBI-Plantlife survey leaflet, pp. 1-4. FOLEY, M. J. Y. & CLARKE, S. (2005). Orchids of the British Isles. Griffin Press, Cheltenham, Gloucs. GODFERY, M. J. (1933). Monograph and iconograph of native British Orchidaceae. Cambridge. HAPEMAN, J. R. & INOUE, K. (1997). Plant—pollinator interactions and floral radiation in Platanthera (Orchidaceae). In T. J. GIVNISH & K. J. SYTSMA (eds), Molecular evolution and adaptive radiation, pp. 433-454. Cambridge University Press, Cambridge. HARRAP, A. & HARRAP, S. (2005). Orchids of Britain and Ireland. A & C Black, London. HOLZINGER, K. E. & WALLACE, L. E. (2004). Bayesian approaches for the analysis of population genetic structure: an example from Platanthera leucophaea (Orchidaceae). Molecular Ecology 13: 887-894. HUGHES, A. (2007). Beetlemania. Journal of the Hardy Orchid Society 4(3): 88-90. HUNT, P. F. (1975). Platanthera Rich. In C. A. STACE (ed.), Hybridization and the flora of the British Isles, p. 488. Academic Press, London. LAURI, R. (2007). The systematic and phylogenetic study of the subgenus Piperia (Orchidaceae) and closely related Platanthera. Botanical Society of America abstracts (Chicago), p. 241. LAVERY, L. (2007). Species Action Framework report on Lesser Butterfly Orchid (Platanthera bifolia) [in Scotland]. Unpublished report to Scottish Natural Heritage. 30 pp. LITTLE, K. J., DIERINGER, G. & ROMANO, M. (2005). Pollination ecology, genetic diversity and selection on nectar spur length in Platanthera lacera (Orchidaceae). Plant Species Biology 20: 183-190. MAAD, J. (2000). Phenotypic selection in hawkmoth-pollinated Platanthera bifolia: targets and fitness surfaces. Evolution 54: 112-123. MAAD, J. & ALEXANDERSSON, R. (2004). Variable selection in Platanthera bifolia (Orchidaceae): phenotypic selection differed between sex functions in a drought year. Journal of Evolutionary Biology 17: 642-650. MAAD, J. & NILSSON, L. A. (2004). On the mechanism of floral shifts in speciation: gained pollination efficiency from tongue- to eye-attachment of pollinia in Platanthera (Orchidaceae). Biological Journal of the Linnean Society 83: 481-495. MCKEAN, D. R. (1982). x Pseudanthera breadalbanensis McKean: a new integeneric hybrid from Scotland. Watsonia 14: 129-131. MULLER, H. (1865). Beobachtungen en westfalischen orchideen. Verhandlungen des Naturalische Vereins fur Province Rheinlandeen Westfalischen Jahrg. 25(3): 36-38. NEILAND, M. R. M. & WILCOCK, C. C. (1998). Fruit set, nectar reward, and rarity in the Orchidaceae. American Journal of Botany 85: 1567-1581. NILSSON, L. A. (1978). Pollination ecology and adaptation in Platanthera chlorantha (Orchidaceae). Botanisk Notiser 131: 35-51. NILSSON, L. A. (1983). Processes of isolation and introgressive interplay between Platanthera bifolia (L.) Rich. and P. chlorantha (Custer) Reichb. (Orchidaceae). Botanical Journal of the Linnean Society 87: 325-350. 16 R. M. BATEMAN AND R. SEXTON NILSSON, L. A. (1985). Characteristics and distribution of intermediates between Platanthera bifolia and P. chlorantha (Orchidaceae) in the Nordic countries. Nordic Journal of Botany 5: 407-419. PAULUS, H. F. & GACK, C. (1990). Pollinators as prepollinating isolating factors: evolution and speciation in Ophrys (Orchidaceae). Israel Journal of Botany 39: 43-97. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. (2002). New atlas of the British and Irish flora. Oxford University Press, Oxford. RUDALL, P. J. & BATEMAN, R. M. (2002). Roles of synorganisation, zygomorphy and heterotopy in floral evolution: the gynostemium and labellum of orchids and other lilioid monocots. Biological Reviews 77: 403-441. SELL, P. D. & MURRELL, G. (1996). Flora of Great Britain and Ireland. 5. Butomaceae—Orchidaceae. Cambridge University Press, Cambridge. SEXTON, R. & MCQUEEN, A. E. D. (2005). Greater Butterfly-orchids: Platanthera chlorantha (Custer) Reichenbach. Forth Naturalist and Historian 27: 77-90. SOLIVA, M. & WIDMER, A. (2003). Gene flow across species boundaries in sympatric, sexually deceptive Ophrys (Orchidaceae) species. Evolution 537: 2252-2261. STACE, C. A. (1997). New Flora of the British Isles (2nd edn). Cambridge University Press, Cambridge. STACE, C. A. (in press). Platanthera Rich. In C. A. Stace (ed.), Hybridization and the flora of the British Isles (2nd edn). Botanical Society of the British Isles, London. STPICZYNSKA, M. (2003a). Floral longevity and nectar secretion of Platanthera chlorantha (Custer) Rchb. (Orchidaceae). Annals of Botany 92: 191-197. STPICZYNSKA, M. (2003b). Nectar resorbtion in the spur of Platanthera chlorantha Custer (Rchb.) Orchidaceae: structural and microautoradiographic study. Plant Systematics and Evolution 238: 119-126. SUMMERHAYES, V. S. (1951). Wild orchids of Britain (\st edn). Collins, London. TOLLSTEN, L. & BERSTROM, L. (1993). Fragrance chemotypes of Platanthera (Orchidaceae): the result of adaptation to pollinating moths. Nordic Journal of Botany 13: 607-613. TREMBLAY, R., ACKERMAN, J. D., ZIMMERMAN, J. K. & CALVO, R. N. (2005). Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84: 1-54. VAN DER CINGEL, N. A. (1995). An atlas of orchid pollination: European orchids. Balkema, Rotterdam. WALLACE, L. E. (2003). Molecular evidence for allopolyploid speciation and recurrent origins in Platanthera huronensis (Orchidaceae). International Journal of Plant Sciences 164: 907-916. WALLACE, L. E. (2004). A comparison of genetic variation and structure in the allopolyploid Platanthera huronensis and its diploid progenitors, Platanthera aquilonis and Platanthera dilata (Orchidaceae). Canadian Journal of Botany 82: 244-252. WALLACE, L. E. (2006). Spatial genetic structure and frequency of interspecific hybridization in Platanthera aquilonis and P. dilata (Orchidaceae) occurring in sympatry. American Journal of Botany 93: 1001— 1009. WEBB, D. A. (1980). Platanthera L.C.M. Rich. In T. G. TUTIN, V. H. HEYwoop, N. A. BURGES, D. M. Moore, D. H. VALENTINE, S. M. WALTERS & D. A. WEBB (eds.), Flora Europaea 5, pp. 331-332. Cambridge University Press, Cambridge. WHILD, S. & LOCKTON, A. (2007). New records. Shropshire Botanical Society Newsletter (Autumn 2007): 4— 5 + cover. (Accepted November 2007) 17 RA 4 4 SPUR LENGTH IN PLATANTHE ased }xau uo ponuluos | xipueddy “Inds 110s Ayfeuondaoxe uv YIM juRTd duo papnfoxe UvOL . :sJaseIep JasIV] JayIO aeor[dnp sjoseyep asay) sv ‘panoyd rou _ :sjuvfd om} JO uo ATUO azts a[dures se ‘payojd jou, 8-Tl g-9 9-6 9-01 r-V1 8-11 coll (%) AD VES 8L-1 €8-1 LOC ESS© I¢-] VN VN Iv: I 79-1 Sal 9-7 6E-€ 79-1 VV-C LO: L6:C SL‘C 6L-¢ dss DPIELG GEC 6l S61 OS-E€ 8e-81 (a) CN ued te ine COG) (St Ol N LiF (Ea Fa 4 7A Zi AZ N pspeys pur]poom pur[poom tpeay tpeay puv[sseis asIvoo Soq/uIeay JOM puvysseis asieo0o pUuv[SseIs asIeoOd pUuv]SseIs asIvOd (Moproul Aey) puvysseis (Mopevoul Avy) purvysseis puv]poom asuap puvysseis adojs Ayoo1 uado ysnyf snoaiesyeo puv][poom poxtu qnios uado puvyssvis asieo0o qn.i9s /puv|SSeIS ISIBOD reyiqey xassng ‘TTTH AUNANALS TOM xossng “TITH WHAANLLMAN sjueH § “aspoT Auusg ‘wonog TTIHda.Ls sjuvH § ‘aspoT Auusq ‘HSN Old JaSIO ‘SUIOQUITAA MAN “Yormdeys ‘ssury AUNGAVA sjuey § ‘preyeuryy “GNA AVA ‘Ari s weyylq [[eMulog ‘yoorstavy ‘fedeyD §.ANNV LS ‘Moprayy $.PIATAS [[e@MuIoD ‘yoostawy, ‘JodeyD § ANNV LS “Mopeay] Ss PIATAS [[eMuslod “yooistavy, ‘TedeyD §.ANNV LS ‘Moprayy $,PIATAS eLOSNYy M “Zuepny{g q UowWY Ig MN WY €T HOATYHEO eLySNYy Ad ‘Zuapn[g q ‘UO I JO SUITYSINO A\N ‘SOOW RLISNY AA ‘Zuepnyg q UoUyY I$ J Wy OT ‘AANLLAd eLasny AM ‘U8ZOG AWN ‘UdJOY preMo? [NAA WOI, AQ peoy BIysny MA U9Z0g MN “TOUNONAO Peol ure ATeyy ‘saUMojog “UaXxtig FS ‘VAVAYOO Jouve] ‘SIODIIA MN ‘SueAoYy AN ‘SHANV SANOL sep X1o1D 20urLy AS ‘810919 JO FS ‘S2q[1D AAN ‘PRPOY ANNVSAAL 0URLY AS ‘S1OII9A JO TS ‘QVIVINVHD § ‘Aris eIUrqoy s0ueLy AS “SIOoIaA JO AS “91d § “"AAVNUVA S Ayyeoo'7T LOOT “991838 C/N LOOT “P-u4RAd C/N YT COO? “epny d/uewmsyeg Y COOT “Tlepny d/uewmoaye gd & poOoT Tepny d/uewaye gq Y COO? “Tlepny d/uewmoyweg Y COO? ‘lepny d/uewmoyweg Y COOT Tepry d/uewsaegd Y COO? “Tepny d/uewmoyeg &Y LOOT ‘seusny V LOOT ‘seusny V LOOT ‘seysny V LOOT “SeysnH V LOOT ‘seyusny V 900T “UOIxaS Y LOOT ‘Tlepny d/uewmaye g Y LOOT ‘Tlepna d/ueusieg Y LOOT ‘Tepny d/uewsieg Y LOOT ‘Tepny d/uewsieg Y (S)a1Vq ‘(S)1apsIOOy eyue.o[Yyo eyUeIo[Yyo BTOFTG BTTOJTG eyUeIo[Yyo eTOsIG SPLqAy eyyUesO[YO BITOFIq MM EIOgr) BTTOFIG d}VUTULIO}OpUT BTOFIG | BORG BOF PTOFIG PTOFIG ETOsIG Sd IV sorsads ACNLILVT ONISVAYONI OL ONIGUODOV GALSIT VYFHINVIVTd HSI GNV HSI Lad dO ACNLS LNASHad FHL Ad GHSSVdINOONA SLNAWAYNSVAW UNdS GNV SHILITVOOT AO STIVLAG "I XIGNdddV puvyssevis L EC:C 61e Ol XxX SSIBOO/GNTOS syong “ANT sudsey NOLNITO NOLSV poo? ‘Hepry d/uewiowg Yo ByURIO[YS puv[poom VN VN rs I XxX uodo/qn.ios syong “UOT UOISY “ANT GNA SYFONVA pOOT “UvUaIeg Yo | ByJURIOTYS 9 or" | SVC $ XxX Gatos SOOT “PNons AN “OLOVA “SSOUO S. TIN LOOT ‘SousnH V 26 SPligay ES) GS | ee. NS XxX puelpoom SONO[H “PNO.NS AN ‘OLOPA “SSO S$. TING LOOT ‘SousnH V BTOFTG VIN SYN SC:0@, |! XxX puv[poom oy A “WeYsIOAR] § “HOIMATAHS MN 99!ddo5 pOOT Tepry d/uvuoweg zy PHLOSTG Ss VN VN fel, || N puryssvis os.e09 SEA “UOISUTART MAS “IH TIAAAAAHD Wos5 PONT TePpry d/Uvuowe gq zy PELOSI ; qnsos m : VN VN YG | N_ /puvysseis as.iv09 AdsIng ‘Ad[SIOH ISOM ‘POOM SVAIdHAHS COOT “Hepry d/uvwuaywg yo, PyJULAOTYO 6-9 CCW? Ke (6 a qnios yyeyo Judy Aq ‘Aunqinuey AS “MNVA IVSNOW LOOT ‘Uosuyor Gq —- ByURIO[YO 7: Gas Gis © A pueypoom usodo Judy Fy weyyeAM ‘WUeG SLEATMOOA ~~ -b/EOOT ‘Wepny d/uvwmawg ys PYUBIOTYO VN VN Oe || N puvysseis Juoy FY “uUeYl” ‘UMOG ALVD MUVd POOT ‘Hepny d/uewawg Yo, PyJULAOTYO VN VN Ol N puvysseis osiv0o Woy A ‘OAM AS ‘SUMO AA M ‘UMOC peolg POOT ‘Tepny d/uewawg Yo | wyyULIO[YS 7 UI See EG aS XxX puv[poom X9SSN§ THH NOLAV TO LOOT P91P9g G/NOIS A BYJURIOTYS Sel SHS SISANG GI x puel[poom X9SSNS “TH AYNANALS TOM LOOT 99189 G/NOS A _-BYJULIO[YS >. 18} (%)AO dSS ueyy N_ popeys wyqey AV[VO0'T (s)ayeq ‘(S)19p.1099x] sorsods 18 “"*ENODO [| XIGNHddV 19 RA — ) SPUR LENGTH IN PLATANTHE ased }xou uo panunuos | xipueddy ‘Inds y10ys Ayyeuondaoxe uv yA yuryd ouO papnfoxa UvOUT . ‘sjaseyep Jose] JIYI0 ajeorfdnp sjaseyep asay) se ‘panojd jou _ :syueyd om) Jo auo ATUO azIs a[dures sv ‘panofd jou | a) CO O’ CO 9-11 rel (%) AD Ihe 6r- on QN : for ANN 00 N © © oA An GN GN eit WMWaADMNNN So | NAN Ct = dss KOIEG S1-9C €8-9C urd 6 Ol N N N N N ZZ Sap Sa a 2 a og nn Sd Sy Sig 7 Sd VA > poepeys PULISSLIS SIRO PpULISSPID ISIVOD pUvISSeID ISIVOD (a)seM) pUv]SSP[S asIvO9 (MOproll) puv[sseis puv]Sseis as1eo0o (MOproll) puv[sseis (puvysseis) puv[sseis pur[poom puvysseis asieoo PpUPISSPIS dS1POD pur]poom pur[poom Mopevoul Avy puv[poom,, puvysseis puvy|ssevis puvy[ssevis aatu AdT[eA pUuvISseIs dS1eOd puvysseis asieo0o pur[poom pur[poom uado pur]poom puvysseis asie0o wiqey oJf4é MAPTRY “Ssou Og “YNT Mopra SNIVINOG StH, MPTP “Ssou Og “UNT Mopeay SNIVINOG St, SUPT “Ssou og “AUNT Mops SNIVINOG oilysuoyrequing seq “NO.LNAY d1TYSuoylequn«d “A “UO VIE QUING “HOOUMUNNWN oirysuoyrequing “M “SSOUdYVO ailysuoyiequing 3seq ‘qIAVOTIN vLiquiny Ys ‘usydaig Aqyry M “Ss8stIUIAIN A GLIVAA BLiquiny Ys ‘usydaig Aqyry M “SsstUIAIN A GLIVAA vliquiny Ys ‘usydaig AqyTy MUNN ATVGUVWNS paeAyoinys YNV AMA BLIquIND FS ‘[epusy § ‘uonounl [6¢Vv/9V BLIquIND FS ‘[epusy AS “uonounl [6S¢V/06SV BLIquUIND YS ‘[epusy AMS ‘uonounl [6¢V/06SV AIVG.LNAG ‘TED stat BLIqUIND FS ‘Tepusey MS POOM SNAAAT asplig KAMAN ‘IIH AuueD BLIquIND AS ‘[Tepuey MS “UNT MOUYVAYALLVI eliquing AS ‘Tepuey MS “UNT MOUNVAYALLVI SSO AT.LNO Arend) ACISGNVS Arend) (OUHLVAWN ArajowlaDd ACISNUV sdoiys ‘Ansamsc ‘uouUTWIOD SK THONATI YOJJNS M WVHNAAVT MA 19218 osplig q ‘poony asvoury shod ‘TPOMOWUD ACAANV TI “surumu05-A 1 AV]eooT LOOT “UOIXES U 9007 “UOIXOS U SCOOT “UOIX9S U LOOT “SSLISUOT § LOOT ‘881ISUOT § LOOT “SSLISUOT § LOOT ‘SSLISUOT S LOOT “UO We, f LOOT ‘AIPUPD V LOOT “UO1g1eH N+T LOOT ‘AIPUEH V LOOT ‘A1PUEH V LOOT ‘AIPUEH V LOOT “UO WA f LOOT ‘A1PUPD V LOOT “UOWS f LOOT ‘A[PUEDH V LOOT “A1PUPH V LOOT “UO f LOOT “AIPUSH V LOOT ‘A1PUSH V LOOT “AIPUSD V LOOT “AIPUSH V FOOT “USI I O07 “J9IAL f/uewayeg y vOOc ‘\Woquie’y7] f+H/uewaeg y (s)ajeq ‘(S)1Op1OSOy -eyyueIO[YO -eyurso[yo eyUeIOTYO vyyueIoyo eyUeIO[YO eyyUeIO[Yyo eyuRIOyYo Prjoyiq “PITOTIQ ByURIO[YS eyUeIOTYo eYyURIO[YS -PyuR1oyyo eyjUeIOyYo eyUBIOTYo ey URIO[YS eyURIO[YS /_RyyURIOTYO eyUeIO[YS BTTOFTG euyuero[ys jeyjueropys veyUeIO[Yo eyUPeIO[Yo eyUeIO[Yo eyyUeIOTYyO >. 18 sorseds R. M. BATEMAN AND R. SEXTON COL SV-e Lee@ Le 68°C 98-1 6V-C Nee II lee SEC EMC > |L OG LO-C tS-C v9-1 8e-1 C1-¢ e/EP It Lvl LS-I LPG OG 99 v6:1 IS-c dss 9V-LC LV-9C 61 60-LC 9-0 8P-81 98°81 EEC 8-LC SEL C81 69-SC KEG $9-6C S0-91 VE-8¢ 90-2¢ CL-81 Iv-LI eS-9C $-8C 76:9C 88°C 86:6C 6:8C uv 9 N 66 N 8 N LC N Cl N VC N Ne N Ol N VC N 8 N ¢ N LV N VC N roa N OC N 9 N SC N I N CC N 61 N 9 N 8C N SL N OC N OC N N Ppepeys puv[sseis Is1vO0d puvysseis as1v09 puvysseis os1v09 (MOpvour) puvysseis pUv]SSeIS dSIBOO puvlsseis as1v09 PpUv]SSeIS ISIBOO puv]sseis dsIBOO puv[sseis as1ev0o (MOPPOUT) puvysseis MOPBOU JOM (MOpeoul) puvyssei3 puy[esseis IsIvOd pUv]SSeIS dSIBOO MOPBOU JOM pUv]SseIs dSIvOD PUP]SSPIS ISIVOD MOPBOU JOM MOPPOUT JOM (MoOpProlw ) puvy[ssevis (MOpeow ) puvy[sseis (MOpeoul) puvy[sseis (MOPPOULT) puvyssevis PUL]SSBIS ISIROD PULISSPIS ISIVOO SUITING M “Jopuryyeo ‘Wwiey ANA TAVUG SUITING AA “Jopuey[eD “Wey XNATAVYA SUTJING “ASINOD JOD YAGNVTIVO dflq ‘SalaD ‘Mopea] SC TNVARONA TA yyod “Mopeay] SOIONO yylag “Mopra SOIONO yyod “Mops SOIONO Yilag “Moprayl SOIONO yilag ‘Mopra SOIONO Uieg ‘auejqung ‘qTaIIHSV SUILING ‘Mopra, MAYONV TIVE SSOUNIM ‘SIH WOM BUTLINS “Plety 9sIOH (VW) NAddIM BUT[INS “Paty e1OH (VY) NAddIM ang 2 [[ASsry “SST uaID BUTLINS (Gq) NAddIM “peoy suuryyH BUTLANS (Gq) NHddIM ‘peoy suuryH SUTLINS “YIVOTVE J01soM BUTLING “YTVOTVE J21som JOOHDS Azewlig NVATd avd AnunoD NVATd waVd Anunoy NVATd wavd AnunoD NVATd STYSSUTpINS “sTwUIeIOID “TTVHAINUId dtTYSuoyre Quind Sb “HOOTIVA yeyqey AV [eoo'T] 9007 “UOIXES Y LOOT “UOIXES UY 9007 “UOIKES U SOOT “UOIX9S Y 9007 UOIXES Y LOOT “UOIXOS U LOOT UOIXES oY 9007 “UOIXeS UY LOOT “UOIXES Y 900T “UOIXES U 9007 “UOIXIS U SOOT “UOIXAS Y LOOT “UOIXES Y QO0T “UOIXES Y LOOT ‘B3LISUOT § 9007 “UOIXES Y LOOT “UOIXES UY QO0T “UOIXES Y LOOT “UOIXIS Y LOOT UOIXES A 9007 “UOIXES A SOOT UOIXSS a LOOT UOIXES A LOO? “88180 § LOOT ‘S8LISUOT S (S)aleq ‘(S)Jap1OIyY -eyjueso[yo eyUeIOTYO BITOJIG eyUeIOTYS ZPIOFIq ZPHOFT BTTOFIG -eyjueIO[Yo vyUeIOyyo eyUeIO[YO EYOFIG eyUe.IO[Yo _eyyue.IO[Yo “ey UPIOTYO BITOJTG -eyue1opys eyUPeIO[YoO PHOS BIOFIG eyyUueIO;Yyo -eyuesropys -eyuesopys eyUPIO [yo eyUPIO[YS eyUe.IO[YO M0 sorsedg “"“ENOD I XIGNHddV SPUR LENGTH IN PLATANTHERA Aypeuondaoxe uv YI Jueyd uO papnypoxs uvalU . ‘sjaseyep JosIV] JayIO oJvor|dnp sjosviep asoy) se ‘payoyd jou _ :sjuejd omy eV VN S-Ol (%) AD UvdI\ € N I N Ol N 8C N tl N L N ib N OC N Cl N Cl N L N OC N OC N OC N G N CC N C N OC N 8 N I N roa N N_ pepeys MOpvoul puvyssvis yreoy /Ysn{y snoaresyeo yeoy /YsnyJ snoaiesyeo 30q Assvis yeoy /Ysnyy snoaresyeo pury1oow pury1ooul purvy..0oul puvysseis puv|ssvis SsOq/U oy aM yeoy /Ysnyy snoaresyes purysseis as.1eo0o Mopeatu durep Moproul AYs‘eUl pury.0oul puv[ssvis as1vO0o SIDA PULTSSLIS puvy.0ooul puvssvis wiqry SSOY JOISIM “OMITOOd MN NMOLCIW SSOY JISOMA “AMAYAANI N 9pisproy SSOW AA ‘yooprey AN “oMajood A “AUWSNYAM Yo] Jueq N SSOY M ‘Yoopey AN “9Molood A “AUVSNYAM Yoo] Yue N SSOY ISOM “AANMVW YOOT MN ‘Aresusay Ss. SsOy A ‘osrowr.ig ‘VINONCHOO'T Wuey winapyso'7 AN SsOY JOISOM “ATVGYAMO TA ‘OlepsAiayy SSOY JaIsaA “o[PPJOMO]Y § “HOOT HAN A SSOY IIISIAMA “B[LPIOMOLY S “AHOVNGYO-Y bu yoo] SSOY 1d}S9AQ ‘DIL YOO] S “A TVAVTIV.L A Opispeoy SSOY J0Is9M “AMAHOO INIA “oP ispeod sopliqoH Join “ejnoequeg N “NVNa&VvO SSOY M “SSOUOA TddV § “UUTTINYAY & YooT ssautaAuy ‘AyaUaqy ‘udVIeD Jo 1kOg S ‘NUIVWONA TO SSOUIOISIM “UJIAIOP “NINIATYG snsuy “YVAUOA “OPIspoo A SSOUIOISO AA. ‘UIDAIOI, “VHINV-Y bu UUlog ‘oprsurejunosy od “[[Psunsoy “ANT NUNGAANL TA SSOUIOISI MA “UIIAIOP “HSINYO.LAYUV “Oprspeoy SSOUIOISO MA “WIOAJOY “OUTTRYDO'T “ANTV HOOT ‘p!syoo'y SSOUIOISI AM “UIDAIOY “ANTTVHOO' “opisproy AVVO] “ands y10ys Jo uo Aju oZIs o]duues sev ‘panoyd jou, LOOT “SeysnH O+d LOOT ‘SP4SNH D+ LOOT “seysnyH D+ LOOT “Wyoyuod J[/uvwowg yy LOOT ‘SPuSNH D+ LOOT “Woy J/ueUseg Y LOOT “Sseyusny O+qd LOOT “Seusny D+d LOOT ‘sousny O+qd LOOT ‘seusnH O+d LOOT “Sseysny D+d LOOT “WOYUoG J/urWaIeg xy LOOT “WOyuod [/UBUTO]e | Pal 9007 “WoYUs | LOOT ‘SPusNH O+d LOOT ‘S94SNH D+ - LOOT ‘SeusNH D+ 9007 “Wyoyusq | LOOT “SeusNH O+d LOOT ‘SeuSNH D+ LOOT ‘SPUSNH D+ (S)ayeVq ‘(S)1ops1Odayy PTTOJIG eujueropyo -PIOJIq eIfOyIq erpoyiq BIOFIq BIOFIq BIOFIq eIOFIq BIOFIq BIOFIq eIOFIq BIfofiq eroyiq PuULIO[YS BITOFIQ PHOFIG vyUe.O yo vyuPIo[Yyo euyuero[ys vyUuPIO[Yo MN sorseds¢ Watsonia 27: 23—36 (2008) Small Cow-wheat Melampyrum sylvaticum L.; Scrophulariaceae in England D. J. TENNANT Low Missise Farm, Laverton, Ripon, North Yorks. HG4 3SY ABSTRACT Small Cow-wheat (Melampyrum sylvaticum L.); Scrophulariaceae occurs in Britain as a rare plant in Scotland and in Northern Ireland. It formerly occurred in North Wales and more recently in northern England (Durham, v.c. 66 and Yorkshire, v.cc. 64, 65). The English distribution and ecology of M. sylvaticum are given together with its taxonomy, and the reasons for its decline are discussed. KEYWORDS: ecology, extinction, plant records, rare species, taxonomy. hemiparasitism, INTRODUCTION Small Cow-wheat (Melampyrum sylvaticum L.), Scrophulariaceae (or Orobanchaceae, as redefined by Olmstead ef al. 2001), is an annual, hemiparasitic herbaceous plant of open woodland which has very special ecological requirements. It is a Nationally Scarce Species in Britain (Rumsey 1994) and is classed as Vulnerable in the Vascular Plant Red Data List (Cheffings & Farrell 2005). It occurs very locally in submontane habitats throughout most of the Scottish Highlands and in the Scottish lowlands from where it has now largely disappeared. It still occurs very locally in Northern Ireland and quite recently it was found that it had formerly occurred in North Wales (Rich, 1997). It has also been known since the beginning of the 19thCentury from northern England, in Durham (v.c. 66) and in Mid-west and North-west Yorkshire (v.cc. 64 and 65), where it is now on the verge of extinction or extinct. A record for Staffordshire (v.c. 39) exists, based on an authentic, but probably mis-labelled, specimen. In spite of the distinctive appearance of M. sylvaticum, its characters have been widely misunderstood until comparatively recent times, often by otherwise experienced botanists. This has resulted in M. sylvaticum being included in the published records of very many counties throughout England, whereas in nearly all of these cases a form of the commoner M. pratense L. was intended. The English distribution and records are therefore updated here, details of the ecology of M. sylvaticum in its former English habitats are given and its past taxonomic treatment briefly considered; the reasons for its decline are also discussed. Outside Britain, M. sylvaticum occurs in Europe from Iceland and Scandinavia to the former U.S.S.R., the Alps, the Pyrenees and extends to central Italy, Bulgaria and the Balkans; it is endemic to this area of distribution. TAXONOMY Melampyrum sylvaticum is quite distinctive in its characters, yet surprisingly has previously been widely confused with M. pratense in Britain, particularly in English localities. Smith (1961) summarised and illustrated the main character differences between the two species. A further, more extensive list of the important characters of M. sylvaticum (Table 1) with illustrations (Fig. 1), which has been compiled from the characters of fresh plants growing in situ in its last known English locality, and a comparison with M. pratense, is therefore given below. However, there appear to be no significant differences between plants in this English population and those seen by me in two separate populations in Mid Perthshire (v.c. 88) in the Central Highlands of Scotland. A single gathering of M. sylvaticum from a different locality in Upper Teesdale (v.c. 66) made in 1874 (BM) however, does appear to differ from the plants described in Table 1 and from those collected elsewhere in the north of England, and this is discussed below. Most Scottish and English examples of M. sylvaticum have uniformly rather deep yellow flowers which deepen to orange when withering, although a pale-flowered variant was described locally from Scotland, — var. pallidiflora White, and more variation occurs in mainland Europe (Britton 1943). Melampyrum pratense is more variable in this respect, although the majority of variant forms are often somewhat paler yellow, the colour varying 24 D. J. TENNANT TABLE 1. CHARACTERS SEPARATING MELAMPYRUM SYLVATICUM FROM M. PRATENSE. BASED ON OBSERVATIONS OF M. SYLVATICUM IN SITU IN UPPER TEESDALE, NORTH-WEST YORKSHIRE (V.C. 65), DJT 1975-1980, WITH AMENDMENTS FROM BRITTON (1943), SMITH (1961) Character M. sylvaticum Longest upper bracts except in largest plants Flowers Suberect Calyx lobes mature capsule Corolla Fruit suture Lax to spreading, linear-lanceolate, straight, equal in size and shape, sharply acute; remaining lax and markedly widening in fruit, +equal to or exceeding Typically 8-10 mm long, tapering abruptly to a narrow tube at base, with a pronounced, often bulbous, helmet and wide gape at the mouth; lower lip strongly deflexed with 3 equal, oblong- ovate, obtuse or subacute lobes with distinct, usually rounded basal sinuses Ovoid, rather abruptly tapering to a sharp apical point or beak, somewhat flattened (broadly elliptic in cross-section); 2- seeded, dehiscing by dorsal and ventral M. pratense Almost as long as mid-stem leaves, entire Usually distinctly shorter than mid-stem leaves, dentate except in some small plants Spreading to semi-erect More appressed to corolla, linear-subulate or triangular, upturned, unequal, upper pair usually wider, sharply to very finely acute; remaining +appressed, not enlarging in fruit, shorter than mature capsule 11—20 mm long, tapering gradually to a broader tube at base, helmet less pronounced and gape at mouth narrower; lower lip projecting forward or sometimes deflexed, with 3, often unequal, usually triangular, +acute teeth with narrowly acute basal sinuses Ovoid-obpyriform, more gradually tapering to an apical beak, distinctly flattened (narrowly elliptic in cross- section); 2— 4-seeded, dehiscing by dorsal suturey *+Smith (1961) stated that the fruits of M pratense contain four seeds, however the capsules examined in this study of the variety hians (Druce) contained only two large, unequal seeds (5-0—7:0 x 2-0—3-0 mm), so this character is variable. from rather deep yellow in var. hians (Druce 1885) to pale yellow or bicoloured with a pale yellow or white tube and deeper yellow or occasionally crimson lip. Although the Upper Teesdale plants described above from v.c. 65 appear to match Scottish examples of M. sylvaticum with no significant differences, one specimen collected in Upper Teesdale’ (vic. 66)" by. EF. AS ees (BM) (undated, but collected with others in 1874 from the same locality, ¥2 mile above Middleton Bridge) differs considerably from others collected in Upper Teesdale or elsewhere in England. This specimen is robust, up to 30 cm tall, with broad leaves up to 12 mm wide (described as “succulent”, and clearly fleshy) and with large flowers. This BM specimen had also been noted by C. E. Britton (1939: 96) who linked it with similar specimens which had been collected in Northern Ireland (County Antrim, v.c. H39 and County Londonderry, v.c. H40) and he referred these to M. silvaticum subsp aestivale Ronn. (M. aestivale Ronn. & Schinz) (Britton 1943: 378), but there is no clear inference that Britton ascribed other Teesdale examples of M. sylvaticum to subsp. aestivale, as stated by Rich et al. (1998). C. E. Salmon (1929) sent British material of M. sylvaticum, which had been collected by J. Beanland at Colt Park Wood, Mid-west Yorkshire (v.c. 64) (confirmed in Bradford Naturalists report, December 1930), and other specimens from Upper Teesdale, Durham (v.c. 66) to R. V. So6, who determined all of the plants as M. silvaticum var. subsilvaticum (M._ silvaticum subsp. subsilvaticum (Schinz & Ronn.) Soo, thus linked English material (except the 1874 Lees specimen mentioned above) with Scottish plants (Salmon 1929), and not with the subsp. aestivale. G. C. Druce suggested that M. pratense var. hians Druce, which occurs in northern and western Britain was often confused with M. sylvaticum (Backhouse 1884), and _ this statement has been repeated by later authors, most recently Rich et al. (1998), however the former is quite unlike M. sylvaticum and could MELAMPYRUM SYLVATICUM IN ENGLAND E L J | FIGURE 1. Melampyrum sylvaticum and M. pratense: Flowers, calyx and fruit. A-F. M. sylvaticum, Upper Teesdale (v.c. 65); A—-C. flower; D. Developing fruit; E. Mature fruit; F. Calyx. G-K. M. pratense; G, H, J, L. Var. hians; G, H. Typical flower with projecting lip, Ullswater, Cumbria (v.c. 70) and Londonderry; (v.c. H40):H. Flower with deflexed lip, Upper Teesdale (v. c. 65);J. Flower showing typical lip; L. Mature fruit; I, K. Smaller form, Aberfeldy, Perthshire (v.c. 88); I. Flower; K. Calyx (scale bars 1 cm). 26 D. J. TENNANT not be confused with it by anyone familiar with these two plants. The floral morphology in var. hians 1s typical of that found in other M. pratense variants and its flowers are much larger (13)15—20 mm long) than those found in M. sylvaticum. The only connection between the two is the deepish yellow flower colour and the deflexed lower lip which can sometimes occur in var. hians, as well as the fact that the two taxa have frequently occurred together in northern England. Melampyrum sylvaticum 1s much more likely to be confused with small- flowered variants of M. pratense, but the characters of the calyx, widely gaping flower and the capsule will readily separate such plants. Down-turned corolla lips also occur in some Scottish montane variant populations of M. pratense, notably small plants from parts of the Cairngorm Mountains and sometimes in a wider-spread Scottish montane variant with a crimson lip, referred to as forma purpureum by Smith (1963: 366), and possibly in other variants of M. pratense. DISTRIBUTION IN ENGLAND The authentic English records of M. sylvaticum which are supported by a confirmed specimen seen by me, or based on a reliable source of information are listed below by vice-county. Rich et al. (1998: Fig. 1, p.37) showed no post 1969 English records, although the species persisted here until c. 1990. MID-WEST YORKSHIRE (V.C. 64): Moors above Middlesmoor, June 1886, R. M. Norman (BM) (Rich 1998). Location uncertain, either near Keasden (SD7063), Kettlewell (SD9571) or Nidderdale Head (SE0974), however “moors” a highly unlikely habitat. Specimen confirmed(?). Near Ribblehead, scars under Park Fell at 1000 ft (305 m), 18 July 1891, J. Beanland (Annotation in G. A. Shaw’s copy of Lees (1888) flora (G.A.S. pers. comm.)). Horton in Ribblesdale, July 1891, J. Beanland (Specimen exhibited at B.N.C. meeting, 20 July 1981 (Anon 1891) Ribblehead at 1500 ft (457 m), August 1891, J. Beanland, herb. C. E. Salmon (BM). Ribblehead District. “Still there 1 July 1905” (J. Beanland, B.N.C. card index, July 1905 (Anon 1905). Colt Park Wood (Ribblehead). Specimen sent to C. E. Salmon by J. Beanland in 1907, from this locality. B.N.C. report Dec. 1930 (Anon 1930). West of north slope of Pen-y-ghent, J. Beanland. (F. A. Lees 1939). . Park Scar Scrub (=Colt Park Wood), 1891, J. Beanland (F. A. Lees 1941). All of the Ribblehead records in 1891 and 1905 probably refer to Colt Park Wood, SD7777 at 340-350 m, although Lees’ 1939 description of the locality is vague, and the altitudes given in G. A. Shaw’s annotated flora and on the sheet in C. E. Salmon’s herbarium (BM) may be errors in the annotations. The B.N.C. report by J. Beanland dated 15 December 1930 confirms that the site was Colt Park Wood and there is no evidence that there was any other site in the area. Outershaw (=Oughtershaw, Langstrothdale), 30° July 1904, HH. Foster; in hersbree Cryer (LDS). Outershaw Wood, 30 July 1904, W. Ingham (NMW). Conf. DJT. Buckden, 30 July 1904, J. Beanland (with F. A. Lees). Yorkshire Naturalists field meeting (Naturalist 1904: 279: B.N.C. report November 1904 (Anon 1904). Outershaw Wood, 1905, R. Butterfield (Annotation in G. A. Shaw’s copy of. Lees (1888) flora). Woods about Buckden. (F. A. Lees 1939). Oughtershaw Hall, 1904, J. Beanland (F. A. Lees 1941). These records all refer to Oughtershaw Hall Wood, SD8781, at c. 345 m, and the initial discovery is credited to J. Beanland, although no specimen collected by him here has been traced, nor any further records post 1905 from Mid-West-Y orkshire (v.c. 64). NORTH-WEST YORKSHIRE (V.C. 65): (A) UPPER TEESDALE RECORDS: Winch Bridge (near Bowlees, NY9027, at c. 270 m) (Winch 1805; HAMU: Baker 1906). Winch Bridge, 1844, J. Backhouse junior (BM); 1850, G. S. Gibson (2 sheets BM); Just above Winch Bridge, 1852, A. W. Bennett (BM); Winch Bridge, 1884, J. E. Gmniffith (NMW): 1923, A. J. Wilmott (BM): Teesbank at Winch Bridge 1968, J. M. Mulllin, sighting of single plant (J.M.M. in pers. comm., and not 1969 as hitherto reported. A later visit by DJT & J.M.M. showed that J.M.M.’s exact site, a steep rocky bank immediately above the river by the bridge had been searched carefully by me in the years preceding and following this record, and later, without success, suggesting that this was a stray seedling. Although this record was never officially accepted in the absence of a specimen (A. J. E. Smith and J. Gaman in pers. comm.) it is likely that JMM would have known the true plant. MELAMPYRUM SYLVATICUM IN ENGLAND Z| Holwick, 1853, J. G. Baker (BM): Below Holwick (Baker 1906). Park End Wood, Holwick, NY9225—26, 1888, Yorkshire Naturalists field meeting (Naturalist 1888: 279; 1889: 284); F. A. Lees (undated, period 1871-1883) (LDS). Teesbanks near High Force, NY8828, 1934, herb. H. Stanley Redgrave (BM). Between Middleton and High Force, rocky bank above river at 285 m, NY82, 17 June 1975-c.1990, D. J. Tennant, conf. A. J. E. Smith, herb DJT (Naturalist 104: 70 (1979)). (B.) OTHER VICE-COUNTY 65 RECORDS: West bank of Whitfell (=Whitfield) Gill above the main force, Wensleydale, SD9292, at 400 yards (365 m), 1884, J. Percival, herb. F. A. Lees (KGY). Yorkshire Naturalists field visit to Hawes area, 28 June 1884 (Naturalist 10, 1884-1885: 18: December 1890: 368: Baker (1906). Mill (=Whitfield) Gill, 1892 (?), J. Wheldon (Anon 1892): Baker (1906) Not found here during the Yorkshire Naturalists field meeting in 1934 (Naturalist 1934: 161). Cantley (—Cautley), near Sedburgh (=Sedbergh), 1909, F. Booth (Specimen exhibited at B.N.C. meeting November 1909 (Anon 1909). F. Booth (Vice-president of Bradford Natural History Society at the time) apparently collected the specimen on _ the Yorkshire side of the vice-county boundary with Westmorland (v.c. 69) near Cautley Spout, SD6897, but the specimen has not been traced and there are no other records from this well visited area, creating slight doubt about the record. However, J. Beanland, as president and recorder for B.N.C. at the time knew the species well and “spared no pains to check the correctness of a record” (Naturalist, June 1933: 125). J. Beanland published the record (Beanland 1910) and there was no _ later suggestion by him that this record was incorrect. UPPER TEESDALE RECORDS, VICE-COUNTY (65 OR 66) UNSPECIFIED: The records from various sources all refer to Winch Bridge NY9027, or near Winch Bridge on the boundary of these two vice-counties, but which one is not specified (Records which specifically refer to the island in the River Tees immediately above the bridge, NY9027, have been included in Durham (v.c. 66) records. nSiieeieeAe Lees (BM): 1873, EF. A. Lees (KGY); 1886, W. C. Barton (BM); 1887, C. A. Wright (BM); E. Backhouse (1810, E; BM): 1888, H. E. Fox (BM); 1889, J. A. Wheldon (NMW);: 1892, Miss Hazlehurst (BM); 1895, F. J. Hanbury (BM); 1895, Hanbury (NMW). DURHAM (V.C. 66) All certain Durham records are from Upper Teesdale, either from the north bank of the River Tees or from the island in the river immediately above Winch Bridge, NY9027. The classic site was the south tip of this island (G. G. Graham in pers. comm.) and E. Vachell noted in her diaries (held at NMW) c.1930 that M. sylvaticum was fairly abundant there. There was a sighting here in 1927 by G. W. Temperley (Graham 1988) and a final sighting in 1947 (A. J. E. Smith in pers. comm.). An extensive search in the area in 1959 by A. J. E. Smith, D. H. Valentine and party failed to find it here. Other records are: Winch Bridge. (Winch, 1805; HAMU); 1839, R. B. Bowman (?7HAMU); 1871, J. E. Leefe (BM); 1892, H. E. Fox (E). North bank of the Tees below Middleton, c. NY9524, 1874, F. A. Lees (BM). Tees-side at Breckholm, mile above Middleton Bridge, NY9325, 1874, F. A. Lees (2 sheets BM: LDS); Above Middleton Bridge, 1874, F. A. Lees (BM); Near Dent Bank Cliff by the Tees above Middleton, NY9325, 1874, F. A. Lees (BM); Tees-side 1 mile above Middleton Bridge, NY9325, 1874, F. A. Lees (KGY; LDS). Two miles above Middleton, NY9226, 1883, R. P. Murray (BM). The island, Winch Bridge, 1883, H. T. Mennell (BM); 1883, F. J. Hanbury (BM); 1907, J. F. Pickard (LDS); 1913, Mrs Wedgwood (BM); 1927, herb. J. E. Lousley (BM). Wood by Tees immediately above High Force (Durham) (Record source mislaid, but seen in 1934 nearby on the opposite river bank, so likely to be correct, NY8728 (c. 315 m). ERRONEOUS, DUBIOUS OR UNCONFIRMED ENGLISH RECORDS OF M. SYLVATICUM The majority of these records are almost certainly errors for M. pratense (other records which certainly refer to M. pratense are omitted). In all cases a specimen has not been traced. MID-WEST YORKSHIRE (V.C. 64): Grass Woods, Grassington, SD9865, 1881. (Yorkshire Naturalists field meeting 1881; B.N.C. records, per Malins Smith (Anon 1881; 1895), R. Dewhirst, B.N.C. records (Anon 28 D. J. TENNANT 1895); 1905, W. West & C. A. Cheetham, B.N.C. records (Anon 1905); 1905, J. Beanland (Annotation in G. A. Shaw’s copy of Lees (1888) flora). Several correct records for M. pratense are known from this locality, e.g. 1902, J. F. Pickard (KGY), and a specimen which is named as M. sylvaticum collected here in 1882 by P. F. Lee (NMW, det. D J.T.) is ™. pratense, making all of these records doubtful. However, if the annotation in G. A. Shaw’s flora is correct and J. Beanland really did see it in Grass Woods in 1905, then this record could almost certainly have been correct, but the record was never published. (M. pratense var. hians, which often occurred with M. sylvaticum in v.cc. 64, 65, 66, was known in Grass Woods). Helks Wood, Ingleton, SD6974 (Baines 1840), and repeated in later Yorkshire Floras. In this former locality for Cypripedium both M. pratense var. montanum Johnston and_ var. hians were recorded by F. A. Lees, so it must remain a doubtful site for M. sylvaticum. Kettlewell, Upper Wharfedale, SD9772, 1864, W. Horsfeld (J. Windsor 1873). Very probably an error for M. pratense, but neither species appears to be at the described site now. Feizor Wood, Austwick, SD7668, R. Clapham (undated) (J. Windsor 1873). M. pratense, which I have seen in this area, has other records here Clapham’s record is no doubt an error for this species; about Buckden (F. A. Lees 1939) refers to the correct records at Oughtershaw (v.c. 64). NORTH-WEST YORKSHIRE (V.C. 65): Muker Woods, Swaledale, SD9198. (Naturalist, 181, 1890: 254). W. Whitewell (Naturalist 183, 1890:305 suggested that this record probably referred to M. pratense, which was well known here). DURHAM (V. C. 66): Egleston (=Eggleston), near Barnard Castle, NZ0023: Shipley Wood, NZ0120 (Winch 1805: HAMU), specimen confirmed? Dryerdale (Near Hamsterley, NZ0833), Graham ef al., (1972). Specimen (DHM) requires confirmation. Barnard Castle Woods, NZO1I, 1794). Specimen not traced. Traced specimens require confirmation, but Shipley Wood would have been a likely locality at this early date. (E. Robson NORTHUMBERLAND (V.CC. 67, 68): V.c. 67: Woods near Hexham, F. Scott (Winch 1831); Newbiggin Scar, NY9449,? Vachell, 1930 in Swan (1993). Vie, 68; Lesbury NU2411 1892 \((iueklex; 1893); Cheviot, NT9120, H. Thomas (in Swan NOS): Swan (1993) suggested that all Northumberland records were likely to be errors (for M. pratense). CUMBRIA (WESTMORLAND, CUMBERLAND, V.CC. 69, 70): All records from these two vice-counties were regarded as errors by Halliday (1997), as no authentic specimens have been traced. Records from widespread localities throughout v.cc. 69, 70 are now regarded as errors for forms of M. pratense. This is perhaps surprising as there appear to be numerous favourable habitats and some of the records for M. sylvaticum here were made by very experienced botanists, such as W. Hodgson, W. H. Pearsall, D. A. Ratcliffe and T. G. Tutin, although none of these records are supported by a specimen and more recent searches have revealed only M. pratense at the recorded localities. It 1s suggested below that the more acidic nature of numerous otherwise suitable habitats in Cumbria is the main reason for the absence of M. sylvaticum there. STAFFORDSHIRE (V.C. 39): Lark (=Park?) Edge, North Staffords., ?G.R., July 1891, J. W. White, in herb Ida M. Roper (LDS). The specimen is correctly named (det. DJT), however, assuming the collector was James Walter White there is some discrepancy. White (1912) stated in his Flora of Bristol that M. sylvaticum “was apparently unknown south of Yorkshire”, so he had either forgotten about his 1891 record or the specimen is from some other locality and mislabelled. It is very unlikely that White would have given his only specimen to I. M. Roper, but there are no specimens of MM. sylvaticum in White’s herbarium at Bristol. Although dismissed as errors for M. pratense, records for M. sylvaticum appeared in floras of nearby English counties (Gloucestershire (1948) and Shropshire (1841)), but there are no records in the Flora of Staffordshire (1972). Finally, although many older records of M. sylvaticum clearly referred to M. pratense it 1s perhaps dangerous to reject all the old records MELAMPYRUM SYLVATICUM IN ENGLAND 29 of the former simply because a more recent visit to the locality concerned only revealed the presence of M. pratense. The two species have nearly always occurred together in Britain, and M. sylvaticum appears to be the earlier of the two to have declined or disappeared from its confirmed localities. The above records suggest that other small herbarium collections might be worth examination. ECOLOGY HABITATS Melampyrum sylvaticum in Britain has several different specific ecological needs, which explain its local occurrence and its rarity; a change in any one of these requirements may have led to its rapid decline. Its major ecological requirements can be summarised as: 1. Sensitivity to the level of light. 2. High atmospheric humidity. 3. Soil conditions 4. As an obligate hemiparasitic plant, the presence of suitable host species In Britain, light tree cover, as provided in natural open woodland, at woodland fringes, tree-lined streams in rocky ravines, or tree- lined riverbanks provide the ideal light conditions, and M. sylvaticum has disappeared where the tree canopy has increased to give higher levels of shade. The only British records where it has been found in the absence of trees are very few and in Northern Scotland, where it is recorded on rock ledges at fairly high altitude. In Britain it has always been found under deciduous trees, although elsewhere it also occurs in coniferous forests, and I have seen it in such habitats in several places in the Central Alps. In its last known English locality it occurred under Betula pubescens, Corylus avellana and Sorbus aucuparia. The high level of atmospheric humidity which appears to be necessary for M. sylvaticum seems clear from its chosen habitats in partial shade, but especially on sheltered banks or in rocky ravines close to rapidly flowing streams, and it is almost certainly no coincidence that many known sites are close to a substantial waterfall, where the aerosol effect from spray increases the atmospheric humidity very locally, enhanced by shelter from wind in the narrow gorge carved out at the base of a waterfall. In the absence of streams or rivers other English habitats are known to have nearby flushes from natural springs and are very rich in mosses and hepatics, indicating a significant level of humidity. Soil conditions where M. sylvaticum has been recorded are somewhat acidic but in English localities this invariably appears to be Where rather shallow acidic soils have accumulated over more mineral-rich or base- rich rock formations, such as basalt and/or carboniferous limestone, and particularly where the soil is thinner or bare rocks are exposed, this has led to a herb-rich community in the adjacent area. Where the habitat is close to a river there is no doubt that some of these sites are flushed during times of spate by river sand and grit whereas other sites above the flood zone must only be enriched by seepage from the underlying rock. Soils on woodland slopes are probably enriched by flushing from nearby springs. M. sylvaticum has not been found on more acidic, herb-poor and mineral-poor soils, habitats in which at least some variants of M. pratense often occur, and M. sylvaticum never occurs on open moorland soils amongst Calluna or Vaccinium myrtillus in the absence of trees, where M. pratense is often found. There appear to be numerous suitable habitats for M. sylvaticum in montane parts of the Lake District in Cumbria (v.cc. 69, 70), and M. pratense var. hians, which often accompanies it, is still to be found here, but the more acidic, hard, nutrient-poor underlying rocks may be the main reason why M. sylvaticum has not been found there. There appear to be at least three differing habitats where M. sylvaticum has been recorded in England, although there is some overlap between these. The first of these is ancient upland open woodland, dominated by Betula pubescens, which occurs in England now only as small, isolated relics of former extensive native forest, and it might be speculated that M. sylvaticum once occurred throughout this habitat, but as such woodland has declined to small remnants or the shade from tree canopy has increased, it has had nowhere to go and largely become extinct there. The best example of such ancient woodland in England where M. sylvaticum formerly occurred is probably Park End Wood (v.c. 65), which J. Backhouse (1896) described as a remnant of the ancient Forest of Teesdale. This small wood occurs on a slightly raised pavement of basalt, and in places carboniferous limestone rock, with scattered tree coverage. Geranium sylvaticum and Paris quadrifolia were recorded here, showing the presence of 30 D. J. TENNANT basic habitats, with M. sylvaticum (Naturalist, Sept.—Oct. 1888:279). Rumsey (1994) noted the similarity of some Scottish sites to montane Geranium sylvaticum-rich Birch woodlands found in Norway. In more recent times Park End Wood has been heavily grazed and trampled by cattle and many of its former plant species are no longer present. A second example of such woodland is Colt Park Wood (v.c. 64), a rather unique remnant of woodland situated on a highly raised limestone pavement, the wood being bordered by carboniferous limestone pillars some 3 m high. Many other less common plants have been recorded here, mainly on very shallow limestone soils, including species which are rare in England, such as Actaea spicata, Carex muricata subsp. muricata, Cardamine impatiens, Crepis mollis, Polypodium x font-queri, Potentilla crantzii, and close by, Pseudorchis albida and Pyrola rotundifolia. M. sylvaticum must formerly have occurred here where slightly deeper, more acidic soils had accumulated over the limestone in places where Vaccinium myrtillus and other acid-loving plants can be seen today. There is little doubt that the barrier of vertical limestone rocks surrounding this wood largely prevented access and grazing by sheep, which is responsible for the survival of this woodland and most of its plants. Another remnant of ancient woodland where WM. — sylvaticum formerly occurred is Oughtershaw Hall Wood (v.c. 64), although a small river runs through it and the exact site where M._ sylvaticum occurred here is not known. Pyrola rotundifolia, a rare plant in Yorkshire, occurred here at the time M. sylvaticum was discovered in 1891. A major reason for the decline of M. sylvaticum here was increasing shade from the tree canopy, probably increased by _ the introduction of additional tree species, the site being much more open at the beginning of the 20th Century as photographs show (W. A. Sledge in pers. comm.). Whitfield Gill in Wensleydale (v.c. 65) is another locality which may represent ancient woodland, but is closer to many of the habitats which are found in Scotland where M. sylvaticum occurs. The site where M. sylvaticum occurred here was well described and is a small area of steeply sloping open woodland over limestone with several wet flushes and a small river below, not far from a large waterfall. Cypripedium calceolus once grew here, and other scarce plants in this area, Persicaria vivipara and Pyrola minor are still to be found. The site is especially rich in mosses and hepatics, with over 100 species recorded here, particularly Eurhynchium striatum, Kindbergia praelonga, Rhytidiadelphus triquetrus and many rarer species (Naturalist, 1934: 161). This wood is seemingly unchanged, so the reason for the decline of M. sylvaticum here is not obvious, but there are no records since 1892, so it must have been very scarce then and a slow decline though climatic factors is the most likely reason. The second type of habitat for M. sylvaticum in England is also similar to many of the habitats known in Scotland: that is, a steep rocky bank above a fast flowing stream, often near a waterfall and fringed by light tree coverage. Examples of this habitat in England occur mainly in Upper Teesdale (v.cc. 65, 66) and, if the record was correct, at Cautley near Sedbergh (v.c. 65). The last known English locality in Upper Teesdale (v.c. 65) for M. sylvaticum has been examined in detail by me between 1975 and 1990. This was also found to be surprisingly species-rich, being a very steep bank near a small waterfall with areas of exposed rock, comprised here of Whin Sill (quartz-dolerite, a hard igneous rock) overlying or intruded horizontally into carboniferous (Yoredale) limestone. The small population (c. 20 plants) of M. sylvaticum occurred here on ledges in an area rich in both mosses and grass species. The relatively dry, rather acid soil was situated above the normal flood zone, so soil enrichment would be primarily by seepage of ground water from the underlying rock, although nor far below the population there was evidence of deposition of river sand and grit left during spate conditions. A list of the species accompanying M. sylvaticum here 1s given in Table 2. The species normally occurring on acidic soils were in most cases close associates, whereas close by on shallower soils or on bare rock a richer flora was present, due to the higher level of nutrients there. It can be seen in Table 2 that almost 90% of the closely associated flowering plant species and many of the grasses at these two v.c. 65 sites alone are also recorded from localities of M. sylvaticum in Scotland, showing the close correlation between them in this respect, although no detailed list of bryophytes from Scottish sites has been seen. There appears to be a third type of habitat in England, possibly confined to Upper Teesdale MELAMPYRUM SYLVATICUM IN ENGLAND 31 (v.cc. 65, 66), which is very similar to that which is described above, being a river-bank site, but in this case very clearly well within the zone of frequent river spates, where a high level of river-deposited sand and grit can be seen in the soil. The upper part of the River Tees is especially prone to sudden and extra- ordinarily heavy spate floods, as several tragic incidents record, so it is possible that even the habitat described above where M. sylvaticum was last seen, although situated above the level of the frequent flood-zone may be flushed very occasionally at times of exceptional river spates. The former main site for M. sylvaticum on the island at Winch Bridge, the site where a single plant was seen by J. M. Mullin in 1968, and other sites where specimens were collected downstream from Winch Bridge on_ the Durham banks of the River Tees would all be subject to such conditions and be affected by flood-water on frequent occasions. F. A. Lees described the habitat of a specimen collected above Middleton (v.c. 66) in an annotation on the herbarium sheet in BM as “Tees-side amongst brushwood” and “in sand”, both obviously deposited there by floodwater. It is highly unlikely that so many collections made here, so far downstream from Winch Bridge, merely arose from seeds washed down from that location, and therefore waterside locations amongst deposited silt appeared to be one favoured habitat. It is interesting that the unique Lees specimen in BM described above (under Taxonomy) and ascribed to UM. silvaticum subsp. aestivale Ronn. by C. E. Britton (1943) was collected at this site and presumably in this habitat. Few perennial plants could survive long in such a habitat, but the annual M. sylvaticum may have some advantage there, and the nutrients released by friction from deposited river grit might result in rapid growth. The host plants associated with M. sylvaticum are discussed later in this paper under ‘parasitism’. The following species which are _ less common or rare in Yorkshire and Durham are recorded in more than half of the seven separate localities where Melampyrum sylvaticum occurred with certainty in the counties: Actaea spicata, Cardamine impatiens, Cirsium heterophyllum, Cochlearia pyrenaica, Galium boreale, Geranium _ sylvaticum, Melampyrum pratense var. hians, Parnassia palustris, Poa nemoralis, Persicaria vivipara, Potentilla crantzii, Pseudorchis albida, Rubus saxatilis and Trollius europaeus, and other rare species occur in some of the same localities. POPULATION SIZES Apart from the single Upper Teesdale site in v.c. 65 studied between 1975 and 1990 there are no other specific details of population sizes in England and Wales. However, it can be assumed that all Yorkshire populations (v.cc. 64, 65) apart from those in Upper Teesdale were extremely small and had probably declined there through natural factors by the time of their initial discovery. The three definite sites there, Colt Park Wood, Oughter-shaw and Whitfield Gill yielded only single or few specimens in the short period from 1884 to 1905, with no later records, and probably represented relic populations in that period. The same applies to the two less certain records from v.c. 64, Middlesmoor and Cautley, being only single gatherings in 1886 and 1909 respectively. However, M. sylvaticum was clearly once wide- spread in Upper Teesdale, the records showing that it had occurred by the River Tees from above High Force downstream to below Middleton, and at the beginning of the 19th Century possibly further downstream to Eggle- ston or even as far as Barnard Castle. At least 12 separate collections or accepted sightings were made from four of the known Yorkshire (v.c. 65) sites in Upper Teesdale (Park End Wood, Winch Bridge (2) and DJT’s locality) but there is no confirmation available from these localities on population sizes, other than the single plant recorded only once by J. M. Mullin in 1968 at Winch Bridge and DJT’s locality described above, where over 20 plants in 1975 were reduced to less than 10 by 1982 and these had probably gone by 1990. In Durham (v.c. 66), where all the records are from the north bank of the River Tees or from an island in the same river at Winch Bridge, there appear to be seven distinct localities where at least 30 separate gatherings or sightings were made. Ten of these collections were from a 6 mile stretch of riverbank in the period 1874—1875 and in 1883, but not since, and although there are no details of population sizes it would seem that M. sylvaticum was not infrequent locally there. A detailed search of the north bank of the Tees and adjacent woodland here in 1974—1975 revealed nothing. The remaining 20 or more v.c. 66 collections or accepted sightings were all from Winch Bridge (nearly all from the island immediately above the bridge), where M. sylvaticum was known to have been abundant (e.g. E. Vachell diaries (NMW), but diminished rapidly at the end of the 19th Century. The last collection traced from here was in 1927 (BM) and the final sighting made there was in 1947 (A. J. E. Smith in pers. comm.). Oo i) TABLE 2. SPECIES ASSOCIATED WITH MELAMPYRUM SYLVATICUM IN D. J. TENNANT UPPER TEESDALE (V.C. 65) SPECIES NOTED BETWEEN 1975 AND 1990 BY DJT IN THE TWO MOST RECENTLY KNOWN SITES (NY9027-28) (a) Close associates Anemone nemorosa* Betula pubescens* Campanula rotundifolia* Conopodium majus* Corylus avellana* Dryopteris filix-mas* Galium saxatile* Hyacinthoides non-scripta Hypericum pulchrum* Lathyrus linifolius* Luzula sylvatica* Melampyrum pratense (incl. var. hians) * Mercurialis perennis* Oxalis acetosella* Polygala serpyllifolia* Succisa pratensis* Teucrium scorodonia Viola riviniana* * Grasses Agrostis canina subsp. capillaris+ Agrostis canina subsp. montanat Arrhenatherum elatius* Brachypodium sylvaticum Briza media Deschampsia caespitosa* Deschampsia flexuosay* Elymus caninus Festuca ovina subsp tenuifoliat* Festuca rubra Festuca vivipara* Holcus mollis* Bryophytes: Bartramia pomiformis Calypogeia muelleriana Campylium stellatum Dicranum bonjeanii Dicranum majust* Dicranum scopariumy Diplophyllum albicans Eurhynchium striatumt Hylocomium splendens* Aypnum jutlandicum* Lophozia ventricosa Mnium hornum+ Plagiochila asplenioides Plagiothecium undulatum Pleurozium schreberi* Polytrichum formosum* Pseudoscleropodium purum Rhizomnium punctatum Rhytidiadelphus squarrosus Rhytidiadelphus triquetrus}* Sphagnum quinquefarium Thuidium tamariscinumy* Tortella tortuosa Tritomaria quinquedentata (b) Species close by Allium ursinum* Cirsium heterophyllum Cruciata laevipes Equisetum pratense* Galium boreale* Geranium sylvaticum* Primula vulgaris* Rubus saxatilis* Sanguisorba officinalis Sorbus aucuparia* Trollius europaeus* Grasses: Helictotrichon pratense Melica nutans Poa nemoralis* Species marked + were abundant or dominant, and those marked * have also been reported from Scottish habitats of M. sylvaticum or recorded by DJT there. ALTITUDE both 340-350 m, and Whitfield Gill in V.C. 65 In England the known altitudinal range of at 365m. Melampyrum sylvaticum lies between c. 210 m In Scotland it has been recorded as high as and 365 m, the highest localities being at Colt 760 m (Rich ef al. 1998). Park Wood and Oughtershaw Wood in v.c. 64, MELAMPYRUM SYLVATICUM IN ENGLAND 313) FLOWERING PERIOD In Upper Teesdale, flowering commenced in the second or third week in June according to season, at least a week or more earlier than M. pratense in the same area. There is one herbarium specimen collected in Yorkshire in advanced bud collected in late May, and several herbarium specimens collected in good flower are dated well into August. PARASITISM Melampyrum sylvaticum is an obligate hemi- parasitic annual by means of root haustoria (Rumsey 1994) and obtains both nutrients and water from its host plant. Smith (1963) studied the related M. pratense and found that the host plants seemed to be either trees such as Betula pubescens, Corylus avellana and Sorbus aucuparia, or woody shrubs such as Calluna vulgaris and Vaccinium myrtillus, and although herbaceous plants occurred with M. pratense he concluded that they were unlikely to be involved as hosts, as M. pratense never occurred with these alone, but was always found near either trees or specific woody shrubs. However, Weber (1976) specifically studied M. sylvaticum in Europe and found that herbaceous plants including a grass species (Briza media) and Melampyrum pratense acted as host plants, in addition to the woody shrubs mentioned above. Although M._ sylvaticum occurs on slightly more nutrient-rich soils than M. pratense in some of its habitats, the same tree species, the same woody shrubs, grasses and M. pratense are all usual associates of M. sylvaticum, and one or more of these probably acts as its host plant. Smith (1963) stated that it was not clear whether M. pratense is host- specific but that there was some indication that this might be the case although there is little evidence for this, but if so, this is also likely to apply to M. sylvaticum, which would be an additional reason for its localised occurrence. Smith (1963) also suggested that the reliance of a shallow-rooted plant like M. pratense on woody plants for water, as well as nutrients, was advantageous in times of drought, and it was noticed that associated herbaceous plants had wilted in such conditions whereas M. pratense appeared to be unaffected. The same would no doubt apply to M. sylvaticum, through attachment to one of the tree species or woody shrubs listed above. | REASONS FOR THE DECLINE OF MELAMPYRUM SYLVATICUM It is clear the Melampyrum sylvaticum must have declined markedly by that time that the majority of the earliest English recordings were made in the relatively short period between 1871 and 1905, and that many of the localities then held only relic populations. The decline in this case is most likely to have been over a very long period through two main causes: gradual climate change and a slow reduction of ancient forests through early forestry and grazing by domestic animals. Many other species of open woodland habitats declined or were lost in northern England in the same period. FORESTRY The former extensive ancient upland forests have been reduced in England to small remnants and subsequent grazing has prevented their re-establishment. Melampyrum sylvaticum may have thrived in such woodland habitats, but once such woodland was reduced to small, discontinuous remnants there was nowhere for the species to go. There is also evidence in a few former English localities, as well as some in Scotland, that the introduction of non-locally indigenous trees, especially Beech (Fagus sylvatica), which cast a deeper level of shade, may have led to a more rapid decline of M. sylvaticum. The large-scale planting of coniferous forests which has reduced the number of suitable habitats in Scotland is hardly a factor in England. GRAZING The effects of grazing are still apparent in several former English localities of M. sylvaticum. In Park End Wood (v.c. 65) both grazing and trampling were very obvious through the access of cattle, and the general flora was adversely affected. Extensive sheep farming in the parts of northern England where M. sylvaticum once occurred has prevented the re-establishment of woodland through the grazing of saplings. It is probably no coincidence that M. sylvaticum survived longer in Colt Park Wood (v.c. 64) as this wood uniquely has a high perimeter barrier of vertical limestone blocks, largely preventing access by sheep. In the last known v.c. 65 site in Upper Teesdale of M. sylvaticum a major factor in its rapid decline there between 1975 and 1990 was undoubtedly an increase in the local rabbit 34 D. J. TENNANT population. The immediate area was covered in excavations by rabbits and some plants of M. sylvaticum itself were seen to be damaged (the single specimen taken from this site and its identity confirmed by A. J. E. Smith had been chewed-off at the base by a rabbit). Little could be done at this site to prevent access by rabbits, such as special fencing, as entry was clearly being gained via the river bed in times of low water level. The same may have occurred in other former localities in Upper Teesdale, as there was ample evidence of rabbit damage even on islands in the river which were largely inaccessible at times when water levels were normal. CLIMATE CHANGE Melampyrum sylvaticum may have declined over a very long period through gradual climate change. It is found in Britain only in areas where relatively high rainfall occurs and appears to have a requirement for a high level of atmospheric humidity, so would therefore be vulnerable to a reduction in rainfall. Rainfall records for Yorkshire, kindly provided by William Foggitt, show that there was a more- or-less progressive decrease in precipitation there between 1892 and 1910 when compared with the earlier decade 1882-1891, particularly in summer months. This period coincides with the time that M. sylvaticum disappeared in Mid-West Yorkshire (v.c. 64), however, the reduction in rainfall then was not very dramatic (c. 12%), and no figures were available for a longer period, so this may be coincidental. Two successive, abnormally hot, dry summers occurred immediately after M. sylvaticum was found in Teesdale in 1975, and thereafter the population was noticeably reduced. Whereas hot dry summers have always occurred periodically, the very unusual effect of two consecutive ones may have played a part in adversely affecting germination of seed or establishment of seedling plants. AIR POLLUTION One suggestion for the apparent rapid decline of M. sylvaticum in its English localities in the late 19th Century is the effect of acid rain through industrial pollution, as several other rare species from similar upland woodland declined or were lost in the same period in northern England, but this theory must remain speculative. OVER-COLLECTING In the vast majority of its former English localities over-collection was not responsible for the decline or loss of M. sylvaticum, as it was already an extremely scarce plant in most localities at the time of its initial discovery. However, in Upper Teesdale (v.cc. 65, 66) collecting may have played a part, as there are numerous specimens in many herbarium collections from localities there, as seen in this study. REPRODUCTION Melampyrum sylvaticum produces few large seeds with a coating which is attractive to ants, which transport the seeds (Gibson 1993a, 1993b). In the absence of water, such as in ancient woodland habitats, the large seeds are largely dependent on ants for their dispersal and may be continually moved to more favourable habitats in this manner, such as to woodland fringes or clearings as the level of shade from the tree canopy increases. Thus a decrease in the local ant population, perhaps through disturbance by animals such as rabbits, would have an adverse effect. PARASITISM Melampyrum sylvaticum, as an obligate hemi- parasite, depends on its host plant for both nutrients and water. The loss of the required host plant or plants would result in immediate extinction. The presence of Corylus avellana in many sites where M. sylvaticum has occurred may be of significance, as the former is found on more nutrient-rich, less acidic soils than most other tree species under which M. sylvaticum usually occurs in Britain, namely Betula pubescens, Quercus petraea and Sorbus aucuparia. M. sylvaticum may occur on rather acidic, nutrient-poor soils, but the roots of trees such as Corylus avellana penetrate deeper into underlying mineral-rich strata and this would provide the parasitic Melampyrum with minerals unavailable at the shallow depth which its roots can reach. Melampyrum pratense, which usually occurred with M. sylvaticum in England is still frequent in Cumbria in Northern England, but has declined or been lost in many of its Yorkshire and Durham localities, including those where M. sylvaticum formerly occurred. The reasons for the decline of M. pratense are very probably the same as those for M. sylvaticum, and where it still survives this was because of its relatively larger population sizes. CONSERVATION Finally, it is probably too late to consider conservation measures for Melampyrum MELAMPYRUM SYLVATICUM IN ENGLAND 35 sylvaticum in its recorded English localities, as it has not been seen in any of these very recently, the last sighting being in 1990 at a single site in Mid-west Yorkshire (v.c. 65). However, there may be lessons to be learned that can be applied in Scotland and Northern Ireland, where controllable measures involving forestry activities, woodland management, restriction of grazing, research into the preservation of local ant populations and host- specificity of this species would prove to be beneficial. The remarkable assemblage of plant species which are uncommon or rare in northern England which are recorded in Yorkshire localities where M._ sylvaticum formerly occurred is probably no coincidence, and although many of these species probably occurred in differing habitats there, it shows that these were all special places which would have merited conservation for their rich general botanical diversity. HERBARIA The following herbarium collections were examined. during this study. BM, Natural History Museum, London; E, Royal Botanic Garden, Edinburgh; KGY, Cliffe Castle Art Gallery and Museum, Keighley, West Yorkshire (ncludes the herbaria of J. Beanland, J. W. Carter and F. A. Lees (part), and Bradford Naturalists Club (B.N.C.) records and reports; most of which were formerly held at Cartwright Hall Museum (CMM), Bradford): LDS, Botany Department, University of Leeds. The following were also consulted: BRIST, Herbarium of the University of Bristol; DHM, Botany Department, University of Durham; HAMU, Herbarium of the Hancock Museum, Newcastle-upon-Tyne; NMW, Herbarium, National Museum of Wales, Cardiff, and the Naturalist (Field Meeting reports, plant records and Transactions), Journal of the Yorkshire Naturalists’ Union, Hull. ACKNOWLEDGMENTS I am grateful to Kathy Fallowfield, Bill Foggitt, Julie Gaman, D. Geedhill, Gordon Graham, Margaret Hartley, Douglas McKean, T. F. Medd, Mike Mullin, David Pearman, Tim Rich, Martin Sanford, Mark Seaward, G. A. Shaw, Arthur Sledge, A. J. E. Smith and Rosalind Smith for information provided, the keepers of the herbarium collections for access to specimens, Gordon Graham and Alan Orange for help with the determination of grasses and bryophytes respectively and Fred Rumsey for helpful comments on the manuscript. REFERENCES ANON. (1881, 1891, 1895, 1904, 1905, 1909, 1930). Bradford Naturalists Club (Records and Reports), Cliffe Castle Museum, Keighley, West Yorkshire. ANON. (1892). Transactions of the Yorkshire Naturalists Union 17, October 1982, Hull. BACKHOUSE, J. jun. (1884). Teesdale Botany: historical and personal recollections. Naturalist Hull 10: 10-13. BACKHOUSE, J. frat. fil. (1896). Upper Teesdale Past and Present [Ed. 2. 1898]. London. BAINES, H. (1840). The Flora of Yorkshire. Longman, London. BAKER, J. G. (1906). North Yorkshire: studies of its botany, geology, climate and physical geography Ed. 2. A. Brown & Sons, London. BEANLAND, J. (1910). Bradford Scientific Journal 23: 346. Bradford Scientific Association, Bradford. BRITTON, C. E. (1939). Notes on Melampyrum. The Botanical and Exchange Club of the British Isles Report for 1938 12: 95-99. BRITTON, C. E. (1942). The genus Melampyrum in Britain. Transactions and Proceedings Botanical Society of Edinburgh 33: 357-379. CHEFFINGS, C. M. & FARRELL, L. 2005. Vascular Plant Red Data List for Great Britain. J.N.C.C., Peterborough. DRUCE, G. C. (1885). A new variety of Cow-wheat; Melampyrum pratense var. hians. [A paper read before the Linn. Soc., London, 1884.] Naturalist Hull, 10: 35-36. EDEES, E. S. (1972). Flora of Staffordshire. David & Charles, Newton Abbot. GRAHAM, G. G., SAYERS, C. D. & GAMAN, J. H. (1972). A checklist of the vascular plants of County Durham. University Dept. of Botany, Durham. GRAHAM, G. G. (1988). The Flora and Vegetation of County Durham. The Durham Flora Committee and Durham County Conservation Trust, Durham. GIBSON, W. (1993a). Selective advantages to hemiparasitic annuals genus Melampyrum of a seed dispersal mutualism involving ants. I. Favourable nest sites. Oikos 67: 334-344. GIBSON, W. (1993b). Selective advantages to hemiparasitic annuals genus Melampyrum of a seed dispersal mutalism involving ants. II. Seed predator avoidance. Oikos 67: 345-350. 36 D. J. TENNANT HALLIDAY, G. (1997). A Flora of Cumbria. University of Lancaster. LEES, F. A. (1888). The flora of West Yorkshire. Lovell Reeve & Co., London. LEES, F. A. (1939). The Vegetation of Craven in Wharfdale. T. Buncle & Co., Arbroath. LEES, F. A. (1941). A supplement to the Yorkshire Floras. C. A. Cheetham, W. A. Sledge (eds). A. Brown, London. LEIGHTON, W. A. (1841). Flora of Shropshire. Van Voorst, London. LUCKLEY, J. L. (1893). Botanical rambles: a flora of Alnwick. Alnwick and County Gazette and Steam Printing Co. (Limited), Alnwick. OLMSTEAD, R. G., DEPAMPHILIS, C. W., WOLFE, A. D., YOUNG, N. D., ELISENS, W. J. & REEVES, P. J. (2001). Disintegration of the Scrophulariaceae. American Journal of Botany 88: 348-361. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. (2002). New Atlas of the British and Irish flora. Oxford University Press, Oxford. RIcH, T. C. G. (1997). Small cow-wheat (Melampyrum sylvaticum L.; Scrophulariaceae) present in Wales. B.S.B.I. Welsh Bulletin 62: 11. RIcH, T. C. G., FITZGERALD, R. & SYDES, C. (1998). Distribution and ecology of Small Cow-wheat (Melampyrum sylvaticum L.; Scrophulariaceae) in the British Isles. Botanical Journal of Scotland 50: 29-46. RIDDELSDELL, H. J., HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. Cheltenham, Cotswold Naturalists’ Field Club. ROBSON, E. (1794). Plantae Dunelmenses, or a catalogue of the more rare Plants which grow wild in the County of Durham, arranged according to the Linnean system. Manuscript folio volume in the library of the Darlington and Teesdale Nat. Fld. Club. RUMSEY, F. J. (1994). Melampyrum sylvaticum L., in A. STEWART, C. D. PRESTON & D. A. PEARMAN eds. Scarce Plants in Britains. Joint Nature Conservation Committee, Peterborough. SALMON, C. E. (1929). Notes on Melampyrum. Journal of Botany 67: 105-107. SMITH, A. J. E. (1961). A note on the differences between Melampyrum pratense L. and M. Sina ie Proceedings of the Botanical Society of the British Isles 4: 151-152. SMITH, A. J. E. (1963). Variation in Melampyrum pratense L. Watsonia 5: 336-367. STACE, C. A. (1997). New Flora of the British Isles (2nd edn). Cambridge University Press, Cambridge. SWAN, G. A. (1993). Flora of Northumberland. Natural History Society of Northumbria, Hancock Museum, Newcastle-upon-Tyne. WEBER, H. C. (1976). Host plants and parasitism in some middle-European Rhinanthoideae (Scrophulariaceae). Plant Systematics and Evolution 125: 97-107. WINCH, N. J., THORNHILL, J. & WAUGH, R. (1805). The botanist’s guide through the counties of Northumberland and Durham Vol. 1. Hodgson, Newcastle-upon-Tyne. T. & J. Hodgson, Newcastle. WINCH, N. J. (1831). Flora of Northumberland and Durham. T. & J. Hodgson, Newcastle. WINDSOR, J. (1873). Flora Cravoniensis. Cave & Sever, Manchester. (Accepted November 2007) Watsonia 27: 37-49 (2008) 37 The distribution, population size and growth of the rare English endemic Sorbus bristoliensis A. J. Wilmott, Bristol Whitebeam (Rosaceae) L. HOUSTON & A. ROBERTSON School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG and T. C. G. RICH* Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP ABSTRACT The distribution, population sizes and growth of the rare endemic Sorbus bristoliensis A. J. Wilmott (Rosaceae) are summarised, based on literature and herbarium studies, and field surveys in 2004 and 2005 using GPS to localise trees. It is endemic to the Avon Gorge, where 262 trees are currently known in woodland, on woodland edges, on open rocks and in disused quarries. It occurs on a range of soils derived from. carboniferous limestone, plateau clays and more acidic sandstone rocks. Data on height and girth show a continuous range of age classes indicating regular regeneration. Tree ring data suggest that the oldest tree is at least 227 years old, and many trees are more than 100 years old. Comparison against 1970s survey data from Leigh Woods shows the population is increasing at c. 10% per decade. Under the 2001 IUCN Threat Criteria it is categorised as “Vulnerable”. KEYWORDS: Gloucestershire, Somerset, conservation. INTRODUCTION Sorbus bristoliensis A. J. Wilmott, Bristol Whitebeam, is a rare English endemic tree confined to the Avon Gorge, Bristol (Wigginton 1999). Flanagan (1998) listed it as declining and in need of recovery, with most of the plants in the Avon Gorge being mature trees with little natural regeneration, and it was regarded as one of the four species of Sorbus with the highest priority for conservation in Britain. *e-mail: tim.rich@museumwales.ac.uk A full history of the discovery of S. bristoliensis 1s in given by Lovatt (2007); here only a summary related to the increase in the number of trees known with time is given. The first record and earliest herbarium specimens of S. bristoliensis are those of Miss M. M. Atwood on 10 June 1852 from the summit of Nightingale Valley, Somerset, in herb. Watson at K (Lovatt 1982; Swete 1854, as Pyrus aria Sm. y intermedia Ehrh.). This tree has been widely suggested (e.g. by Green et al. 2000) to be the multi-coppiced tree by the Stokeleigh Camp viewpoint, Leigh Woods, on the north side of Nightingale Valley (grid reference ST56107328), but a large multi-coppice tree hanging over the edge of the cliffs on the south side of Nightingale Valley below North Road (ST56217305) fits the description of the location, and White (1912, as Pyrus latifolia Syme) noted that only the tree on the southern edge of the valley was known to the early botanists. White (1912) reported that only two or three trees were known until 1901, when five more were found to the north by A. Ley (“Nightingale Valley, south side, at the edge of the rocks nearly opposite the first cross roads. Also on the north side’; BRISTM), and subsequently more were found by J. W. White in 1902 (BRISTM, NMW, UPS; White 1912), but other herbarium specimens suggest it was also found elsewhere despite T. B. Flower’s insistence to White (1912) that there was only one tree. It was reported as first found on the Gloucestershire side in 1909 by White (1912) 38 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH who noted that “a very old tree with several boles overhangs some steep rocks on a thickly wooded slope of Clifton Downs” (a large old tree in a similar location was still present in 2006 but in poor condition). However, there is an earlier specimen in GLR which can be traced to a field trip by W. H. Purchas, T. B. Flower and G. A. St Brody to the Gully in 1869 (Lovatt 2007). When Wilmott (1934) described S. bristoliensis as a distinct species, he noted that there were “many trees, occurring on both sides of the gorge” but he only noted one tree from the Gloucestershire side in Riddelsdell er al. (1948). His type specimen was collected from this tree on Clifton Down in 1933 (BM), and not the tree on the summit of Nightingale Valley as often reported (e.g. Hendry & Pearson 1973). Bailey et al. (1972) and Hendry & Pearson (1973) noted it was present in substantial quantities in Leigh Woods, both within the National Nature Reserve (N.N.R.) and above Quarries |, 2 and 3, with smaller numbers between. Russell (1979, Appendix 3) listed 146 trees (80 mature, 66 regeneration) in the Leigh Woods NNR. Nethercott (1998) suggested that there were probably about 100 trees, mainly on the Leigh Woods side. It is difficult to be certain of population trends given the extent of quarrying and destruction of the original woodland vegetation in the gorge, but given how rare White (1912) regarded S. bristoliensis as and the frequency with which it has now colonised secondary habitats, it has either increased markedly over the last 100 years or its true extent had simply never been established. As part of work on the rare Sorbus species of the Avon Gorge, we have compiled records of S. bristoliensis from herbaria and carried out field work between 1996 and 2005, and here report the first detailed estimate of the distribution and population size. We have also collected some of the data on its growth and performance to assess the population trends. Full details of the records have been lodged with the National Trust, Natural England, Bristol Environmental Records Centre, the Threatened Plants Database and the National Biological Records Centre, Monks Wood. Sorbus bristoliensis is a relatively distinct member of the S. latifolia group, characterised by the obovate, shallowly but sharply lobed leaves with a cuneate base, and the orange ripe fruits (Sell 1989). We have not observed yellow ripe fruits as reported by Sell (1989) and, in exposed sunlit sites, some ripe fruits can be distinctly reddish on one side at least. Sorbus bristoliensis occurs with three other members of the S. latifolia group in the Avon Gorge, S. latifolia (Lam.) Pers. sensu stricto (occasional), S. croceocarpa P. D. Sell (locally frequent), and S$. decipiens (Bechst.) Irmisch (rare), aS well as other Sorbus species; they may be identified using the leaf silhouettes in the Plant Crib (Rich & Jermy 1998), and T. Rich will be happy to determine specimens, which should preferably consist of leaves from the short lateral rosettes, and fruit if present. METHODS Historical records were compiled from herbaria (BIRA, BIRM, BM, BRISTM, CGE, E, GLR, LIV, LTR, MBH, NMW, OXF, RNG and UPS), published and unpublished literature and reports (e.g. Bailey et al. 1972, Hendry & Pearson 1973, Russell 1979, Rich & Houston 1996), and from correspondence with botanists. These records were used to direct field surveys, though most, with the notable exception of Russell (1979), lacked sufficient detail to be any use in the field. RELIABILITY OF GPS READINGS AND GROWTH MEASUREMENTS Small, hand-held, cheap GPS units are being widely used to collect botanical survey data, but there are few published estimates of their reliability for repeating survey work. The reliability of the GPS readings was tested by repeat measurements at the same point by the same and different units, and for some individual trees. The repeatability of height and girth measurements was also assessed for some trees. DISTRIBUTION AND POPULATION SIZE FROM FIELD SURVEY A detailed survey was carried out between August 2004 and October 2005 by the authors, and on 2 October 2004 with help of 13 volunteers from the Somerset Rare Plants Group and the B. S. B. I. The best time to search for trees was found to be late September and early October, when some leaves have blown to the ground, which can then be traced back to the tree. Towards the end of October, too many leaves of all species cover the ground, making the fallen S. bristoliensis leaves less conspicuous and consequently the trees are harder to trace. Fallen fruits, which tend to fall directly under the trees, were also a SORBUS BRISTOLIENSIS 39 useful aid to finding the trees quickly. Efforts were made to avoid recording the same trees twice by temporarily marking the trunks with chalk. Trees and saplings were searched for and their locations pinpointed using five Garmin Etrex units as well one other hand-held GPS. In general it was straightforward to define an individual. We have only observed S. Bristol- iensis suckering once from exposed roots c. | m from the main trunk on the edge of the Gully, and have generally treated separate maiden trunks as separate plants even if very close together. On two occasions, two trees have been observed growing together at the base but with no sign of having been coppiced, so they were treated as maidens. No seedlings or young S. bristoliensis plants less than 0-75 m high were recorded, either because they were difficult to see amongst undergrowth or were difficult to identify, though in general there is good regeneration of most Sorbus species in the Avon Gorge, especially in the open areas. Young saplings were difficult to identify as the characteristic short lateral shoot leaves were not developed, and leaves of young, shaded S. bristoliensis plants are difficult to tell apart with confidence from leaves of young, shaded S. latifolia and S. croceocarpa, both of which are _ also regenerating in the gorge. GROWTH AND PERFORMANCE The height of the tree was estimated by eye, the girth of the largest trunk measured with a tape- measure at 1-3 m above the ground, and notes were made on the growth form (maiden or coppiced) and the presence or absence of fruit. Where ivy could not be removed or prised away from the trunk, a visual allowance was made to correct for the increased girth. Coppiced trees may have originated from deliberate coppicing as part of woodland management, or may be naturally coppiced following loss of a trunk in a storm, etc. (S. bristoliensis, like many Sorbus species, appears to coppice readily from the base of the trunk following damage). Where trees occurred on cliffs or in inaccessible situations, it was not possible to collect some data. Four trees for which girths were measured in 1996 (Rich & Houston 1996) were re-measured in 2004. AGE OF TREES Tree ring data were collected to help assess tree ages and trends in the population. One core was taken from each of 26 trees (22 maidens, 4 coppice) at, with a few exceptions, 1-3 m above the ground using a Haglof 5-15 mm increment borer on 17 March 2005. Trees were selected from a range of growth conditions, from shaded woodland on plateau clays to dry, open rocks, and their girth was noted at the height of coring. Where trees were on slopes or adjacent to wood edges, cores were taken from the side of the trunk parallel to the slope to avoid reaction wood and minimise asymmetrical growth. In total 17 cores reached the middle of the trunk, and nine cores either missed the centre or did not reach it. No core holes were plugged as this is regarded as unnecessary (Grissino-Mayer 2003). In the lab, the annual growth rings were marked with pencil under a microscope and their widths measured with a graduated lens to the nearest 0-1 mm. The rings do not always stand out clearly in this species, and checks showed that rings had occasionally been missed. The most recent rings were the hardest to measure, partly due to their small size In recent years, and partly due to distortion caused by the increment borer. To estimate at what age a sapling would reach 1:3 m tall, six saplings were aged from the annual bud scars. The girth data were also analysed for maiden trees in eight areas (two woodland, two scrub, two natural cliff and two quarry sites) with reasonable populations of trees. The girths were converted to estimated age from the tree ring data and summarised into age class by decade. COMPARISON OF 2004-2005 SURVEY DATA WITH RUSSELL (1979) DATA Russell (1979) mapped 80 mature and 66 regenerating trees in grid squares within Leigh Woods N.N.R. between 1976 and 1978. The 2004-2005 field data for mature and/or fruiting trees were compared with his data to assess loss and gain of mature trees. RESULTS RELIABILITY OF GPS READINGS AND GROWTH MEASUREMENTS For the location and growth data to be useful for monitoring in the future, it is important to have some estimate of their reliability. The 10- figure (i.e. to the nearest metre on the ground) grid references from the Garmin Etrex GPS units were tested for repeatability by recording the coordinates for the gated entrance to Leigh 40 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH V.c. 34 West Gloucester River Avon The Gully V.c. 6 North Somerset North Road 7% | : , \ “Nightingale Valley FIGURE |. Detailed distribution of Sorbus bristoliensis in the Avon Gorge, Bristol, 2004—2005. SORBUS BRISTOLIENSIS 4] Woods on North Road at the start and end of the survey on 2 October 2004. The units reported their accuracy as an average of +8-6 m (range +6-12 m). Relative to the overall average reading at the start of the day, five units varied by an average of 4:25 m east-west (range 2—10 m) and 8:25 m north-south (range 1-18 m). At the end of the day, three of the units retested varied by an average of 4:33 m east-west (range 0-8 m) and 13 m north-south (range 5—24 m) from their individual readings at the beginning of the day. Readings for the same trees, made on different days by different Garmin units (n = 12, mean reported accuracy +15-4 m, range +6—47 m), varied by an average of 20 m east-west (range 2—74 m) and 17 m north-south (range 2-47 m). The accuracy of the GPS unit readings varied depending on the location and canopy cover. Where there was a clear view of the sky and no tree canopy, the accuracy was usually reported on the units as +6—7 m, but under closed canopy in tall woodland, it was much more variable (and often over +20 m), and readings for the individual trees varied markedly. Nonetheless, the use of these simple, cheap GPS units provided significantly more accurate locations than could be obtained from traditional maps, and considerably speeded up localising individual trees. Basic GPS readings of this level of accuracy are unlikely to be of significant use for monitoring individual trees in clustered populations, where other means of localising and identification may also be needed. More expensive GPS units with differential correction could be used to give greater repeatability, but are unlikely to be affordable to many amateur botanists. Repeated height measurements for three trees on different days varied by an average of 0-67 m (range O-1 m); tree heights are notoriously difficult to estimate reliably by eye. Repeated girth measurements on the three trees varied by an average of 3 cm (range 1—4 cm); the variation may be due to a combination of different people measuring at slightly different heights on the tree and the difficulty of keeping the tape level and tight round large trees. DISTRIBUTION AND POPULATION SIZE FROM FIELD SURVEY The distribution of plants is shown in Figure | and the number of plants in each area summarised in Table 1. A total of 262 S. bristoliensis trees are currently known, but there is little doubt that there are more on the less accessible parts of cliffs and disused quarries, and in dense woodland. The map shows that most trees occur on the Leigh Woods side of the Avon Gorge in v.c. 6 North Somerset (234 trees), with the highest densities associated with the natural rock outcrops, the quarry tops and edges, and the change of slope from the valley sides to the plateau. On the Leigh Woods plateau there are scattered trees, the lower density in the northern Forestry Commission area reflecting the woodland clearance and replanting undertaken in that area. A total of 28 trees occur on the east side in v.c. 34 West Gloucester, mainly associated with young woodland. PLANTED TREES There is one planted tree on The Plain, Leigh Woods, and one outside the former Bristol University Botanic Garden in Church Road on the Somerset side (v.c. 6). There is one planted tree on the Promenade near Merchant’s Hall and three trees grafted onto S$. aucuparia L. rootstock on Durdham Down _ on _ the Gloucestershire side (v.c. 34). OTHER RECORDS Specimens: “collected” by. D>” Fry; ~ 1886; Oakleigh, near Keynsham, which had part of a label missing, led Salmon (1899) mistakenly to publish a new locality; it subsequently turned out that the specimens were from a tree in a shrubbery which had been planted by a member of the Fox family (Fry 1899; White OND): GROWTH AND PERFORMANCE Height and girth are highly correlated measurements of growth and performance and, as might be expected, the taller the tree the larger the girth (Figure 2; for maidens alone r = 0-808, n = 170, p<0-001; for all trees r = 0-806, n = 190, p<0-001). There is a weak but not significant trend for stems of coppiced trees to have smaller girths than maiden trees of the same height. For trees where measurements were available, height (n = 241) and girth (n = 200) were analysed separately for the maiden and coppiced trees. The mean height of maidens was 7-7 m + 0-32 ms.e. (n = 226), and of coppiced trees 5-9 m + 0:54 m s.e. (n = 35); these were significantly different (p = 0-046, 2-tailed t-test with sample variances assumed to equal). The mean girth of maidens was 40-6 cm + 2:5 cm s.e. (n = 180), 42 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH TABLE 1. NUMBER OF SORBUS BRISTOLIENSIS TREES AND SAPLINGS OVER 0:75 M HIGH IN EACH AREA, EXCLUDING PLANTED TREES V.C. _—_ Locality No. plants 6 North Road, gardens 2 6 Nightingale Valley, rocks on south side 19 6 Leigh Woods National Trust southern area (including The Plain) 11 6 Stokeleigh Camp 6 6 Stokeleigh Camp, cliffs below viewpoint 48 6 Stokeleigh Camp, valley NE of 1 6 Leigh Woods NT, northern plateau 5) 6 Leigh Woods, Donkey Slide/Goat Path south to railway tunnel 17 6 Leigh Woods, Forestry Commission area 1 6 Leigh Woods, towpath north of Quarry | 9 6 Leigh Woods, north end west of railway 3 6 Leigh Woods, Quarry 2 and top edges Ve 6 Leigh Woods, Quarry 3 (Police Quarry) and top edges 42 6 Leigh Woods, ridge between Quarries | and 2 10 6 Leigh Woods, towpath near Quarry 3 yy 6 Leigh Woods, Quarry 4 and top edges + 6 Leigh Woods, Quarry 5 and top edges 13 6 Leigh Woods, between Quarries 2 and 3 3} 6 Leigh Woods, between Quarries 3 and 4 5 34 Bridge Valley Buttress (Jack’s Hole, Great Fault) 3 34 Bridge Valley Road, above >) 34 Bridge Valley Road, below 3 34 Clifton Down, wooded slopes north of Great Quarry 9 34 The Gully/Walcombe Slade, south side 3 34 The Gully/Walcombe Slade, north side > Total 262 and of coppiced tree stems 31-8 cm + 4°8 cm s.e. (n = 23); these were not significantly different (p = 0-114). Similar proportions of the maidens (50%) and coppiced (47%) trees were fruiting. Figure 3 shows the variation in the estimated heights of the maiden trees, with the proportion of trees in each size class with fruit. The graph shows a skewed distribution with most trees in the 2:1-8 m height categories, and progress- ively fewer larger trees. There are relatively few saplings in the smallest size category up to 2 m tall, either because they were small and overlooked or not identifiable with certainty, or because they may be genuinely rarer. In forest, the tree height tends to match the general height of the canopy, which at the changes of slopes preferred by S. bristoliensis is about 10 m high, and in the taller woodland on the plateau about 15 m high. The proportion of trees with fruit increases as the trees get taller, as might be expected. The smallest tree with fruit was 3 m tall with a girth of 12°5 cm. The largest tree without fruit was 18 m tall with a girth of 78 cm; it was growing in a tall area of dense woodland on the plateau and had no leaves in the canopy. Figure 4 shows the frequency of the different girth sizes (2005 data corrected to 2004). The girth size -classes are again strongly skewed towards the smaller size classes (the smallest size class again being under-represented due to the practicalities of survey) but a large range of girth sizes 1s present, with a few large trees. The proportion of trees with fruit increases as the girth sizes increase, again as expected. AGE OF TREES The tree ring growth data for the 26 trees (c. 10% of the population) are summarised in Table 2. The oldest cores were from tree 23 (107 rings) and tree 26 (103 rings). Individual rings varied in width from 0-1 mm to 5-0 mm, with an overall mean of 1-31 mm + 0-018 mm s.e. The mean ring width for each tree ranged from 0-75 mm (tree 8) to 3:05 mm (tree 25), with an overall average of 1-53 mm + 0-092 mm s.e. a year, suggesting an average increase in girth of 9-6 mm per year. The data for copp- iced and maiden trees were lumped together as a t-test indicated no significant differences (p = 0-46) between them, though this is partly as only four coppiced trees were cored. SORBUS BRISTOLIENSIS 43 160 . ® 140 | 120 | v °¢ 100 + Girth (cm) 25 Height (m) FIGURE 2. Relationship between height and girth of maiden Sorbus bristoliensis trees (n = 170). A best fit line is also shown. 70 | Number of trees Ke) KK oO) CO — — — — is NO G my ae a a iS nD is m G0 s) {& 1 1 iDe — — — SS — sy o © = Be aR Sn on. nS ine) aN [@>) (oe) 8 NO Height (m) FIGURE 3. Variation in height of maiden Sorbus bristoliensis trees. m proportion with fruit. GO proportion without fruit. 44 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH Number of trees Ss 15 10 5) 0 — §6ff =k an he) a) aN on (op) NI foe) Co) = = =i at = | aan aa rtd Te irik ae TT =" 7s T i = N o BE >) Ne) er) aS cn rep) NI fo) Ke) = = ae F a aT fo ro) ro) fo) Oo jo) Oo Oo ro) =n a4 =x a =< fo) aa Nh @ KR on jo) o) ro) ro) Girth (cm) FIGURE 4. Variation in girth of maiden Sorbus bristoliensis trees at 1-3 m (data for 2005 corrected to 2004). w Proportion with fruit. O Proportion without fruit. The age of the individual trees can be estimated from the girth data as the girth and core age data are highly correlated (r = 0-825, p<0-001). On average, the bud scars showed that saplings took 14-8 + 1-6 s.e. years to reach 1-3. m tall (the height at which cores were taken), so 15 years can be added to each age estimated from the cores. Treating the trunks as cylinders, the oldest trees cored, both on the Somerset side, are estimated to be 159+15 = 174 years (tree 26) and 141+15 = 156 years (tree 23). The first tree found on the Gloucestershire side (cf. White 1912) has the one remaining trunk estimated to be potentially 227 years old (currently girth of 93 cm and mean annual ring width of 0-66 mm/year) but the large coppiced base suggests it may be much older. Two trees at the bottom of the woodland slope north of the Great Quarry were estimated to be 160-170 years _ old. Observations in the field show that the assumption the trunks are cylinders may overestimate the age of some trees, as, for the 17 cores which reached the centre, the actual core length to the centre was 87% of the calculated core length (assuming a cylindrical trunk). This is due to the asymmetrical trunk growth in natural conditions and the cores being taken from the side of the tree parallel to the slope. The mean ring growth of trees in woodland was 1:44 mm in width per year, and for trees in the open 1-62 mm per year, but these are not statistically different (p = 0-38), suggesting that other unmeasured factors such as climate, soil depth and soil nutrient status may also be important for growth. The ring data for some woodland trees (e.g. tree 2 from the edge of the ramparts of Stokeleigh Camp) also showed that there could be sudden changes from periods of slow growth to fast growth and vice versa, perhaps related to woodland management, loss of adjacent trees or closure of the canopy. The age distributions of individual trees for four different habitats in eight areas of the gorge are shown in Figure 5. Whilst the age estimates are fairly crude, there are some interesting differences between the habitats. The densely wooded slopes north of the Great Quarry (Figure 5a) have only mature, old trees present (estimated ages 80—170 years) with no regeneration. In contrast, the wooded area of the Leigh Woods plateau (Figure 5b), once pasture woodland, have a broader range of SORBUS BRISTOLIENSIS 45 TABLE 2. TREE-RING DATA FOR 26 CORED SORBUS BRISTOLIENSIS TREES No. rings Mean ring Maiden or Girth (mm) Core length Estimated age Tree coppice at 1:3 m Habitat analysed(mm) measured width(mm) _ of tree (years)* i Coppice 240 wood 31 3) 1-30 +4 D, Coppice 390 wood 52 60 0-95 80 3 Coppice 410 open a3 42 1-47 59 4 Coppice 1010 open 112 62 keg/2 108 5 Maiden 170 open 42 40 1-51 38 6 Maiden 230 open 29 18 1-87 35 7 Maiden 240 wood 3) a2 ONG; 65 8 Maiden 340 wood 38 68 O75 87 9 Maiden 340 open 43 49 E27) 58 10 Maiden 350 open 44 >i 1-25 49 itl Maiden 360 open D2 48 IL Xe) Sil 12 Maiden 540 wood 79 75 1:46 74 13 Maiden 630 wood 106 58 PINS) 62 14 Maiden 770 wood 79 58 ony 120 15) Maiden 780 wood 117 73 log 88 16 Maiden 5 wood 92 68 OF, 133 7/ Maiden 810 open 94 59 EO 83 18 Maiden 870 open 100 1S |95)5) 104 19 Maiden 920 wood 73) 101 135 124 20 - Maiden 980 clearing 119 93 1-34 Syl 21 Maiden 1020 open 136 70 2:05 94 22 Maiden 1080 wood 96 IS leSy7 124 235 Maiden NTS wood 121 107 133 156 24 Maiden 1190 clearing 154 76 Pall 105 25 Maiden 1380 wood 110 45 3:05 87 26 Maiden 1500 wood Sy 103 1250 174 Mean 707 85 64 1:53 89 *Age includes 15 years for the trees to reach 1-3 m, and assumes trunks are cylinders; the age is then calculated from the radius and mean ring width. trees, including some young regeneration, especially along the edges of woodland rides and in some of the slightly more open areas. The open woodland-scrub-grassland areas around the Goat Path and The Gully (Figures 5c, d) both show a range of age classes, with some older trees up to 120 years old. The largely natural cliffs at Stokeleigh Camp and on the south side of Nightingale Valley (Figures 5e, f) show a strongly skewed distribution towards younger plants, with few trees up to 100(—120) years old. This young age distribution may be partly a result of trees on the cliffs often being naturally small and frequently damaged through exposure. The age distribution for Quarries 1 and 2 (Figure 5g) 1s strongly skewed towards young trees to 60(— 80) years old, and indicates colonisation since 1935 following cessation of quarrying. The age distribution for Quarry 3 (Figure 5h) is bimodal but again skewed towards younger trees which are colonising the gentler northern slopes of the quarry. COMPARISON OF 2004-2005 SURVEY DATA WITH RUSSELL (1979) DATA Of the 82 mature trees in total mapped by Russell (1979), it is probable that 61 were refound in 2004-2005. The comparison is difficult because it is not always clear that a mature tree mapped by Russell (1979) is exactly the same as the one refound in the same grid square in 2004-2005, or how Russell defined a mature or a regenerating tree; data for _ his grid squares OF and NE are not specified, and some trees may have been overlooked. Some 23 trees appear to have been lost 46 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH A. Wooded slopes north of Great Quarry B. Leigh Woods plateau area (n=20) (n=9) 3 7 es a) € D) Pr 2 SEZ eo - | = 5 = = 2 = 2 eI i ay x | i | | 0 +— — = 0 4 ie) & ON (oe) — — — — — ihe) Or — — — — — So tS © © © ES ag & Sta (Oo “S (OS eSaacweae Siero sore tS oS ©. 6, ore FKstimated age (years) Fstimated age (years) C. Goat Path area (n=17) D. The Gully (n=8) 4 2 i 3 ma = 5 5 5 37) 5 | = 4 =a i) >) = = os i ¢ ; 0 1 0 Sa NO £ a ow - — = — a oO - — _ = OS Sa Se 6S Sa eo S Se Te SS Sa oe FKstimated age (years) Estimated age (years) E. Stokeleigh cliffs (n=51) F. Nightingale Valley cliffs (n=19) IL) 6 5 > Pm = 10 2 4 5 5 3 S 3 Y 5 oOo 979 ‘= w pay ft Fay fa I 0 sy 0 Vs a i) aN oy Oo _ _ — — — i) on = — = — — Sree "So gor too NY ESF ao\meic So’ oS & 6 Syms Pesce eS & “S “Si yee S. Se eto so Estimated age (years) FKstimated age (years) G. Quarry | and 2 (n=33) H. Sane (n=38) 10 8 8 Se > 6 = 6 5 = = 4 Sa 2 (a i 2 0) i 0 =: i) & en 00 — — — — — i) en = _ = = SY Foe SS Vier WW Neoy, too So. 6 §S So (6) iS) Ves aoe Estimated age (years) —_— Kstimated age (years FIGURE 5. Age distribution of maiden Sorbus br istoliensis trees in eight different areas of the Avon Gorge SORBUS BRISTOLIENSIS 47 between 1976-78 and 2004-2005. We have observed a few trees with heart-rot during our surveys, but have not seen any dead trees for certain. In 2004-2005, we found 118 mature trees in the same area covered by Russell (1979). Allowing for growth of trees since 1976-1978 (i.e. excluding trees with a girth of less than 40 cm, our estimate for 28 years old), we found 53 trees which had matured since that time. Some trees apparently additional to Russell’s survey partly resulted from more detailed survey of the hazardous cliffs below Stokeleigh Camp and on the south side of Nightingale Valley. The net population increase of 30 trees between 1976-78 and 2004—2005 suggests an increase of c. 10% per decade. DISCUSSION The data in this paper are the first detailed population census of S. bristoliensis. The census indicates there are at least 262 plants, and their height and girth data, coupled with the population increase observed in Leigh Woods N.N.R. since 1976-1978, suggest that the population is currently healthy and re- generating. We attribute the absence of the smallest size class of plants to their being largely overlooked during our surveys rather than a sudden and recent lack of regeneration; Russell (1979) found frequent regeneration in S. bristoliensis and there is no reason for this situation to have changed. These new data allow Flanagan’s (1998) assessment of S. bristoliensis as ‘declining and in need of recovery to be updated to ‘increasing and not in need of recovery’. Similarly, the IUCN iihreatwCatesory “(-U:E.N. 2001) of S. bristoliensis can now be_ revised to “Vulnerable” as the population size exceeds 250 plants, updating the “Endangered” assessment in the recent Red Data List by Cheffings & Farrell (2005) which was based on our unattributed, incomplete 2004 survey data and which also lacked our caveats about the data. The age distribution of the trees estimated from the girth measurements indicates that at least four trees from different areas of the gorge are likely to have been present at the time of Miss Atwood’s first record in 1852, suggesting that S. bristoliensis had been present for some time prior to the first record and that her tree was not the first to originate; we cannot however say when it did originate. The number of trees known has increased due to a combination of recruitment and increasing knowledge of its distribution, and it is clear that it has been widely overlooked in the past. Observations in autumn 2005 and autumn 2006 of the holes created by coring in spring 2005 showed that there were often splits in the trunks up to 20 cm long, and once up to | m long associated with the holes (this splitting was not apparent at the time of coring). Few holes had healed fully and a few were still weeping sap by autumn 2006. Observations show that the trees have healed from other wounds and that they survive coppicing, but given this unexpected damage we do not advise coring other rare Sorbus unless it is imperative to obtain such data for their conservation. Sorbus bristoliensis is relatively widespread in the Avon Gorge, where it is probably the second commonest Sorbus after S. aria (L.) Crantz. It is especially characteristic of Leigh Woods on the upper edges of the open, natural rocks and disused quarries and of the broad ledges and screes in the gorge, where it occurs with many woody species including Tilia cordata Miller, Taxus baccata L. and other Sorbus species over a field layer dominated by Brachypodium sylvaticum (Hudson) Beauv. It also occurs on the woodland plateau soils in Leigh Woods where it can occur as a tall tree in ancient woodland with Fraxinus excelsior L., Acer pseudoplatanus L. and Quercus spp. over Rubus fruticosus s.\. and Hedera helix L. On the Clifton side of the gorge S. bristoliensis mainly occurs as scattered trees in secondary woodland and scrub with Quercus ilex L., as much of the original vegetation has been destroyed. At the northern end of the gorge some saplings occur on the acidic sandstone slopes with Quercus spp., Betula spp. and Luzula sylvatica (Hudson) Gaudin (soil pH 4-9). Relatively little detail is known about the ecology of S. bristoliensis. Russell (1979) recorded it on shallow to medium depth soils (2-19 cm deep) with a pH range of (5:3—)6-0- 7:2. One sapling was seen as an “epiphyte” on a pollarded oak. He noted that, although S. bristoliensis occurred on deeper soils than S. aria, it seemed less competitive in most situations. Squirrels, blackbirds, — thrushes, crows, magpies, tits and finches were seen to take the fruits and, compared to S. aria, S. anglica Hedl. and S. wilmottiana Warb., S. bristoliensis had the highest number of viable seeds per fruit in 1976, with a range of (O—)74— 182 seeds per hundred fruits (n = 10, mean 48 L. HOUSTON, A. ROBERTSON AND T. C. G. RICH 1:22 seeds/fruit). Seeds had a high mean germination rate of 59%, and germination was higher in sheltered growth conditions than in open growth conditions. Russell found that drought was the biggest cause of mortality in young plants, followed by invertebrate feeding and vertebrate grazing. Given that virtually all of the trees except those in gardens occur within the statutorily protected areas of the Avon Gorge SSSI and SAC, there are few obvious threats to its long term survival other than mismanagement. Occasional plants have been cleared for public safety or access: Hendry & Pearson (1973) noted that 15 small trees or tall saplings present in May 1969 on the towpath below Stokeleigh Camp had been felled in 1972. Based on past experience with S. wilmottiana (Rich & Houston 2004), loss of some trees during conservation work might also be expected but so far this has not been noted and, once cut down, the trees usually respond by coppicing from the base. The development of closed, tall woodland dominated by shade-tolerant species such as Tilia cordata and Fagus sylvatica could limit S. bristoliensis to the shorter, more open areas of woodland at the changes of slope and on open outcrops (cf. Figure 5a). An increase in browsing from deer populations could limit regeneration (we have observed deer several times in Leigh Woods, but not on the Clifton side of the gorge). Live trees, possibly mostly or all originating from the well known tree at the Stokeleigh Camp viewpoint, are held in cultivation at Bristol University Botanic Garden, Cambridge University Botanic Garden, Ness Botanic Garden, Westonbirt Arboretum, the National Botanic Garden of Wales, and elsewhere. Some seed has been deposited in the Millennium Seed Bank, Wakehurst Place. Sorbus bristoliensis and other rare Avon Gorge Sorbus species were distributed by N. J. Wray as part of the Bristol University Botanic Garden Millennium Tree Project in 1999-2000 to the Forest of Avon Project, Bristol Zoological Gardens, Avon & Somerset Police Head Quarters at Portishead, Bristol University Gardens Department, and elsewhere in the UK to members of Bristol University Alumni and a few private collectors; records were kept of where each species of tree was sent (pers. comm. N. Wray 2006). ACKNOWLEDGMENTS — We would like to thank the Keepers of the herbaria for access to material; Lorne Campbell, Rochelle Campbell, David Cann, Peter Hilton, Nick Hudson, Mark Jannick, Mark and Clare Kitchen, Brian Laney, Clive Lovatt, Liz McDonnell, Pam Millman, Bill Morris, Sharon Pilkington, David Price, Mikhail Semenov, Tony Smith, Tim and Jeanne Webb, and Nick Wray, for help with the 2004 survey and information; and Alex Lockton and the Threatened Plants Database. The Sorbus research project at Bristol University is funded by the Leverhulme Trust. REFERENCES BAILEY, E., HENDRY, G., PEARSON, D. & SHEPHERD, T. (1972). A botanical survey of the distribution and state of the rare plants in the Avon Gorge, Bristol. Part I. Clifton side. Unpublished report, Department of Botany, University of Bristol, Bristol. CHEFFINGS, C. M. & FARRELL, L. (2005). Species status no. 7. The vascular plant Red Data List for Great Britain. Joint Nature Conservation Committee, Peterborough. FLANAGAN, M. (1998). The conservation status of Sorbus in the U.K. — summary of the afternoon discussion. In JACKSON, A. & FLANAGAN, M., eds. The conservation status of Sorbus in the UK. pp. 48-50. Royal Botanic Gardens, Kew. FRY, D. (1899). Pyrus latifolia in N. Somerset. Journal of Botany 37: 488. GREEN, I. P., HIGGINS, R. J.. KITCHEN, C. & KITCHEN, M. A. R. (2000). The flora of the Bristol Region. Pisces Publications, Newbury. GRISSINO-MAYER, H. D. (2003). A manual and tutorial for the proper use of an increment borer. Tree-Ring Research 59: 63-79. HENDRY, G. & PEARSON, D. (1973). A botanical survey of the distribution and state of the rare plants in the Avon Gorge, Bristol. Part II. Somerset side. Unpublished report, Department of Botany, University of Bristol, Bristol. I.U.C.N. (2001). LU.C.N. Red List Categories. Version 3.1. The World Conservation Union, Gland. LovaTT, C. M. (1982). The history, ecology and status of the rare plants and the vegetation of the Avon Gorge, Bristol. PhD thesis, University of Bristol. LovatrT, C. M. (2007). “Only one tree I believe; I can always find it with ease”: Thomas Bruges Flower and the knowledge of Sorbus bristoliensis, Bristol Whitebeam, in the Avon Gorge, 1852-1901. Bristol Naturalists’ Society Bulletin 466: 15-18. © SORBUS BRISTOLIENSIS 49 NETHERCOTT, P. J. M. (1998). Conservation status of Sorbus in the Avon Gorge. In JACKSON, A. & FLANAGAN, M., eds., The conservation status of Sorbus in the UK. pp. 40-43. Royal Botanic Gardens, Kew. RicH, T. C. G. & Houston, L. (1996). Sorbus survey of the Gully, Avon Gorge. September 1996. Unpublished contract report to English Nature, Taunton. RICH, T. C. G. & JERMY, A. C. (1998). Plant Crib 1998. B.S.B.I., London. RICH, T. C. G. & HOUSTON, L. (2004). The distribution and population sizes of the rare English endemic Sorbus wilmottiana E. F. Warburg, Wilmott’s Whitebeam (Rosaceae). Watsonia 25: 185-191. RIDDELSDELL, H. J.. HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. T. Buncle & Co., Arbroath. RUSSELL, R. V. (1979). Avon Gorge National Nature Reserve. The current status and distribution of Sorbus species. Unpublished report, Nature Conservancy Council. SALMON, C. E. (1899). Somerset plants. Journal of Botany 37: 408-411. SELL, P. D. (1989). The Sorbus latifolia (Lam.) Pers. aggregate in the British Isles. Watsonia 17: 385-399. SWETE, E. H. (1854). Flora Bristoliensis. J. Van Voorst, London. WHITE, J. W. (1912). The flora of Bristol. John Wright & Sons, Bristol. WIGGINTON, M. J., ed. (1999). British Red Data Books. 1. Vascular Plants. 3rd ed. Joint Nature Conservation Committee, Peterborough. WILLMOTT, A. J. (1934). Some interesting British Sorbi. Proceedings of the Linnean Society of London 146: 73-79. (Accepted September 2007) a ae oa | ua: ied Saanet reay oe A ee _ : a - i Pate mn Ae pura aul os oak ma OY ¥ wnat a RerY Li ea Pgh Bc ne ie ee a By BL Ae Pee hes at ie e) ay oe Pay i ib - - = . a i- > ~ ™~ y I — ~— Teer po i vl i }. sort ~ = 1 sy Wd oo ey ; Rei IP .),% Watsonia 27: 51-57 (2008) jh Description, ecology and establishment of Carex salina Wahlenb. (Saltmarsh Sedge) — a new British species M. DEAN'AND P. A. ASHTON? NGAS, Edge Hill University, St Helens Rd., Ormskirk, Lancs. L39 4OP K. HUTCHEON 13 Silverknowes Gardens, Edinburgh EH4 5ND A. C. JERMY The Natural History Museum, Cromwell Road, London SW7 5BD and JACAVOUERTIE Agriculture and Agri-Food Canada, Ottawa, Ontario, Canada KIA 0C6 ABSTRACT Carex salina (Cyperaceae) section Phacocystis is a species new to the British Isles, first discovered on a saltmarsh on the west coast of Scotland in 2004. Elsewhere it is found in the northern latitudes of western Europe and the north eastern seaboard of North America. The species is distinct from the closely related C. recta on the basis of habitat, distribution and morphology. It is also morpho- logically distinct from other closely related members of the same section. It is a component of saltmarshes similar in composition to those found elsewhere in its distribution. It has a hybrid origin, with neither parent recorded from the British Isles. It is most likely to have colonised following long distance dispersal, its establishment possibly facilitated by isostatic uplift (i.e. rise in land level due to relaxation of ice age conditions) and recent climatic changes. KEYWORDS: Colonisation, hybridisation, saltmarsh, Scotland. Cyperaceae, INTRODUCTION The addition of a new species to the British flora is a rare event. Hence the discovery of a new Carex (Cyperaceae) species on the west coast of Scotland by one of the authors (KH) in July 2004 and the subsequent identification of the species as C. salina Wahlenb. (Saltmarsh Sedge) by authors JC, MD, ACJ and Arthur Chater is an exciting find and was reported by ‘e-mail: deanm @edgehill.ac.uk “e-mail: ashtonp @edgehill.ac.uk Dean et al. (2005). C. salina is a coastal species with an amphi-Atlantic distribution. It has a hybrid origin, the product of a cross between C. paleacea Schreb. ex Wahlenb. and C. subspathacea Wormsk., neither of which is recorded from the British Isles. Thus its discovery raises interesting questions on the distribution, dispersal and evolutionary potential of the species. This paper details what is currently known of the ecology, identifi- cation and distribution of the British population of the sedge, expanding upon the areas outlined by Dean et al. (2005). Additionally, it includes a detailed description of the sedge in Britain and the characteristics that separate C. salina from its close relatives. It also discusses the possible origins of the population and speculates on the reasons for its establishment. DESCRIPTION AND IDENTIFICATION When seen in its habitat as an extensive colony, C. salina can be confused from a distance with Elytrigia juncea (L.) Nevski. The following description is expanded from Jermy et al. (2007 Jermy et al.) following further field studies in Norway (PA) and Scotland (MD) and on N. American material (JC). Rhizomes far-creeping; shoots tufted; roots rich brown-grey; scales red-brown, often blackened, persistent. 52 M. DEAN, P. A. ASHTON, K. HUTCHEON, A. C. JERMY AND J. CAYOUETTE Stems 16-30 cm, stiff, obtusely trigonous, smooth, spongy, typically shorter than the leaves of the vegetative shoots. Leaves <30 cm (—60—70 cm where more sheltered) x 3—4 mm, mid-green, + glaucous and matt on both surfaces, stiff, V-shaped, upper 3—-¥2 and tip toothed; amphistomous; sheaths herbaceous, pale brown, soon decaying; inner face hyaline, transverse septa distinct; ligule acute, c. 5 mm; apex concave. Inflorescence 4 to almost 2 length of stem; lowest bract leafy, somewhat invaginated at the base, slightly shorter to slightly longer than inflorescence, often spathe-like, enclosing spike. Male spikes 1—2(—3), 1-2 cm long; male glumes 4-6 mm, oblanceolate-elliptic, blackish- brown; margins hyaline; apex obtuse. Female spikes 2—3, 1-3 cm, erect; narrow, cylindrical, upper almost sessile, lowest spike with peduncle up to 1:5 cm, sometimes lax- flowered at base; female glumes 4-5 mm, ovate-elliptic, mostly as wide as the utricles or wider, green at the start of flowering and maturing to mid - to dark-brown with broad pale centre with obvious midrib, some glumes three-veined; margins sometimes asymmetrical in upper florets or tending to be retuse; apex of those in upper florets apiculate, lower florets all awned; awn + | mm long, toothed. Utricles 2:5-3:5 mm X 1-2 mm, ellipsoid, apex acute, slightly papillose with O-3 nerves beneath beak; beak 0:2—0:-4 mm, mostly cylindrical, sometimes conical, lacking spinules around the beak orifice, obliquely truncate or not; stigmas 2; nut orbicular or obovate, apex truncate, biconvex, glossy, usually deeply constricted on one face. Fruiting occurs in June and July. C. salina in Scotland is very sparse flowering. No fertile specimens were found in 2004-2006. Therefore the important characteristics of the truncate apex and constricted nut have not been seen in British material. Low fertility may partly be an intrinsic feature of the species. Other maritime members of section Phacocystis which have a hybrid origin exhibit lowered pollen fertility, partly as a result of disturbed meiosis (Cayouette and Morisset 1985). The few Norwegian populations examined failed to produce fertile seed. However, ecological factors may also influence fertility. Tidal inundation may cause a failure in pollination. Canadian C. salina populations which are covered by tides during summer exhibit sterility. By comparison, Canadian populations covered only, or mostly, by spring and autumn high tides can produce good seeds. Within the British flora C. salina is closest in appearance to C. recta Boott which, however, has a different ecological niche to C. salina. C. recta is found along the edges of the higher reaches of estuaries, usually amongst tall vegetation, e.g. in the Oykel estuary it can be dominant with Phragmites, Phalaris arundinacea and other grasses in species-rich sedge communities of tall-herb fen (S27), with C. rostrata, C. nigra and Juncus acutiflorus (Jermy et al. 2007). In other places (e.g. Bonar Bridge) it may form extensive and dense populations, excluding any other species. By contrast, C. salina is found on the mid-to- llower saltmarsh where it frequently forms single-species stands up to 10 m by 3 m in the saltmarsh creeks. It also creeps into the Puccinellia maritima sward where it has shorter leaves, flowers very infrequently if at all, and is much sparser and could easily go undetected (Dean et al. 2005). | In Britain C. salina and C. recta are also geographically separate, C. recta being located on the northeast coast of Scotland, C. salina on the west. C. recta is a much larger plant typically up to | m tall with vegetative leaves up to 130 cm, whereas C. salina grows up to 30 cm unless the plant is in the shelter of creek banks where the leaves can be longer. Morpho- logically the two species are separated by the female spike length (1-3 cm in C. salina vs 2— 8 cm in C. recta), longer and typically narrower female glumes with a more acuminate apex in C. recta (4—5 mm long in C. salina vs 3—7 mm in C. recta); the beak of C. recta tends to be shorter and narrower (0-2—0-4 mm in C. salina vs c. 0-2 mm in C. recta). Both have apiculate or awned female glumes, with the glumes from the lowest florets usually awned in both species. Two other species within the same section, C. subspathacea and C. vacillans Drejer are broadly similar to C. salina, but a third species, C. paleacea, is relatively easy to distinguish. None of these three species have been recorded from Britain (Dean et al. 2005). C. salina is a larger plant than C. subspathacea, which is typically a very short stemmed (3-15 cm, although in some Hudson Bay and_ north- eastern North American populations it can grow as tall as 25—30cm), narrow leaved (1-3 mm), low growing early flowering species of CAREX SALINA 53 saltmarshes. C. paleacea is a larger species than C. salina but smaller in stature than C. recta. C. paleacea is distinguished by its pendent female spikes and pale brown female glumes with a long, finely serrated awn 6-15 mm _ long (Mossberg & Stenberg 2003). In size C. vacillans is closest to, but typically taller than, C. salina. The two species are also separated by utricle appearance. Specifically, C. vacillans has densely papillose and nerved utricles, with spinules around the beak orifice, dark pistillate glumes with a narrow central band, sometimes incomplete (not reaching the apex) on unawned glumes, usually a scabrous stem, and dull surfaced nuts. It also has some divergent to even drooping spikes and usually a sharp trigonous stem. C. salina lacks this suite of characters. Both share the characters of frequently scabrous awned _ glumes _ and constricted nuts. Small plant size, spathe-like lowest bract and the female glumes, which are mostly found with a short to long scabrous awn, are the best character to identify C. salina and are consistent across specimens of C. salina in northern Europe (including Scotland) and eastern North America. A key to C. salina and closely related, potent-ially confusing species is given below. KEY A key to Carex salina and related coastal species in Carex section Phacocystis in northwest Europe. This section is identified by separate male and female spikes with the upper usually entirely male and the lower usually entirely female, two stigmas, and utricles and nuts biconvex or planoconvex. 1 ie Vegetative leaves involute and 1-3 mm wide; leaves lying on substrate surface; on lower PSHEIES OWE SANETIRITS D5, hoceasnstesceseee eee aces ecco eS S UNDE Se ET Ca nae eee ee ae Tee Carex subspathacea 1. Vegetative leaves keeled or V-shaped or widest more than 3 mm; leaves clear of substrate surface; on lower or upper reaches of saltmarsh or adjacent landward vegetation. ............... 2... 2 Fea NomMara predominantly or exclusively on one leaf Surface Only ..:...2..-.....2-.-.--os0ccre-eceecceceees ons 1 PS Ommaraeanundant On both le al SUIACES w....62-....ccssasoneaeecSauseon dese nebeasgentebecacdeueesevaneoe-ducndaseeceoecees 6 3. Leaves with stomata predominantly or exclusively on upper surface; female glumes obtuse or ACWIE, WLDOULE ARDS essccgSesicshcGeee eee ae eee M Se eae NAc cee NR ere Le el 4 3. Leaves with stomata predominantly or exclusively on lower leaf surface; female glumes with LO MMMME May Meatal CASUONAO WCE PLONE US .2..25.ccnccsees = cece - sete te cat ote- cece oscnoeeness i ss, A y y = i = 1 Watsonia 27: 59-64 (2008) 59 Conservation of Britain’s biodiversity: Hieracium pseudoleyi (Asteraceae), Purple-flushed Hawkweed J. SAWTSCHUK Université de Bretagne Occidentale, Institut de Géoarchitecture, CS 93837, 29238 Brest Cedex 3, France W. McCARTHY 5 Tyn-y-coed Road, Great Orme, Llandudno, Conwy LL30 2QA and iC 1 GyRiCk: Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP ABSTRACT Reviews of historical data were combined with field surveys to assess the status of H. pseudoleyi, Purple- flushed Hawkweed, a rare endemic of North Wales. It has been recorded historically in three sites, and there are doubtful or erroneous records from four sites. In 2006, 798 plants were found on the Great Orme and 53 plants on the Little Orme. It was not refound at Pydew. It occurs on limestone rocks and in crevices in open vegetation. Under the [.U.C.N. (2001) threat criteria it is probably best regarded as “Vulnerable”. KEYWORDS: endemic, I.U.C.N. Threat Criteria, Wales. INTRODUCTION Hieracium pseudoleyi, Purple-flushed Hawkweed, is a rare endemic plant confined to North Wales. Although it is a Red Data Book species (Wigginton 1999), relatively little 1s known about it. As part of a project into conservation of Welsh plants, historic data has been combined with field surveys to assess its needs for conservation and to provide an [.U.C.N. Threat Category as required under the Global Plant Conservation Strategy (Secretariat for the Conservation of Biodiversity 2002). In this paper the data are summarised; full details are given in Sawtschuk (2006), updated with additional field work in 2007. 2 *e-mail: tim.rich@museumwales.ac.uk The history of H. pseudoleyi is somewhat confusing as it has been named_ twice independently. It was collected regularly from the Great Orme from the 1890s onwards, initially under the name H. caesium Fries which was broadly interpreted at the time. Linton (1905) noted that some plants from the Great Orme and West Yorkshire differed from H. caesium in having leaves more attenuated below and densely glandular involucral bracts, and described it as a new variety, var. decolor W. R. Linton, also noting that the Great Orme plant had leaves fringed with strong hairs. Specimens were sent by Linton to the Scandinavian expert Dahlstedt, who com- mentedsy not vw caesium: Enes => ‘So Ley (1909) raised Linton’s variety to species rank as H. decolor Ley, and it was recorded under that name for many years. However, the Yorkshire and Great Orme plants recorded under this name are now regarded as different species, and as H. decolor is typified on Yorkshire material, the name H. decolor is no longer applicable to the Great Orme taxon. The current name H. pseudoleyi is a new combination by Roffey (1925) of the name coined by Zahn (1921) for material distributed as cultivated H. leyi in the Set of British Hieracia no. 128 from Ysgolion Duon. Material distributed with this set is composed of both H. leyi F. Hanb. (wild plants, mostly collected by 60 J. SAWTSCHUK, W. McCARTHY AND T. C. G. RICH FIGURE 1. Hieracium pseudoleyi (del. T. Rich). A, Whole plant. B-E, Outer rosette leaves, upper surfaces. F_I, Stem leaves. J, Involucral bract. Scale bars 1 cm. HIERACIUM PSEUDOLEYI 6] A. Ley or W. R. and E. F. Linton) and H. pseudoleyi (cultivated plants, grown by E. F. Linton), the latter being cited as types by Zahn. It is not clear if H. pseudoleyi has ever occurred at Ysgolion Duon or whether this was a result of confusion of labels on specimens; irrespective of their origin, they are the types (Sell & Murrell 2006). Hieracium pseudoleyi is now included in section Stelligera Zahn by Sell & Murrell (2006). It has previously been included in section Vulgata (Griseb.) Willk. & Lange and section Oreadea (Fries) Dahlst. (Linton 1905; Ley 1909; Pugsley 1948; Sell & West 1968). Summary of synonymy: Hieracium pseudoleyi (Zahn) Roffey in F. Llanbeond Cart (Brit: Pl. ed. 11, 26 (1925). =H. caesium Fries var. decolor W. R. Linton, Brit. Hier. 59 (1905) pro parte. =H. decolor Ley, in J. Bot. 47: 10 (1909), and auct. =H. saxifragum Fries subsp. pseudoleyi Zahn in Engl., Pflanzenreich IV. 280 (Heft 75): Precio) basionym. Type H. leyi cult.) Set of British Hieracia no. 128 pro parte. Ystolion [sic] Duon. =H. smithii (Baker) Druce, List Brit. Pl., ed. 2. 67 (1928), pro parte? Hieracium pseudoleyi is illustrated in Figure 1, and a full description is given by Sell & Murrell (2006). Key identification features are the basal rosette of narrow, purple-flushed, weakly glaucous leaves typically with the margins up-curved, the outer rosette leaves oblong-lanceolate, toothed to sharply toothed and cuneate at the base, the usually solitary stem leaf, the forking inflorescence with 1-6 medium-sized heads, the dense stellate hairs on the involucral bracts and yellow styles (Pugsley 1948). It is quite variable, probably related to the environment; robust plants to 35-40 cm with strongly toothed leaves in sheltered places contrast with smaller plants to 15 cm with few, sparsely toothed leaves on windswept crags. Other glaucous-leaved species which may grow with it are H. cambricum (Baker) F. Hanb. which has deeply laciniate rosette leaves, H. holophyllum W. R. Linton which has much broader, ovate-lanceolate basal leaves, and H. britannicoides P. D. Sell which has sparse stellate hairs on the involucral bracts. Hieracium vagense (F. Hanb.) Ley is very similar and also grows in the area but has not yet been found growing with H. pseudoleyi; it has sparse stellate hairs on the involucral bracts and usually laciniate leaves. METHODS Historical data were compiled from herbaria (mainly BM, CGE and NMW, with some records from BIRM, LIV and MANCH), the literature and correspondence with local botanists. Field surveys were carried out mainly in 2006 using the historical records to target areas to search. As H. pseudoleyi occurs on carbon- iferous limestone cliffs and quarries, it was searched for by walking along the tops and bottoms of the cliffs, and by using binoculars. Reasons for possible loss of populations and threats to extant populations were assessed in the field. Voucher specimens and photographs are deposited in NMW. Associated species were recorded in estimated 2 x 2 m quadrats around the species. Soil pH was measured with a pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water on soil samples collected from around the roots. RESULTS DISTRIBUTION The historic records traced are listed in Appendix 1, which show H. pseudoleyi has been recorded in three sites, and there are doubtful records for four sites. These records were used to direct the field surveys in 2006; all sites were visited with the exception of Ysgolion Duon. ane We \ ra és Ww ys i Zz ia Lease salt // Ar ee JS fr eae) ie FIGURE 2. Distribution of Hieracium pseudoleyi. g, 2006. G, pre-2006. x, doubtful or error. 62 J. SAWTSCHUK, W. McCARTHY AND T. C. G. RICH GREAT ORME (SH78; V.C. 49) There are many records of H. pseudoleyi from the southwest, south and southeast sides of the Great Orme over approximately 4 km of slope cut by numerous cliffs at different levels. Indeed W. McCarthy has reported seeing many plants in these areas during the last 20 years. Surveys were carried out by in June 2006 by W. McCarthy and J. Sawtschuk. All 10 pre- cisely localised historic records were refound, with many new subpopulations. Although the whole population could be regarded as one large site, it can be conveniently broken down into six main subpopulations. The first subpopulation occurred on southwest facing cliffs to the south of Great Ormes Head above and below the marine drive at Gogarth. A total of 114 plants were counted, and there were no obvious threats other than over-grazing by goats. The second subpopulation occurred on cliffs from the east of Gogarth to the west of Pen-y- ffridd farm and near the old copper mine. At least 145 plants occur with H. britannicoides and another species. Again there were no obvious threats other than over-grazing by goats. The third subpopulation occurred on_ the cliffs above Cwlach east of Pen-y-ffridd Farm. The 51 plants counted are potentially threatened by scrub expansion. The fourth subpopulation occurs on the rocks from the West Toll Gate Lodge to Haulfre Gardens above Llandudno. A total of 132 plants occurred on rocks and path sides where there is significant encroachment of vegetation and scrub. Some areas have already been cleared for conservation of other grassland species such as Hypochaeris maculata, which has benefited H. pseudoleyi too. The fifth subpopulation occurred on cliffs above Ty-Gwyn Road and Wyddfyd Road. 171 plants were counted, mostly on inaccessible cliffs where it is not threatened. The final subpopulation of 185 plants occurred on rocks and cliff above Happy Valley. This site is not threatened. Thus at least 798 plants have been recorded across most of the south side of the Great Orme. Given the difficulty of covering such a large area of cliffs and rocks thoroughly, this 1s clearly a conservative estimate. LITTLE ORME (SH8182; V.C. 49) Hieracium pseudoleyi was first found on the Little Orme in May 1953 by V. Gordon, and was refound by W. McCarthy in 2002. Three sub-populations of H. pseudoleyi were found on 8 June 2006 by W. McCarthy and J. Sawtschuk. A total of 35 plants occurred in a small, amphitheatre-shaped cliff with UH. holophyllum, and another five plants occurred in a second amphitheatre 100 m to the north. Six plants occurred on a small, northwest facing cliff, above a bigger cliff above the sea. Thus at least 53 plants occur. There are no obvious threats to this site as the plants are mostly on low cliffs inaccessible to sheep, though there is a longer term potential for expansion of scrub on some of the more sheltered cliffs. PYDEW (SH8179; V.C. 49) J. M. Brummitt collected plants on limestone pavement at Pydew on 27 June 1969 at SH816797. The site is now a nature reserve. On 8 June 2006 W. McCarthy and J. Sawtschuk searched the site, but found only two other Hieracium species. Although the site is difficult to search thoroughly and H. pseudoleyi could still be present, the amount of scrub has increased significantly during the last 30 years, and it may have been shaded out. Some areas have been opened up recently, but it has not yet been refound. DOUBTFUL OR ERRONEOUS RECORDS The cultivated type specimens from Ysgolion Duon (SH66; v.c. 49) may be the result of confused labelling (Sell & Murrell 2006), and some records are errors for H. leyi. Hieracium pseudoleyi has not been refound there amongst the many other Hieracium species collected during many visits. A specimen collected in a slate quarry in April 1968 by W. E. Hughes at Llanllechid (c. SH6268; v.c. 49) has been ascribed to H. pseudoleyi (Sell & Murrell 1968). The area was surveyed on 10 June 2006 by J. Sawtschuk. Two plants similar to H. pseudoleyi and the Hughes material were found at SH62846770 but differed in having too many stem leaves. Re-examination of the Hughes herbarium specimen suggests it is not H. pseudoleyi. Hieracium carneddorum Pugsley occurs in other quarries. Some material from Gloddaeth (SH88; v.c. 49) collected by A. Ley on the 19 June 1901 has been named H. pseudoleyi but is H. vagense. Material collected by A. Ley at World’s End, Eglwyseg (SH78; v.c. 50) on 16 July 1903 has been treated as H. pseudoleyi (Sell & Murrell 1968). The material is very similar to H. pseudoleyi in many respects but differs in HIERACIUM PSEUDOLEYI 63 having discoloured styles and involucral bracts with sparse to medium stellate hairs and few, short glandular hairs. There are about 40 plants at World’s End, and whilst obviously closely related to H. pseudoleyi these would be best treated as a separate taxon. ECOLOGY Hieracium pseudoleyi, like other British Hieracium species, is a polycarpic perennial, and is probably apomictic. Its chromosome number is not known. Flowering often begins in April (or exceptionally in 2007 in March), which is early for Hieracium in Britain, continuing into June and possibly July de- pending on the weather. About 46% of the plants were flowering in 2006. Seeds are wind- dispersed. Its populations are probably some- what dynamic and it spreads readily from seeds. Hieracium pseudoleyi typically occurred on steep to vertical slopes (40—90°) on west, south and east facing slopes, in short (up to 40 cm tall), open vegetation with cover 5-—50% associated with Dactylis glomerata L., Festuca ovina L., Helianthemum canum (L.) Hornem., Lotus. corniculatus L., Sanguisorba Scop., Scabiosa columbaria L., Sonchus oleraceus L., Teucrium scorodonia L., Thymus polytrichus A. Kern ex Borbas, and sometimes other Hieracium species. It occurred in a range of calcicolous vegetation types from upper maritime cliffs communities to droughted calcicolous grasslands, which mainly were in terms of British Plant Communities (Rodwell 1992, 2000) CGI Festuca ovina — Carlina vulgaris grassland, MC4 Brassica oleracea maritime cliff-ledge vegetation and the OV39 Asplenium trichomanes — A. ruta-muraria community, but the presence of a rocky, open limestone habitat seems more important than the community and associated species per se. Plants were rooted directly into crevices in the limestone rocks, or into shallow soil pockets. minor Two soil samples collected from around the roots had pH values of 7-0 and 7-4. DISCUSSION A total of 851 plants of H. pseudoleyi were recorded in 2006 and 2007 in one large population scattered across the south side of the Great Orme and in a small population on the Little Orme. It was not refound at Pydew, which may have been a transient, wind-blown colonist from the large population at the Great Orme. These sites are difficult to search thoroughly, and the population estimates are conservative. Under the IUCN (2001) threat criteria, H. pseudoleyi is probably best categorised as “Vulnerable”. However, the large, widespread population on the Great Orme (virtually all within the Great Orme’s Head SSSI) and the second population on the Little Orme (also an SSSI) without any obvious current threats suggest that H. pseudoleyi is actually not very threatened in the short term. Longer term, the main threats are potentially over-grazing, scrub invasion or spread of alien species such as Cotoneaster. We suggest that monitoring the populations every 10 years would be adequate to keep a watching eye on its populations. Seeds from different sites were sent to the Millennium Seed Bank, and it is planned to cultivate the material at the Great Orme Visitor Centre: ACKNOWLEDGMENTS We would like to thank Wendy Atkinson, Jean Green, David McCosh, Gina Murrell, John Osley, Sally Pidcock and Peter Sell for their help and advice, and the Keepers of the Herbaria for access to libraries and specimens. The historic data’ were compiled with funding from the Countryside Council for Wales. The map was plotted using DMAPW by Alan Morton. REFERENCES 1.U.C.N. (2001). LU.C.N. Red list categories and criteria. Version 3.1.1.U.C.N., Gland. LINTON, W. R. (1905). An account of the British Hieracia. West, Newman & Co., London. PUGSLEY, H. W. (1948). A prodromus of the British Hieracia. Journal of the Linnean Society of London (Botany) 54: 1-356. RODWELL, J. S., ed. (1992). British plant communities. Volume 3. Grasslands and montane communities. Cambridge University Press, Cambridge. RODWELL, J. S., ed. (2000). British plant communities. Volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge. ROFFEY, J. (1925). Hieracium, in HANBURY, F. J. ed., The London catalogue of British plants. Edition 11. G. Bell & Sons, London. SAWTSCHUK, J. (2006). Conservation of endemic Hieracium species in the British Isles and assessment of four Welsh species: Hieracium pachyphylloides, Hieracium pseudoleyi, Hieracium rectulum and Hieracium cambricogothicum. E.S.E.B. Masters Thesis, Université de Rouen, France. 64 J. SAWTSCHUK, W. McCARTHY AND T. C. G. RICH SECRETARIAT FOR THE CONSERVATION OF BIODIVERSITY (2002). Global strategy for plant conservation. Secretariat for the Conservation of Biodiversity, Montreal. SELL, P. D. & MURRELL, G. (2006). Flora of Great Britain and Ireland. Volume 4. Cambridge University Press, Cambridge. SELL, P. D. & WEST, C. (1968). Hieracium L. in, PERRING, F. H., ed., Critical Supplement to the Atlas of the British flora. BSBI, London. WIGGINTON, M. J., ed. (1999). British Red Data Books. 1. Vascular Plants, 3rd edition. Joint Nature Conservation Committee, Peterborough. ZAHN, K. H. (1921). Compositae — Hieracium. in ENGLER, A., ed., Das Pflanzenreich TV. 280. Heft 76: 450. Engelmann, Berlin. (Accepted September 2007) APPENDIX 1 — DETAILS OF HIERACIUM PSEUDOLEYI RECORDS TRACED V.C. 49 CAERNARFON Great Orme. Near old copper mines, July 1891, J. E. Gnffith (BIRM, BM, CGE, LIV, NMW),. 31 May, 18 and 28 August 1892, J. A. Wheldon (NMW). 1896, A. Ley (CGE). July 1900, G. C. Druce (BM). 20 June 1901, A. Ley (BM, LIV, NMW, CGE; Set of British Hieracia no. 179). South side, 21 and 22 June 1901, A. Ley (CGE). July 1901, H. J. Riddelsdell (BM). 4 September 1902, H. W. Pugsley (BM). 9 July 1904, A. Ley (BM, CGE). Small quarry near Gogarth Abbey, 15 May 1905, A. O. Hume (BM). Cliffs above Gogarth Abbey end of Marine Drive, 25 May 1905, A. O. Hume (NMW). SW corner, higher cliffs, 27 May 1905, A. O. Hume (BM). Quarry adjoining Happy Valley, 29 May 1905, A. O. Hume (LIV). 26 July 1905, H. W. Pugsley (BM). Rocks, 19 June 1908 and 2 October, T. J. Foggitt (BM). 26 June 1910 and 21 July 1911, J. W. Ellis (LIV). September 1910, N. D. Simpson (BM). 2 July 1912, W. A. Shoolbred (NMW, CGE). 2 July 1912, S. H. Bickham (CGE). Shaded limestone quarry and shaded gully, 2 July 1912, Linton, Marshall (BM, CGE). Exposed cliffs, west side, 9 July 1912, W. A. Shoolbred & E. S. Marshall (BM, CGE, NMW). Cottage, stony patch, 1918, S. H. Bickham, flowered in cultivation 5 August 1919 Ledbury (BM, NMW, CGE). 31 July 1922, H. W. Pugsley (BM). 31 July 1936, H. S. Redgrove (BM). Above Haulfre Gardens, 10 July 1942, H. W. Pugsley (BM). Quarry on landward side, June and 13 July 1948, K. G. Payne (CGE). Southwest side, 26 June 1952, P. D. Sell (CGE). June 1953, C. West (CGE). Limestone cliff, Marine Drive, 26 June 1953 and wall at Craigside Hydro, 25 June 1955, C. A. E. Andrews (BIRM). SH7683, 25 September 1954, V. Gordon (NMW). 28 July 1955, F. R. Browning (BM). Roadside cliffs, SH760832, 24 June 1961, J. N. Mills (MANCH). 8 June 1964, U. K. Duncan & E. P. Beattie (BM). 11 September 1977, B. R. W. Fowler (BIRM). SH755835, limestone rocks at 130 m, 24 July 1977, R. J. Pankhurst (BM). Rocks above Haulfre, SH774824, 2000, W. McCarthy (herb. McCosh). Cliffs above Ty-Gwyn Road, SH775829, cliffs above West Shore toll house, SH768823, cliffs at top of zigzag path on Invalids Walk, SH772824 W. McCarthy. Little Orme (SH8182). 25 May 1953, V. Gordon (NMW). SH813823, 2002, W. McCarthy. Pydew (SH8179). SH816797, 27 June 1969, J. M. Brummitt, thin soil in pockets of limestone pavements (NMW, CGE). DOUBTFUL OR ERRONEOUS RECORDS Llanllechid (v.c. 49). Slate quarry near, April 1968, W. E. Hughes (NMW). Carnedd Dafydd/Ysgolion Duon (v.c. 49). Carnedd Dafydd, 29 Jul 1892, W. R. Linton (NMW),. Ysgolion Duon, July 1893, J. E. Griffith (NMW); both probably H. Jey. Gloddaeth. 19 June 1901, A. Ley (CGE; redetermined T. Rich 2007). Eglwyseg (v.c. 50). Bluff next north of the church; World’s End, 16 July 1903, A. Ley (CGE; redetermined T. Rich 2007). CULTIVATED Ysgolion Duon, cultivated 23-28 June 1896, 1898 & 1899, E. F. Linton (Syntypes. Set of British Hieracia no. 128. BM, CGE, NMW, and no doubt elsewhere too). Great Orme, 8 June 1908, A. Ley. Great Ormes Head, August 1915, S. H. Bickham cultivated at Ledbury (NMW, CGE). Great Ormes Head, 16 September 1925, origin 1922 H. W. Pugsley (CGE). Watsonia 27: 65—72 (2008) 65 Notes EPILOBIUM TETRAGONUM SUBSP. TOURNEFORTI NATURALISED IN BRITAIN In the context of the British flora, issues regarding the infraspecific treatment of Epilobium tetragonum (Square-stalked willowherb) centre on whether the segregate Epilobium lamyi F. W. Schultz should be treated as a separate species, a subspecies or not afforded any distinct rank. There is, however, a further segregate within E. tetragonum which, until 2007, had apparently not been recorded in the British Isles, namely the large-flowered Epilobium tetragonum subsp. tournefortii (Michalet) Léveillé. This subspecies has a Mediterranean distribution, being known in southern Portugal, Spain and France; also Sicily, Corsica, Sardinia and Malta. In Africa it ranges from Tangier to Tunisia; and in the east, it 1s found in Turkey, ranging to eastern Anatolia, Syria and Lebanon (Raven 1967). : On 6 August 2007, an unusual willowherb was found by A.C.L. (in the company of A. Stevenson) growing in quantity along the banks of the A505 Royston by-pass in Cambridge- shire, v.c. 29, approximately at TL364421. A further visit undertaken by G.D.K. and A.C.L. on 12 August 2007 confirmed the identity of the plant as Epilobium tetragonum subsp. tournefortii. Voucher material has been placed in CGE and herb. G.D.K. The find location (site 1) was a steep south- west facing chalk slope, up to 7 m high, and on the northeast side of the bypass. Erosion maintained a semi-open habitat which provided Opportunities for Epilobium — seedling establishment, although occasional plants were to be found in the coarser grassland at the foot of the slope. The flora of the slope otherwise consisted of native calcicoles, with some neophytes or introductions, none of which suggested any particular association with introduction of the Epilobium — viz. Sedum rupestre L., Centranthus ruber (L.) D.C. and Conyza sumatrensis (Retz.) E. Walker. There were hybrids between the latter and Erigeron acer L., of which an account 1s to be published separately. The colony of Epilobium tetra- gonum subsp. tournefortii extended from TL36594196 to TL36394213 alongside the carriageway, with the main grouping concentrated within 45 m of that length. The population appeared well established, with several thousand plants present. A small scattering of plants was also present on the opposite side of the carriageway. Further investigation by ACL later on 12 August revealed the existence of a further population by the A505, approximately 1 km to the northwest of site 1, and located on a similar southwest facing bank alongside a layby (site 2). At least a thousand plants were present. Because of the size of the rose-purple flowers, 2°0—2:5 cm in diameter, their appear- ance was quite distinctive. The impression given was of numerous Epilobium hirsutum flowers borne by robust E. tetragonum plants; but the taxon was readily distinguishable from E. hirsutum xX tetragonum by a number of factors. The stem hairs were short and appressed, as with FE. fetragonum_ subsp. tetragonum, whereas the hybrid carries some glandular hairs as well as long hairs influenced by E. hirsutum, which are upwardly deflexed (and so intermediate between the spreading hairs of E. hirsutum and the closely appressed hairs of the other parent). The hybrid carries a confused stigma, intermediate between 4-lobed and clavate stigma types, whereas Epilobium tetragonum subsp. tournefortii bears a clavate stigma. The hybrid also occurs as single plants, potentially with variable characteristics between the parents, and with reduced fertility; the Royston populations, however, were fully fertile and, except for one plant mentioned below, uniform. Various authors have remarked on distinguishing features of subsp. tournefortii, summarized in Table 1. The key in Flora Europaea (Raven 1968) describes petals as 7-O—11-5 mm long (in contrast to other E. fetragonum subspecies recognized, which carry petals 2°5—7-:0 mm long), with the stigma elevated above the anthers at anthesis. Flower length for subsp. tournefortii has also been described by Raven (1962) as 8-13 mm. None of the authorities cited in Table 1 attributes a measure greater than 8-5 mm for the spetalalencthy or othe sremainder (of = tetragonum. Petals from the Royston populations measured 11—13 mm; flowers were held horizontally at anthesis, with the stigma clearly projecting beyond the anthers. Nieto Feliner (1996) considered that anther 66 NOTES Watsonia 27 (2008) TABLE 1. TAXONOMIC COMPARISON Flower/petal length Anther length Seed length E. tetragonum subsp. tournefortii Raven (1962) Flowers 8-0—-13 mm Raven (1968) Petals 7-0-11-5 mm Smejkal (1995) (1-9)2-0-2-3(2-4) mm Nieto Feliner (1996, Petals 6-11 mm (1-3)1-5—2-4(2:7) mm_ (1)1-1-1:3 mm 1997) Remaining segregates of E. tetragonum, including E. lamyi Raven (1968) Petals 2:-5—7-0 mm Smejkal (1995, 1997) Petals 4—-8(8:5) mm (0-8)0-9-1-1 mm 1-0-1-2(1-3) mm Nieto Feliner (1996, Petals 3—5:5(6:5) mm 0-7-1-1(1-3) mm 0-8—1(1-1) mm 1997) Oredsson & Petals (4:5)5-0-8:0 mm 0-7—1-0 mm (0-85)1-0—1-1 mm Snogerup (1976) Royston plants Petals 11-13 mm (1:5—)1-7—2:2 mm 1-1-1-3 mm length was significant, with subsp. tournefortii measuring (1-3)1-5—2:4(2:7) mm, as distinct from O-7—1-1(1-3) mm for the remainder of E. tetragonum. Measurements for the Royston populations are (1:5—)1-7—2:2 mm, well within the range of subsp. tournefortii. The distinction in anther length is also one made by Smejkal (1995), although the figures used by him for subsp. fournefortii are (1-9)2:0—2:3(2:4) mm, and for the remainder of E. tetragonum (0:8) 0-9-1-1 mm. Nieto Feliner (1996) also commented on the longer seeds of subsp. tournefortii (1-1-1-3 mm), and those of the Royston populations replicate this range. His account in Flora Iberica (Nieto Feliner 1997) recognises a slightly wider range — (1)1-1-1:3 mm. The seeds of British E. tetragonum do not normally exceed 1-1 mm (observations by G.D.K.), and this is in accordance with data from some regions where subsp. tournefortii is not present, e.g. (0-85)1-0—1-1 mm in Scandinavia including the segregate E. lamyi (Oredsson & Snogerup 1976), although Smejkal (1997) identified a range of 1:0-1:2(1:°3) mm from the Czech Republic. The characteristics of subsp. tournefortii as regards flower and anther size and the relative position of stigma and anthers are indicative of a taxon which is out-crossing, as distinct from the remainder of E. fetragonum (and, indeed, most species of Epilobium), which is commonly self-pollinated. Raven (1967, 1968) remarked on the existence of intermediates between subsp. fournefortii and _— subsp. tetragonum and their complete interfertility. The Royston population at site 1 appeared | uniform. Some plants of E. tetragonum subsp. tetragonum were noted on the margins of the population of subsp. tournefortii but there were no indications of interbreeding or intermediacy. At site 2, however, where subsp. fetragonum was also present, one plant was noted with characteristics bearing a degree of inter- mediacy: flower diameter was 15 mm, petal length 9-5 mm, anther length 1-0O—1-2mm and seed length 0-7—1-Imm; some of the smaller seeds appeared to be flattened and empty and perhaps indicate some reduced fertility. The data for this specimen are not included in the general measurements given above for the Royston populations. Vigorous E. tetragonum subsp. tetragonum often takes a bushy form, with strong flower production resulting in a massed appearance of long upright capsules. It does not often exceed 80 cm in height, although it may occasionally reach 110 cm. The Royston plants appeared less bushy, more robust, and _ regularly exceeded 110 cm, the tallest specimen being measured at 123 cm. There is no evidence (of+ the basismior introduction of this taxon to Royston, although the original seeding of the roadbanks or carriage by vehicles both constitute possibilities. Spread has evidently occurred to a degree, but the appearance of FE. fetragonum_ subsp. tournefortii in the British Isles at present comprises an isolated record without suggestion of invasiveness, although affording evidence that this Mediterranean taxon is capable of establishing in the British Isles and thriving. NOTES Watsonia 27 (2008) 67 REFERENCES NIETO FELINER, G. (1996). Notes on Epilobium (Onagraceae) from the Western Mediterranean. Anales Jardin Botanico de Madrid 54: 255-264. NIETO FELINER, G. (1997). Epilobium L. in Flora Iberica 8: 101-131, Real Jardin Botanica, CSIC, Madrid. OREDSSON, A. & SNOGERUP, S. (1976). Drawings of Scandinavian Plants 113-114 Epilobium L. Sect. Epilobium. Botaniska Notiser 129: 193-197. RAVEN, P. H. (1962). The genus Epilobium in Turkey. Notes from the Royal Botanic Garden Edinburgh 24: 183-198. RAVEN, P. H. (1967). A revision of the African species of Epilobium (Onagraceae). Bothalia 9: 309-333. RAVEN, P. H. (1968). Epilobium L., in TUTIN, T. G. et al., eds. Flora Europaea 2: 308-311. Cambridge University Press, Cambridge. SMEJKAL, M. (1995). Sieben neue Bastarde in der Gattung Epilobium L. (Onagraceae). Acta Musei Moraviae. Scientiae Naturales 79: 81-84. SMEJKAL, M. (1997). Epilobium L., in SLAVIK, B., ed. Kvétena Ceské Republiky 5: 99-132. Academia, Prague. G. D. KITCHENER Crown Villa, Otford Lane, Halstead, Sevenoaks, Kent TN14 7EA A.C. LESLIE 109 York Street, Cambridge CBI 2PZ NEW WILLOW HYBRID Salix x permixta J. Webb hybrid. nov S. caprea L. x S. cinerea L. x S. repens L. Type : British Isles, North Devon (v.c. 4), Braunton Burrows, SS4629234468, 27 August 2007, J. Webb 001 (TTNCM: Taunton County Museum Herbarium). Frutex multiramosus tholiformis, circiter 1—1-5 m altus, e surculis subterraneis exoriens; rami numerosi, ascendentes, teretes, primo dense appresso-pubescentes, in maturitate glabri vel subglabri, saturate brunnei, decorticati obscure et sparsissime striatulati; folia late ovata, acuta vel obtusa, breviter apiculata, 2-54 x 1-3-2°8 cm, brevissime petiolata, lamina supra saturate Viridi sparsissime puberula, subtus pallide grisea dense et appresse sericeo-tomentella, nervis prominentibus ascendentibus conspicue ornata, marginibus subintegris vel remote serrulatis; stipulae plerumque minutae, subintegrae vel obscure glanduloso-serrulatae. Amenta foeminea (masculina non _ visa) praecocia vel cum foliis juvenilibus evoluta, cylindrica, subsessilia vel breviter pedunculata, suberecta, primo probabiliter 2-5 cm longa, in maturitate usque ad 6 x 1:3 cm, _ bracteis basalibus ovatis 0-5 cm longis, dense sericeo- pilosis, amentorum bracteolae ovatae, obtusae, 3 x 1-2 mm, subfuscae, pilis sericeis conspicuis vestitae; ovarium attenuato-ovoideum, usque ad 3-5 x 1-3-2 mm, in maturitate breviter stipitatum, viride, indumento sericeo sub- sparsim vestitum; stylus brevis, vix 0-5 mm longus; stigmata 2, parva, obtusa, indivisa. Semina matura non visa. British Isles, N. Devon, (v.c. 4) Braunton Burrows. All specimens in Taunton County Museum Herbarium (TTNCM) 001: 27/08/07 002: 17/06/04 003: 04/05/07 Shrub, about 1-1-5 metres high, arising from suckers and forming a much branched dome- shaped bush; branches numerous, ascending, terete, at first densely appressed-pubescent, later glabrous or almost glabrous, dark brown, when the bark is removed obscurely and very sparsely striatulate; leaves broadly ovate, acute or obtuse, shortly apiculate, 2-5-4 x 1:3-2°8 cm, very shortly petiolate, with lamina dark green above, very sparsely puberulous, pale grey below, densely appressed-silky- tomentellous, with nerves prominent, ascending, and margins subentire or remotely serrulate; stipules generally very small, subentire or obscurely glandular-serrulate. Female catkins (males not known) precocious or developing with the young leaves, cylindrical, subsessile or shortly pedunculate, suberect, at first probably 2°5 cm long, at maturity up to 6 x 1-3 cm, with basal bracts ovate, 0-5 cm long, densely sericeous-pilose; 68 NOTES Watsonia 27 (2008) 2 ee ee | lcm FIGURE |. Salix x permixta. A: leaves x 1; B x 3: C: underside of leaf x 1; D: leaf base x 2. catkin-scales (bracteoles) ovate-obtuse, 3 x 1-2 mm, darkish, clothed with conspicuous silky hairs; ovary attenuate-ovoid, up to 3-5 x 1-3-2 mm, at maturity shortly stalked, green, rather sparsely clothed with silky hairs; style short, scarcely 0-5 mm long; stigmas 2, small, obtuse, not divided. Mature seeds not seen. NOTES This willow was first seen on 5 June 2004 in a clearing of the wet-slack habitat that borders the dune system of Braunton Burrows. The 970 hectares of Braunton Burrows, and _ the surrounding coastline, have been declared a UNESCO Biosphere Reserve. On the eastward margins of the dune system Salix repens carpets the sandy slacks with tall scrub and small trees forming small woodland glades. S. cinerea (presumably subsp. oleifolia) is not uncommon in the area but pure S. caprea is rather rare, as is S. aurita. Where S. caprea and S. cinerea co-exist, the hybrid between the two species, S$. x reichardtii, is common and a whole range of intermediates 1s found. Whilst in one such area, Hog Wood, to the south of Sandy Lane car park, we came across a patch NOTES Watsonia 27 (2008) 69 of willow we could not immediately identify. The upright stems were relatively polished and glabrous, the leaves were broadly ovate with the underside distinctly softly pubescent and demonstrating the reticulate venation of S. caprea. Although S. aurita had been found nearby, the stipules of the unidentified plant were small, not at all resembling the large conspicuous stipules of the hybrid between S. repens and S. aurita, nor were the leaves in any way rugose which one would expect in an aurita hybrid. We were certain it was not a hybrid between S. caprea and S. repens , S. X laschiana, as at no stage had it the dense indumentum of S. caprea and if it was a direct hybrid the leaf margins would have been undulate serrate rather than relatively entire. The hybrid between S. cinerea and S. repens, S .x subsericea, was ruled out as its leaves are much narrower and more oblong. S. repens was abundant all over the clearing, which covered an area approximately 75 m x 75 m, and there was a very large conspicuous bush or small tree of S. caprea x S. cinerea (Salix x reichardtii) in the immediate background. We have come to the conclusion that we have identified a new triple hybrid (Webb 2005.) A visit early the following year (2005) found that the clearing had been cut during management works to clear scrub, this prevented catkins forming so our information was incomplete. Subsequent discussion with the site manager prevented this happening again and catkins were collected in 2006 and 2007 and we are thus able to confirm the identification of the triple hybrid. Five separate bushes of the new willow have been found, but all in the same area. It is hoped to engage the active interest and co-operation of the landowner in order to conserve the single location. However, a number of cuttings have been taken and new plants grown; these have been planted in the Community Orchard, Old Cleeve, beside a natural pond where we hope they will thrive. This report has been prepared in consultation with R. D. Meikle. REFERENCE WEBB, J. (2005). New Willow Hybrid. BSBI News 99: 16-17 and Colour Section, p.3. J. WEBB Hybeck, Old Cleeve, Minehead, Somerset, TA24 6HW CHROMOSOME NUMBERS IN SORBUS L. (ROSACEAE) IN THE BRITISH ISLES In the British Isles, the genus Sorbus L. is composed of four sexual diploid species and at least 19 apomictic polyploids (Stace 1997). Many of the polyploids are endemic, and have arisen from hybridisation between a sexual diploid and a polyploid. Knowledge of the cytology of the endemics can help interpret their origins. The following chromosome counts for Sorbus taxa from Britain and Ireland have been compiled from the literature, the B.S.B.I. Cytological Database (www.bsbi.org.uk) and our own data. Many of the first counts for the British endemics were given by Warburg (1952) and details for some of these have been traced in A. J. Wilmott’s notes held in BM which show that they were from seedlings grown from fruit collected in 1933; Warburg’s vouchers for many of the parents are in BM. Counts marked QK were made by Q. O. N. Kay in the 1970s (some vouchers in NMW). Counts marked JB were made on material collected by TR in 2002 and 2003 (vouchers in NMW),. In most cases, counts have been made on root-tips from seedlings, which might differ in ploidy level from the mother depending on whether pollen of a different ploidy level had fertilised the ovules or not. A few counts from flow cytometry are also included at the gross ploidy level. Some counts are approximate due to the difficulty of counting lots of small chromosomes and should be taken to indicate approximate ploidy level, not exact counts. Counts of 2n = 34 are usually regarded as diploid, 2n = 51 as triploid and 2n = 68 as tetraploid. 70 NOTES Watsonia 27 (2008) Sorbus anglica Hed. 2n = 51: Bottom of gorge (‘one plate seemed 2n =) 52) s1Cheddan wc Ow Eas Vanbune (BM notes); Ban-y-gor cliffs, Woodcroft v.c. 34, JB. 2n = c. 49-53: Craig y Cilau, v.c. 42, OK. 2n = 68: Pant Trystan (=Llanymynech Hill) v.c. 40, E. F. Warburg (BM notes); Darren Cilau v.c. 42, Avon Gorge v.c. 6/34, and Kinkerswell v.c. 3, HMcA; Craig Breidden v.c. 47 and (2n = c. 68) Piercefield Cliffs VC, 355 UB: Sorbus aria (L.) Crantz subsp. aria 2n = 34: Cultivated material, Kew Gardens (Moffett 1931); Leigh Woods v.c. 6, and Lord’s Wood v.c. 36, E. F. Warburg (BM notes); The Gully, Avon Gorge v.c. 34, and Piercetield! Glints v.c235; Js. 2m) — Cx 3234: 2 plants, Tidenham v.c.34, QK (collected as S. eminens; NMW). A tree with sharply lobed leaves from The Gully, Clifton Downs v.c. 34, suggested to be a S. aria x latifolia backcross by T. Rich, is 2n = 34, and may simply be S. aria; further investigation of this tree is under way. Sorbus aria var. longifolia Pers. 2n = 34: Aberystwyth v.c. (collected as S. rupicola). 46, HMcA Sorbus aria is widely reported as 2n = 34 in Europe (e.g. Liljefors 1953; Jankun & Kovanda 1988). Aldasoro et al. (1998) have recently reported both diploid 2n = 34 and tetraploid 2n = 68 S. aria s.l. from Spain, though how _ their tetraploid plants relate to British S. aria s.s. remains to be clarified. Sorbus arranensis Hed. 2n = 51: two plants from Glen Diomhan v.c. 100, HMcA. Sorbus aucuparia L. subsp. aucuparia 2n = 34: Cultivated material, Kew Gardens (Moffett 1931); Silverdale v.c. 60 (Al-Bermain et al. 1993); Cwm Clydach v.c. 42, JB. Bolstad & Salvesen (1999) reported 2n = 33 (1 plant) and 2n = 34 (3 plants) from Norway. Sorbus bristoliensis Wilmott 2n = 51: Leigh Woods v.c. 6, E. F. Warburg (BM notes); Avon Gorge v.c. 6/34, ex Cambridge Botanic Garden, HMcA; Avon Gorge v.c. 6/34, QK; Clifton Down v.c. 34, JB. Sorbus croceocarpa P. D. Sell 2n = 59 (aneuploid): Church and Hazelwood Roads, Clifton v.c. 34, QK. 2n = 68: Lemche (1999); Avon Gorge, Clifton v.c. 34, JB. Sorbus decipiens (Bechst.) Irmisch 2n = 51: Stoke Bishop v.c. 34, JB. Sorbus devoniensis E. F. Warb. 2n = 68: Loxhore v.c.4, QK; Woody Bay v.c. An Be Plants of the ‘No Parking’ form from Watersmeet v.c. 4, 2n = c. 68, QK (cited as for typical form in Sell 1989). Sorbus eminens E. F. Warb. 2n = 68: Symonds Yat top v.c. 34; E. FE Warburg (BM notes); Police Quarry, Avon Gorge v.c. 6, JB. 2n = c. 68: Cheddar, QK; Seven Sisters, Great Doward v.c. 36, JB. 2n = c. 71: Seven Sisters, Great Doward v.c. 36, JB. Trees with 2n = c. 51: Sea Mills v.c. 34 and Worle Hill v.c. 6, QK, may be hybrids of S. aria and §. eminens. Sorbus hibernica E. F. Warb. 2n = (c. 50—)51: Rathdrum, Wicklow v.c. H21, OK. Sorbus lancastriensis E. F. Warb. 2n = 68: Arnside Knott, Humphrey Head and Low Frith, all v.c. 60, HMcA. Sorbus latifolia (Lam.) Pers. 2n = 34: two seedlings from planted tree in Newnham College, Cambridge, QK. 2n = 68: seed from planted tree in Newnham College (Lemche 1999); Fairyland, Avon Gorge v.c. 34, QK; Fairyland, Avon Gorge v.c. 34, JB. The difference between the two counts from Newnham College where there are two trees (Sell 1989) is interesting. De Poucques (1951) reported 2n = 34 for S. Jatifolia originating from Fontainebleau, France (the original locality), and for S. confusa (widely treated as a synonym of S. latifolia) from the Malzéville. The triploid count for S. latifolia from the Menai Straights in the B.S.B.I. Database refers to another taxon whose identification remains to be resolved. NOTES Watsonia 27 (2008) 71 Sorbus leptophylla E. F. Warb. 2n = 68: Unlocalised, Britain, Warburg (1952); Craig Breiddan v.c. 47, HMcA. Craig-y- Cilau v.c. 42, and Craig Rhiwarth v.c. 42, JB. 2n = 67-68: Craig y Cilau v.c. 42, QK. Sorbus minima (Ley) Hedl. 2n = 51: Darren Cilau v.c. 42, HMcA; Craig y Cilau v.c. 42, JB. [2n = 68; Coed Pen Pwyr (=Craig y Cilau), E. F. Warburg (BM notes), possibly an error in the notes, as this sample was cited as 2n = Sil im Werlotiey IO syA, Sorbus porrigentiformis E. F. Warb. 2n = 51: Big tree on rock, road below Yat Top, Symonds Yat v.c. 34, E. F. Warburg (BM notes); Crawley Cliff, Gower v.c. 41, HMcA; Crawley Cliff, Gower v.c. 41 QK; Burrington Combe v.c. 6, M. Chester eg al. (flow cytometry, unpublished); Craig y Cilau v.c. 42, Seven Sisters v.c. 36, and Blaise Castle, Bristol v.c. 34 (2 trees, one 2n = c. 51), JB. These triploid counts may refer to hybrids with S. aria. 2n = 68: Near Observatory, Cheddar v.c. 6, and Dan y Graig v.c. 42, E. F. Warburg (BM notes); Stokeleigh Camp v.c. 6 and Craig y Cilau v.c. 42, JB; Craig y Cilau v.c. 42, QK; Babbacombe v.c. 3, Avon Gorge v.c. 6/34, and Aberdur Rocks v.c. 43, HMcA. 2n = (67—) c. 68 (-69) Heale Ladders v.c. 6, Wick Rocks v.c. 34, Craig y Cilau v.c. 42 (both forms), Craig y Castell v.c. 42, Craig Rhiwarth v.c. 42, Cwm Clydach v.c. 42, Darren Disgwylfa v.c. 42 and Woody Bay v.c. 4, QK (some counts reported in Proctor & Groenhof 1992). Plants from Nantporth, Menai Straits v.c. 49, often included under S. porrigentiformis but probably more closely related to S. eminens/S. hibernica (Proctor & Groenhof 1992), are also 2n = 68, HMcA; 2n = c. 66-68, QK. 2n = 87: Craig y Cilau v.c. 42, JB; this is the first ‘pentaploid’ count in wild Sorbus, though this has also been seen in cultivated material by HMcA. Sorbus pseudofennica E. F. Warb. 2n = 68: Glen Catacol v.c. 100, from buds, P. Harrold (annotated specimen in E); Glen Diomhan v.c. 100, HMcA (McAllister 1986). Sorbus rupicola (Syme) Hed. ite— wos Ceci EFedw v.c. 50) and Craig Eglwyseg v.c. 50, E. F. Warburg (BM notes); Malham Cove v.c. 64, and Tighnabruaich v.c. 98, HMcA; Craig y Cilau v.c. 42, JB. Liljefors (1953) also reported 2n = 68 for a plant from Gotland. Sorbus subcuneata Wilmott 2n = 51 and 2n = c. 67-70: Watersmeet v.c. 4, QK. These counts were made in different seed from the same parent; the seedlings were morphologically similar in cultivation. Sorbus torminalis (L.) Crantz 2n = 34: Cultivated material, Kew Gardens (Moffett 1931). Aldasoro et al. (1998) have recently reported both diploid 2n = 34 and tetraploid 2n = 68 material from Spain. Sorbus xX vagensis Wilmott (=S. aria x torminalis) Die o4s aColdwellssRocks -vec. 34. Ey Warburg (BM notes). Aas et al. (1994) report one triploid and 12 diploid counts from three seed families from the primary S. aria x torminalis hybrid from Birnberg, Germany. Sorbus vexans E. F. Warb. 2n = 68: Woody Bay v.c. 4, JB. Sorbus wilmottiana E. F. Warb. n= l-r Avon) Gorse! vc, 0/34) HIMcA: material from type tree, Clifton Downs v.c. 34, JB. Sorbus whiteana T. Rich & L. Houston 2n = 51: Police Quarry, Avon Gorge v.c. 6, and Ban-y-Gor cliffs v.c. 34, JB. In conclusion, three sexual diploids, seven triploids, eight tetraploids and one hybrid are reported as having one ploidy level. Sorbus anglica and S. subcuneata are reported as both triploid and tetraploid, and S. latifolia as diploid and tetraploid. Sorbus porrigentiformis has triploid, tetraploid and pentaploid counts. There are no counts for S. domestica L., S. leyana Wilm. or S. x thuringiaca (sles) Fritsch. ACKNOWLEDGMENTS TCGR would like to thank Wolfgang Bopp and Jessica Gould, National Botanic Garden of Wales for their expertise in growing Sorbus, and Mark Spencer for finding Wilmott’s notes at BM. 72 NOTES Watsonia 27 (2008) REFERENCES AAS, G., MAIER, J., BALTISBERGER, M. & METZGER, S. (1994). Morphology, isozyme variation, cytology and reproduction of hybrids between Sorbus aria (L.) Crantz and S. torminalis (L.) Crantz. Botanica Helvetica 104: 195-214. AL-BERMANI, A.-K. K. A., AL-SHAMMARY, K. I. A., BAILEY, J. P. & GORNALL, R. J. (1993). Contributions to a cytological catalogue of the British and Irish flora, 3. Watsonia 19: 269-271. ALDASORO, J. J., AEDO, C., NAVARRO, C. & GARMENDIA, F. M. (1998). The genus Sorbus (Maloideae, Rosaceae) in Europe and in North Africa: morphological analysis and systematics. Systematic Botany 23: 189-212. BOLSTAD, A. M. & SALVESEN, P. H. (1999). Biosystematic studies of Sorbus meinichii (Rosaceae) at Moster, S. Norway. Nordic Journal of Botany 19: 547-559. DE POUCQUES, M.-L. (1951). Etude chromosomique des Sorbus latifolia Pers. et Sorbus confusa Gremli. Bulletin de la Société Botanique de France 98: 89-92. JANKUN, A. & KOVANDA, M. (1988). Apomixis at the diploid level in Sorbus eximia (Embryological studies in Sorbus no. 3). Preslia 60: 193-213. LEMCHE, E. B. (1999). The origins and interactions of British Sorbus species. PhD thesis, Darwin College, Cambridge. LILJEFORS, A. (1953). Studies on propagation, embryology and pollination in Sorbus. Acta Horti Bergiani 16: 277-329. LILJEFORS, A. (1955). Cytological studies in Sorbus. Acta Horti Bergiani 17: 47-113. MCALLISTER, H. A. (1986). The rowan and its relatives (Sorbus spp.). Ness Series 1. Ness Botanic Gardens, Liverpool. MoFFETT, A. A. (1931). The chromosome constitution of the Pomoideae. Proceedings Royal Society of London, Series B, 108: 423-445. Proctor, M. C. F., PROCTOR, M. E. & GROENHOF, A. C. (1989). Evidence from peroxidase polymorphism on the taxonomy and reproduction of some Sorbus populations in south-west England. New Phytologist 112: 569-575. PRoOcTOR, M. C. F. & GROENHOF, A. C. (1992). Peroxidase isoenzyme and morphological variation in Sorbus L. in South Wales and adjacent areas, with particular reference to S. porrigentiformis E. F. Warb. Watsonia 19: 21-37. STACE, C. A. (1997). New flora of the British Isles. 2nd edition. Cambridge University Press, Cambridge. WARBURG, E. F. (1952). Sorbus L., pp. 539-556. in CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. (1952). Flora of the British Isles. Cambridge University Press, Cambridge. J. P. BAILEY Department of Biology, University of Leicester, University Road, Leicester LE] 7RH OF ONNMICAN: School of Biological Sciences, University of Wales, Swansea SA2 8PP H. MCALLISTER School of Biological Science, University of Liverpool, Liverpool L69 7ZB dis Cy Gale Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP SERIES REASSIGNMENT OF RUBUS CERDICII In describing a conspicuous reddish-flowered | Rhammnifolii (Bab.) Focke. It has a sufficiently bramble mainly confined to the north half of the New Forest, v.c. 11, as R. cerdicii (Allen 2004), I assigned it to series Vestiti (Focke) Focke. Further study, however, has shown that to be unsustainable. Instead, despite the presence of scattered glands and acicles, it seems best in series marked phenotypic resemblance to R. nemoralis P. J. Mueller, a member of the Rhamnifolii very common in the New Forest, to suggest that that may have been one of its ancestral progenitors, the other presumably having been a member of one of the less strongly glandular series. REFERENCE ALLEN, D. E. (2004). Five new species of Rubus L. (Rosaceae) mostly from central south central south England. Watsonia 25: 157-174. D. E. ALLEN Lesney Cottage, Middle Road, Winchester Hampshire SO22 5EJ Watsonia 27: 73-89 (2008) 73 Plant Records Records for publication must be submitted to the appropriate Vice-county Recorder (see BSBI Year Book 2005), and not to the Editors. Following publication of the New Atlas of the British & Trish Flora and the Vice-county Census Catalogue, new criteria have been drawn up for the inclusion of records in Plant Records. (See BSBI News no. 95, January 2004 pp10 & 11). These are outlined below: e First records of all taxa (species, subspecies and hybrids) included in the VCCC, designated as native, archaeophyte, neophyte or casual. e First record since 1970 of the taxa above, except in the case of Rubus, Hieracium and Taraxacum. e Records demonstrating the rediscovery of all taxa published as extinct in the VCCC or subsequently. e Newly reported definite extinctions. e Deletions from the VCCC (e.g. through the discovery of errors, the redetermination of specimens etc.) NB — only those errors affecting VCCC entry. e New 10km square records for Rare and Scarce plants, defined as those species in the New Atlas mapped in the British Isles in 100 10km squares or fewer. (See BSBI News no. 95, January 2004 pp 36-43). Records for the subdivisions of vice-counties will not be treated separately; they must therefore be records for the vice-county as a whole. However, records will be accepted for the major islands in v.cc. 102-104, 110 and 113. In the following list, records are arranged in the order given in the List of Vascular Plants of the British Isles and its supplements by D. H. Kent (1992), from which the species’ numbers, taxonomy and nomenclature are taken. The Ordnance Survey national grid reference follows the habitat and locality. With the exception of collectors’ initials, herbarium abbreviations are those used in British and Irish Herbaria by D. H. Kent & D. E. Allen (1984). Records are field records if no other source is stated. For all records ‘det.’ or ‘conf.’ appear after the herbarium if the determination was based on material already in an institutional herbarium, otherwise before the herbarium. The following signs are used: * before the vice-county number: to indicate a new vice-county record. + before the species number: to indicate that the plant is an archaeophyte. + before the species number: to indicate that the plant is a neophyte. © before the species number: to indicate that the plant is a casual. The above 3 signs may also used before the vice-county number to indicate the status of the plant in that vice-county. ®) before the vice-county number: to indicate that this is an additional hectad for a Rare or Scarce plant. @ at end of entry: established taxon not in Vice-County Census Catalogue. Name of authority follows plant name. [ ] enclosing a previously published record: to indicate that that record should be deleted or changed. The following list contains the second set of records up to and including the year 2006. Records including the year 2007 will be published in the next issue of Watsonia. [Isoetes xhickeyi (I. lacustris x I. echinospora) 3/1.1 x2. 70, Cumberland: delete as unconfirmed, fide A. C. Jermy. | Equisetum xrothmaleri (E. arvense x E. palustre) 4/1.5x8. *49, Caerns.: edge of cycle track, Pant Glas, SH4647, S. J. Thomas, 2006, conf. C. N. Page. Equisetum xbowmanii (E. sylvaticum x E. telmateia) 4/1.7x9 *106, East Ross: marshy meadow near dense swamp woodland, south side of Loch Ussie, NH55, D. J. Tennant, 2006, det. C.N. Page, E. 74 PLANT RECORDS ¢Pteris multifida Poir. 8/1.mul. *29, Cambs.: one good-sized plant under grating on the Trinity Street side of Michaelhouse church, Cambridge, TL448585, P. H. Oswald, 2006, CGE, conf. F. J. Rumsey. @ See BSBI News 104 for details of this species found in Bath. Pilularia globulifera 9/1.1. ®69, Westmorland: bed of ephemeral pool, near Pewet Tarn, Subberthwaite, SD252870, C. Cornish, 2006. 3 sites in same monad. Ceterach officinarum 15/3.1.. £91, Kincardines.: south-facing dike, Stonehaven, NO862851, D. Welch, 2006. Ist record since 1970 at the location previously recorded. *Matteuccia struthiopteris 16/1.1. *62, N. E. Yorks.: several plants of varying age in damp woodland, originally Victorian planting, by Hodge Beck, Cockayne, Bransdale, SE622985, V. Jones, 2003. Dryopteris xcomplexa nothosubsp. complexa (D. filix-mas x D. affinis subsp. affinis) M/S 2X3e *46, Cards.: shaded roadside bank, 200m NW of Caer-Cadwgan, Cellan, SN614480, K. Trewren, S. J. Thomas & A. O. Chater, 2005. Ist Welsh record of nothosubspecies. Dryopteris xdeweveri (D. carthusiana x D. dilatata) 17/3.8x9. 49, Caerns.: wet heathy field, Mynydd Cennin, Bryncir, SH4444, W. McCarthy, 2006, conf. S. J. Thomas. Ist record since ISTO: tAbies alba 20/1.1. *H12, Co. Wexford: many trees self-sown along ride of conifer plantation, Carrig Hill, S9434, P. R. Green, 2006. *Cedrus deodara 20/6.1. *29, Cambs.: one self-sown tree (c.20ft tall in 2006), wedged between two graves, Mill Road cemetery, Cambridge, TL461582, A. C. Leslie, 2002, CGE. Specimen collected 2006. Ist v.c. record for a clearly self-sown Deodar Cedar. +Pinus radiata 20/7.5. 44, Carms.: mixed mature groups of conifers and single trees mixed with native broad-leaves on open south-facing hillside generally dominated by bracken, West Mead Farm, Pendine, SN263094, R. D. & K. A. Pryce, 2006. Ist record as a neophyte. tCupressus macrocarpa 21/1.1. *29, Cambs.: one self-sown young plant at base of St Andrews churchyard wall, Coldham’s Lane, Cherry Hinton, TL489570, A. C. Leslie, 2002, det. K. W. Page. Ist record for clearly self-sown Monterey Cypress. +Chamaecyparis lawsoniana 21/2.1. © *H12, Co. Wexford: several self-sown trees of various sizes on waste ground, New Ross, S725289, P. R. Green, 2006. t+ Juniperus chinensis L. 21/4.chi. *41, Glam.: one well-established shrub in sandy grassland, on Pennard Burrows, Gower, SS553886, A. S. Lewis, 2006. *44, Carms.: field hedge, Broad Oak, SN579226, V. & A. S. Lewis, 2006, NM'W. Ist Welsh record. One mature tree 3m-4m tall, c.40cm in girth, cultivar with white leaf sprays at end of each branch, no cones visible, cut by hedge cutting. Known here for at least 40 years by landowner but girth would indicate that it is considerably older. @ tHelleborus orientalis 28/3.3. *62, N. E. Yorks.: long-established at edge of copse, by ford in village, Battesby, NZ596076, V. Jones, 1993. May have originated from thrown-out garden rubbish. tHelleborus argutifolius 28/3.arg. *62, N. E. Yorks.: self-sown on wall, Sandsend, NZ859128, V. Jones, 2001. tAnemone blanda 28/9.bla. | *42, Brecknock: gravelly area on river bank, Sennybridge, SN 920287, R. G. Woods, 2006. tAnemone xhybrida (A. hupehensis x A. vitifolia) 28/9.hupxvit. *H12, Co. Wexford: patch on bank of River Barrow, New Ross, $71552722, P. R. Green, 2006. +Clematis tangutica 28/12.3. *91, Kincardines.: path in woodland, Banchory, NO704965, D. Welch, 2006. ¢tRanunculus ficaria subsp. ficariiformis 28/13.17c. *67, S. Northumb.: north bank of R. Tyne, NZ118646, C. Metherell, 2006, hortal origin. Thalictrum minus 28/17.4. *46, Cards.: one small plant in rock crevice in cliff in ancient woodland, bottom of Allt Boeth, Cwm Rheidol, SN737773, C. M. Forster Brown, 2006. Certainly native and Ist record as such. *Berberis thunbergii 29/1.2. *69, Westmorland: old quarry, Levens, SD499870, R. E. Groom, 2006, LANC. This plant is var. atropurpurea. *91, Kincardines.: streamside, Banchory, NO715965, I. P. Green & M. Duncan, 2006. *Berberis julianae 29/1.7.. *©57, Derbys.: one bush in waste ground by disused canal, near Springfield Park, SK475349, R. D. Martin, 2006, conf. A. Willmot. Appears bird-sown. PLANT RECORDS TS tBerberis darwinii 29/1.9. *91, Kincardines.: woodland near housing, Banchory, NO704965, D. Welch, 2006. +Berberis verruculosa Hemsl. & Wils. 29/l.ver. 77, Lanarks.: rough grassland, south-west of Milnwood, NS7459, P. Macpherson, 2006, det. E. J. Clement, herb. P. M. Probably originally planted in the area but now established. +Epimedium alpinum L. 29/EPLalp. *46, Cards.: estate woodland, Castle Hill, Llanilar, SN6274, J. H. Salter, 1935-1942. Annotated by J. H. Salter as ‘naturalised’ in his copy of Flowering Plants and Ferns of Cardiganshire in NUW. © Apparently new as naturalised in the British Isles. ©Eschscholzia californica 30/6.1. 44, Carms.: many plants in lay-by on SE side of road, Glynhir Road, Llandybie, SN635148, BSBI Meeting, 2006. Ist record as neophyte (see BSBI News 105 for photo). *62, N. E. Yorks.: one self-sown plant in field edge, near industrial estate, Whitby, NZ907095, V. Jones & W. A. Thompson, 2005. *H12, Co. Wexford: single plant on newly sown road bank, Blackwater, T1234, P. R. Green, 2006. +Pseudofumaria lutea 31/3.1. *H12, Co. Wexford: several clumps on garden wall in Mill Park Road, Enniscorthy, S9739, P. R. Green, 2006. +Pseudofumaria alba 31/3.2. *62, N. E. Yorks.: a few plants in cracks in wall, Thirsk, SE427824, V. Jones & W. A. Thompson, 2006, herb. V. Jones. Ceratocapnos calviculata 31/4.1. *29, Cambs.: weed amongst bedding plants, Notcutts garden centre, Horningsea, TL493622, A. C. Leslie, 2006. Fumaria capreolata subsp. babingtonii 31/5.1b. | £48, Merioneth: one or two fine plants at car park of the Royal St David’s Golf Club, Harlech, SH53, P. M. Benoit & J. M. Maynard, 2000 & 2006. Ist record since 1970. +Soleirolia soleirolii 36/3/1. *110, Outer Hebrides: walls, Kyles Lodge, South Harris, NG997878, J. S. Faulkner, 2006, E. +Pterocarya fraxinifolia 37/2.1. | *62, N. E. Yorks.: showing prolific natural regeneration in woodland, Mill Bank Wood, Kildale, NZ603098, V. Jones, 2004. *Quercus rubra 39/4.7. 46, Cards.: by Forestry Commission track, Coed Maenarthur, Pontrhyd-y-groes, SN722721, S. P. Chambers, 2006. Ist record of a self-sown sapling. Betula nana 40/1.3 *78, Peebless.: Lee Pen, Innerleithen, NT3238, J. N. Brown, 1855, E, conf. D. J. McCosh from herbarium specimen. Ist localised record. Noted in Hudson’s Flora Anglica (1778). Chenopodium rubrum 43/1.6. *£91, Kincardines.: road verge, Crathes, NO726964, I. P. Green, 2006. Stellaria neglecta 46/5.4. 67, S. Northumb.: fen carr to north of lake, Minsteracres, NZ025561, A. J. Richards, 2006, conf. P. M. Benoit, herb. G. A. Swan. Ist record since 1970. Cerastium arvense 46/7.2.. _*52, Anglesey: dune grassland, Valley, SH3074, I. R. Bonner & R. Birch, 2006, herb. I. R. B. Presumed native — link to eastern Ireland (or from an air base in E Anglia?). +Petrorhagia prolifera 46/24.2.. *62, N. E. Yorks.: stony ground in hollow of long-deposited industrial slag, South Gare N of Redcar, NZ557272, C. I. Pogson, 2000, det. E. J. Clement, herb. E. J. Clement. ©Persicaria capitata 47/l.cap. *62, N. E. Yorks.: prolific over several metres at base of wall, Pickering, SE801837, V. Jones & W. A. Thompson, 2005. ©Fagopyrum tataricum (L.) Gaertn. 47/3.tat. *62, N. E. Yorks.: several plants appearing to be a grain contaminant in arable field, near Ingleby Greenhow, NZ572069, V. Jones, 2004, conf. E. J. Clement, herb. V. Jones. © tFallopia xbohemica (F. japonica x F. sachalinensis) 47/5.1x2. *70), Cumberland: riverbank, N side of A66 east of Scales, NY358268, D. Broughton, 2005, det. J. Bailey, LANC, LTR. Herbarium and determination based on R. E. Groom’s 2006 material. *91, Kincardines.: streamside, Banchory, NO715965, I. P. Green & M. Duncan, 2006. *H12, Co. Wexford: large patch on road bank, Crooked Bridge, $826390, P. R. Green, 2006. *Rheum palmatum 47/7.pal. *69, Westmorland: copse with garden throwouts, Ninezergh Lane, Levens Hall, SD494849, G. Halliday, 2006, LANC. Rumex xschulzei (R. crispus x R. conglomeratus) 47/8.13x14. *44° Carms.: cattle- trampled rough pasture, East of Pont Cellyn, near Twyn Farm, SN535217, G. Kitchener, 2006. 76 PLANT RECORDS *H12, Co. Wexford: in field with both parents, Wexford, T024201, P. R. Green, 2006. Rumex xsagorskii (R. crispus x R. sanguineus) 47/8.13x15. | *44, Carms.: a vigorous plant, with both parents, Dryslwyn castle mound, SN552203, G. Kitchener, 2006. *H12, Co. Wexford: single plant in field with both parents, Coolmela, $946608, P. R. Green, 2006. Rumex xabortivus (R. conglomeratus x R. obtusifolius) 47/8.14x19. *44, Carms.: edge of saltmarsh (with both parents), Laugharne, SN303108, G. Kitchener, 2006. Rumex xdufftii (R. sanguineus x R. obtusifolius) 47/8.15x19. *44, Carms.: one large multi- stemmed plant with R. sanguineus nearby, S side of A40 near Pont Dulas, SN551219, G. Kitchener, 2006. *H12, Co. Wexford: single large plant on road bank with both parents, Mackmine Bridge, S964312, P. R. Green, 2006. *Paeonia officinalis 49/1 .off. *51, Flints.: well established from planted material on old industrial site, Rnydymwyn Valley Nature Reserve, SJ205665, J. Phillips, 2006. Hypericum xdesetangsii (H. perforatum x H. maculatum) 51/1.6x7. *91, Kincardines.: riverside, Crathes, NO722964, I. P. Green & M. Duncan, 2006. Hypericum undulatum x H. tetrapterum 51/1.8x9 = *2, E. Cornwall: on edge of arable field in rushy area, east of Leigh Farm, Pillaton, SX387628, I. J. Bennallick, 2006, conf. N. K. Robson, BM. Growing with both parents. Also seen by I. J. Bennallick on the edge of an aqueduct at the base of a bank in the Luxulyan Valley, SX058570. © 1st confirmed record for British Isles of this hybrid. Hypericum humifusum x H. linariifolium 51/1.10x11. | *49, Caerns.: grassy slope above the sea, Mynydd Cilan, SH22, I. Edgar, 2006. =Tilia tomentosa Moench 52/1.tom. *7(), Cumberland: roadside pasture, Watermillock, NY446225, G. Lines, 2006, LANC. @ Lavatera arborea 53/2.1. ©67,S.Northumb.: South Beach, Blyth, NZ3179, G. & A. Young, 2005. Ist record since 1970. ©Abutilon theophrasti 53/6.1.. *44, Carms.: in flower and fruit in disturbed ground on road construction site with Echinochloa crus-galli and Lactuca serriola, near Morfa Roundabout, SS517986, V. & A. Lewis, 2005. ©Anoda cristata (Reichb.) Hochr. 53/ANO.cri. *29, Cambs.: Thriplow tip, TL445445, G. M. S. Easy, 1976, herb. G. M. S. Easy, det. A. L. Grenfell. The record was of the var. brachyantha. O Viola xintersita (V. riviniana x V. canina) 57/1.4x6 *H12, Co. Wexford: several clumps on sand dunes with both parents, Ballyteige Burrow, S948050, P. R. Green, 2006. £Viola cornuta 57/1.10. *52, Anglesey: corner of pony-grazed field, Bryn Castell, Llanddona, SH563787, Anglesey Flora Group, 2006, conf. N. Brown. *69, Westmorland: pathside, N of Sunbiggin Tarn, NY672079, G. Lines, 2006, LANC. ©Viola xwittrockiana (V. lutea x V. tricolor x V. altaica) 57/1.11x12xalt. ABI €o! Wexford: self-sown at base of wall, Duncannon, $7208, P. R. Green & A. Stevenson, 2006. Viola xcontempta (V. tricolor x V. arvensis) 57/1.12x13. | *70, Cumberland: set-aside by R. Eden, Lazonby, NY558397, G. Lines, 2004, photo in LANC. ©Azara microphylla Hook. f., FLA/AZA.mic. | *29, Cambs.: one self-sown plant at base of street wall of Department of Earth Sciences, Downing Street, Cambridge, TL4558, A. C. Leslie, 2006. @ Apparently 1* British Isles record outside gardens for this species. Populus nigra subsp. betulifolia 61/1.3a. | *35, Mons.: 13 maiden, many-stemmed male trees in hedge line, Great Manson Farm, S0499156, D. E. Green, 2006. Ist record updating the VCCC although known for some time from this and 12 other tetrads in the vice-county. +Salix acutifolia 61/2.7. 52, Anglesey: lakeside, Llyn Alaw, SH388871, I. R. Bonner & M. Stead, 2006, det. A. O. Chater, herb. I. R. B. 1st British record as established neophyte. Originally planted with other willows around the reservoir margin. Salix xstipularis (S. viminalis x S. caprea x S. aurita) 61/2.9x10x12. *©57, Derbys.: marshy grassland, Tideswell Dale, SK154736, A. Willmot, 2006. Possibly planted for landscaping. oh wales 3B 5 x Sorbus aria + + » ‘° IRA oe °¢ uF ° + ¢ +% ° 2 ae @ © Sorbus cf. aria F + 4 + : : + As nes . si + Sorbus Menai Strait | a co cp Ay “ = — ———————— oP + + + rf 0 © te A + ‘ -2 AR rig + 4 v x | -3 1 @ =f Axis 2 FIGURE 4. Principal Components Analysis plot of leaf characters in Sorbus. The group labels relate to the Same groups shown in Figure 2, although not all samples are represented as some voucher specimens were inadequate for morphological analysis; other samples used are listed in Appendix 1. 106 R. S. COWAN, R. J. SMITH, M. F. FAY AND T. C. G. RICH the possibility of hybrids between S. aria and S. hibernica has been proposed here, there is no clear evidence to demonstrate that this occurs, and such plants might be expected to be morphologically significantly different from S. hibernica. Significant sub-sexual reproduction is also unlikely in a triploid. The origin of S. hibernica is currently unknown, but it is likely to result from S. aria crossing with a tetraploid such as S. eminens, S. porrigentiformis or S. rupicola (Syme) Hedl. in Britain where all the taxa occur, with subsequent dispersal by birds migrating to Ireland (Wernham ef al. 2002). Sorbus hibernica is also not as uniform morphologically as some of the other apomictic taxa we have studied (e.g. S. bristoliensis; Houston et al. 2008); Parnell & Needham (1998) also found that it was variable. Although the Menai Strait plants form a distinct genetic group, they are clearly closely related to S. hibernica. Their minor genetic separation could result from long-term isolation across the Irish Sea, or could reflect a similar origin. They differ in chromosome number as the Menai Strait plants are reported as tetraploid and S. hibernica as_ triploid (Bailey et al. 2008), suggesting they have different origins. Morphologically, the two taxa cannot be distinguished from the leaves (Figs 1 and 4), and the limited data on fruit size, shape and colouration available show strong overlap between S$. hibernica and the Menai Strait Sorbus (Proctor & Groenhof 1992; pers. obs. T. Rich). Further morphological characters will need to be sought to separate the Menai Strait Sorbus from S. hibernica. ACKNOWLEDGMENTS We would like to thank Trevor Dines, John Early, John Faulkner, John Harron, Libby Houston, Andy Karran, Wendy McCarthy, Andy McVeigh and Sylvia Reynolds for collecting DNA _ samples. Rosalind Codd helped with some of the morphological data. The work was funded by the National Museum of Wales and the Royal Botanic Gardens, Kew. The Royal Botanic Gardens, Kew wish to acknowledge the financial support of Natural England. REFERENCES AAS, G., MAIER, J., BALTISBERGER, M. & METZGER, S. (1994). Morphology, isozyme variation, cytology and reproduction of hybrids between Sorbus aria (L.) Crantz and S$. torminalis (L.) Crantz. Botanica Helvetica 104: 195-214. BAILEY, J. P., KAY, Q. O. N., MCALLISTER, H. & RICH, T. C. G. (2008). Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles. Watsonia 27: 69-72. BENGTSSON, B. O. (2003). Genetic variation in organisms with sexual and asexual reproduction. Journal of Evolutionary Biology 16: 189-199. CASGRAIN, P. & LEGENDRE, P. (1999). The R Package for Multivariate Analysis. Version 4.0. http:// alize.ere.umontreal.ca/~casgrain/R/ CHAPMAN, H., ROBSON, B. & PEARSON, M. L. (2004). Population genetic structure of a colonising, triploid weed, Hieracium lepidulum. Heredity 92: 182-188. CHASE, M. W. & HILLS, H. G. (1991). Silica gel: an ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215—220. CHESTER, M., COWAN, R. S., FAY, M. F. & RICH, T. C. G. (2007). Parentage of endemic Sorbus L. (Rosaceae) species in the British Isles — evidence from plastid DNA. Botanical Journal of the Linnean Society 154: 291-304. DOYLE, J. J. & DOYLE, J. L. (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, Botanical Society of America 19: 11-15. Fay, M. F., GERNANDT, D. S., COWAN, R. S., KITCHEN, M. A. R., KITCHEN, C. & RICH, T. C. G. (2002). Parentage of an unknown member of the Sorbus latifolia (Lam.) Pers. group. Watsonia 24: 91—100. HAMMER, @., HARPER, D. A. T. & RYAN, P. D. (2001). PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontologia Electronica 4(1): 9pp. _ http://palaeo- electronica.org/2001_1/past/issuel_O1.htm HOUSTON, L., ROBERTSON, A. & RICH, T. C. G. (2008). The distribution, population size and growth of the rare English endemic Sorbus bristoliensis A. J. Wilmott, Bristol Whitebeam (Rosaceae). Watsonia 27: 37-49. PARNELL, J. A. N. & NEEDHAM, M. (1998). Morphometric variation in Irish Sorbus L. (Rosaceae). Watsonia 22: 153-161. VARIATION IN SORBUS HIBERNICA 107 PRAEGER, R. L. (1934). The Sorbus aria group in Ireland. /rish Naturalists’ Journal 5: 50-52. Proctor, M. C. F. & GROENHOF, A. C. (1992). Peroxidase isoenzyme and morphological variation in Sorbus L. in South Wales and adjacent areas, with particular reference to S. porrigentiformis E. F. Warb. Watsonia 19: 21-37. PUGSLEY, H. W. (1934). Sorbus porrigens Hedlund, in Ireland. Journal of Botany 72: 58. RICH, T. C. G., LOCKTON, A. J. & PARNELL, J. (2005). Distribution of the Irish Whitebeam Sorbus hibernica E. F. Warb. (Rosaceae). Watsonia 25: 369-380. RICH, T. C. G., MCDONNELL, E. J. & LLEDO, M. D. (2008). Conservation of Britain’s biodiversity: the case of Hieracium cyathis and its relationship to other apomictic taxa. Botanical Journal of the Linnean Society 156: 669-680. SWOFFORD, D. L. (1998). PAUP*: Phylogenetic Analysis Using Parsimony (* and other methods) Version 4. Computer program distributed by Sinauer Associates, Sunderland, MA. WARBURG, E. F. (1952). Sorbus L., in CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E .F. Flora of the British Isles, pp. 539-556. Cambridge University Press, Cambridge. WARBURG, E. F. (1957). Some new names in the British flora. Watsonia 4: 44-45. WERNHAM, C., TOMS, M., MARCHANT, J., CLARK, J., SIRIWARDENA, G. & BAILLIE, S., eds. (2002). The Migration Atlas: movement of the birds of Britain and Ireland. T. & A. D. Poyser, London. (Accepted May 2008) 108 R. S. COWAN, R. J. SMITH, M. F. FAY AND T. C. G. RICH APPENDIX 1. SORBUS SAMPLES USED IN MORPHOLOGICAL ANALYSIS SORBUS ARIA Loch Bunny, H9 Clare, 19 July 1978, D. A. Webb (TCD). Ardrahey, H15 SE Galway, 21 August 1952, D. A. Webb (TCD). Ballinadareen, HI5 SE Galway, 19 July 1978, D. A. Webb (TCD). Gort, H15 SE Galway, 20 August 1952, T. G. Tutin (TCD). Oranmore, H15 SE Galway, 19 July 1978, D. A. Webb (TCD). Oughterrard, HI5 SE Galway, 1978, D. A. Webb (TCD). Ballynahinch, H16 W Galway, 21 September 2003, L. Houston (NMW). Clontarf golf course, H21 Dublin, 29 September 1978, P. H. Pitkin & D. A. Webb (TCD). Portlick Castle, H23 Westmeath, 16 August 1982, D. L. Kelly (TCD). Tallyho, H23 Westmeath, 23 September 2003, L. Houston (NMW). Whitehill, Castlepollard H23 Westmeath, 8 June 1983, D. A. Webb & F. H. Perring (TCD). Slish Wood. H28 Co Sligo, 1962, D. E. de Vasian (BEL). Lower Kilycolpy, H36 Tyrone, 29 July 1927 and 10 September 1937, F. H. W. Kerr (BEL). Carngaver, 8 August 1986, J. Harron (BEL). SORBUS HIBERNICA (ADDITIONAL TO SPECIMENS LISTED IN TABLE 1) Boolabrien Upper, H6 Co Waterford, 18 September 2003, L. Houston (NMW). Coolbunnia, H6 Co Waterford, 10 June 2003, P. R. Green (NMW). Ballyallia Lough, H9 Clare, C. M. Brady, 22 June 2002 (NMW). Lough Derg, H10 North Tipperary, 18 September 2003, L. Houston (NMW),. Loughmerans, H11 Co Kilkenny, 29 September 2006, T. C. G. Rich, R. N. Goodwillie & D. A. Cann (NMW). Pilltown, H12 Co Wexford, 11 August 1958, D. A. Webb (TCD). Priesthaggard, H12 Co Wexford, 17 September 2003, L. Houston (NMW),. Gort, west of, HIS SE Galway, 20 August 1952, T. G. Tutin (TCD). Clifden, H16 West Galway, 10 June 1990, T. G. Evans (NMW). Errislannan Peninsula, H16 West Galway, 5 September 1979, D. A. Webb (TCD). Cong, H17 NE Galway, 30 August 2005, D. T. Price (NMW). Tullamore, H18 Offaly, 1978, D. A. Webb (TCD). Between Greenane and Rathdrum, H20 Co Wicklow, 11 September 1978, D. A. Webb (TCD). Ballybetagh, H21 Co. Dublin, 25 September 2004, P. H. Carvill (NMW),. Killowen, 25 September 2006, T. C. G. Rich & D. A: Cann (NMW). Loch Fea, H36 Tyrone, 28 September 2006, T. C. G. Rich & D. A. Cann (NMW). Inishmaine, H26 East Mayo, 30 August 2005, D. T. Price (NMW). Clonbur, H26 East Mayo, 30 August 2005, D. T. Price (NMW). SORBUS ‘MENAI STRAIT’ Menai Strait, 19 September 1985, T. C. G. Rich (NMW). Menai Strait, 21 August 1989, G. Hutchinson (NMW). Nant Porth, 12 June 1990, M. C. F. Proctor (NMW). Menai Strait, 9 July 2002, T. C. G. Rich & R. A. Jones (NMW). Upper Bangor, 19 September 2003, T. Dines & W. McCarthy (NMW). Watsonia 27: 109-118 (2008) 109 Conservation of Britain’s biodiversity: status of the two Wye Valley endemics Hieracium pachyphylloides, Carboniferous Hawkweed and H. vagicola, Tutshill Hawkweed (Asteraceae) J. SAWTSCHUK Université de Bretagne Occidentale — Institut de Géoarchitecture CS 93837 29238 Brest cedex 3, France and W, G, Ge lhe Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP, UK ABSTRACT Reviews of historical data combined with field surveys have been carried out to assess the status of two hawkweeds endemic to the Wye Valley, Hieracium pachyphylloides and H. vagicola. H. pachyphylloides has declined from six sites to one site, and H. vagicola from two sites to one site. Both species are ‘Critically Endangered’ under the I.U.C.N. threat criteria. The main threats are rock climbing, closure of woodland canopies and spread of alien plants. Conservation action is urgently required. KEYWORDS: endemic, England, I.U.C.N. Threat Criteria, rock climbing, Wales. INTRODUCTION There are two Hieracium species endemic to the lower Wye Valley on the border between England and Wales, H. pachyphylloides (Zahn) Roffey, Carboniferous Hawkweed, and H. vagicola P. D. Sell, Tutshill Hawkweed (Sell & Murrell 2006). Both are rare plants about which very little is known, so a review of the historical data has been combined with field surveys to determine their current statuses and needs for conservation, as required under the Global Plant Conservation Strategy (Secretariat for the Conservation of Biodiversity 2002). The data are summarised here; full details can be found in Sawtschuk (2006). The glaucous-leaved hawkweed in the Wye Valley gorge between Ross and Monmouth now known as H. pachyphylloides was known to W. H. Purchas and A. Ley for many years under the name H. caesium Fries (Purchas & *e-mail: tim.rich@museumwales.ac.uk Ley 1889). Following the clarification that H. caesium was a Scandinavian plant which did not occur in Britain, and that the Wye Valley plants differed from other plants also once referred to H. caesium, Purchas described it as H. murorum L. var. pachyphyllum (Purchas 1895). Williams (1902) raised this variety to species rank as H. pachyphyllum but the combination was invalid as that name was already in use for a different European species described by Brenner in 1892. Zahn (1921) treated it as a subspecies of H. murorum, selecting the new epithet pachyphylloides as the combination H. murorum L._ subsp. pachyphyllum Brenner also already existed. Roffey (1925) then raised it to its current status as a species using Zahn's epithet. It is a member of Section Hieracium. Hieracium vagicola has only recently been described (Sell & Murrell 2006). Plants were formerly included under H. subbritannicum (Ley) P. D. Sell & C. West, another endemic of the Wye Valley and South Wales, but differ mainly in the dense stellate hairs on the involucral bracts. It is a member of Section Stelligera. The species are illustrated in Figures | and 2, and are described in detail in Sell & Murrell (2006). Key identification features to help separate them from the other glaucous-leaved species recorded in the Wye Valley (H. schmidtii Tausch, H. subplanifolium Pugsley and A. subbritannicum) are for H. pachyphylloides, the weakly toothed, truncate- based leaves with numerous pale, simple eglandular hairs abaxially and the greyish- green involucral bracts with many, black-based 110 J. SAWTSCHUK AND T. C. G. RICH FIGURE |. Hieracium pachyphylloides. A, Whole plant. B—C, Outer rosette leaves, upper surfaces. D, Inner rosette leaf, lower surface. E, Stem leaf. F, Capitulum. G, Involucral bract. Scale bars 1 cm. Del. T. Rich. HIERACIUM PACHYPHYLLOIDES AND H. VAGICOLA 111 FIGURE 2. Hieracium vagicola. A, Whole plant. B, Vegetative rosette. C, Outer rosette leaf, upper surface. D, Inner rosette leaf, lower surface. E, Inner rosette leaf, upper surface. F, Involucral bract. Scale bars | cm. Dele De Rich: 112 J. SAWTSCHUK AND T. C. G. RICH simple eglandular hairs, few glandular hairs and dense stellate hairs, and for H. vagicola the slender stems and strongly toothed basal leaves with the lowest teeth on either side somewhat reflexed at the truncate base with few to numerous simple eglandular hairs abaxially, and involucral bracts blackish-green with a long-drawn out filamentous apex and few to numerous simple eglandular hairs and gland- ular hairs, and dense stellate hairs. METHODS Historical date were compiled from herbaria (BEL, BIRM, BM, CGE, E, LIV, MANCH, NMW), the literature and correspondence with local botanists. The records traced are given in Appendices | and 2. Field surveys were carried out between 2003 and 2007 using the historical records to target areas to search. As both species occur in Carboniferous Limestone cliffs and quarries, they were searched for by walking along the tops and bottoms of the cliffs, and by using binoculars. Some limited climbing with roped access was carried out at Symonds Yat. Access to rocks in one place at Symonds Yat was restricted due to nesting peregrines. Reasons for possible loss of populations and threats to extant populations were assessed in the field. Voucher specimens and photographs have been deposited in NMW. Associated species were recorded in estimated 2 m x 2 m quadrats around the species. Soil pH was measured with a pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water on soil samples collected from around the roots. RESULTS HISTORICAL RECORDS About 12 populations of H. pachyphylloides have been recorded in six localities in three vice-counties (Appendix 1). In three localities it has only been collected once about 100 years ago and all are imprecise or ambiguous. The limestone cliffs near Chepstow on the Glouc- estershire side of the Wye could refer to any of the cliffs and quarries from Tutshill to Ban-y- gor rocks. Brockweir Common is disused name for the general area around St Briavels and the Hudnals, an area of hilly pastures (Peterken 2005), and it is not clear exactly where the hawkweed might have occurred as there are no obvious cliffs. The third record ‘Bigsweir opposite Tintern’ is ambiguous as Bigsweir is c. 5 km north of Tintern, and cannot be described as ‘opposite’; it is possible that the record refers to the Shorn Cliff, opposite Tintern Abbey. The fourth H. pachyphylloides locality around Symonds Yat and Coldwell Rocks includes cliffs, quarries and the railway line, where it was frequently collected in several places in this area until the 1960s. It was collected four times at Piercefield Park but has not been seen for over 100 years. It has occurred in several places on the Great Doward, where it was last collected in 1910. The H. pachyphylloides literature record for Stroud in Riddelsdell et al. (1948) is best not accepted without a voucher. There is a correctly identified specimen from Pwll Byffre in BM which is likely to either be a confused label/specimen or a _ misidentification (no glaucous-leaved plants were present there in 2004 or 2005 and there are no other collections for this well known site). Records for Ireland refer to H. basalticola Pugsley. Hieracium vagicola has been recorded in two areas in the Tutshill to Ban-y-gor rocks area (Appendix 2). Shoolbred (1920) cited it from Beaucliffe, an old name for Cockshoot, but we have traced no voucher material (a possible specimen in NMW is labelled Bannagher [=Ban-y-gor?] cliffs but is not H. vagicola). The cliffs and quarries from Pen Moel to Lancaut form one large locality where it has occurred scattered in various subpopulations, and was last recorded in 1984. A record for the Symonds Yat area, whilst plausible, is based on a rather confusingly labelled specimen, and is best rejected. FIELD SURVEYS 2003-2007 Both hawkweeds grow on cliffs and rocky places which are difficult to search thoroughly, and consequently we made several visits to sites to try to refind the plants in their historic sites (Kéry ef al. (2006) found between two and four visits were usually required to be fairly certain of refinding rare species). The quality of the historical information does not always make refinding sites simple and we often had to use our intuition about Hieracium ecology to find suitable places to search. HIERACIUM PACHYPHYLLOIDES AND H. VAGICOLA 3) Both hawkweeds lost much of their distinctive glaucous leaf coloration later in the season, and we recommend future surveys are carried out in May—June when the glaucous coloration on the early leaves is strong and the plants are flowering. Hieracium pachyphylloides was refound in three subpopulations of 22, 32 and 19 plants in the Symonds Yat and Coldwell Rocks area, where they had not been recorded for nearly 50 V.c. 36 Herefordshire Great Doward Monmouth V.c. 35 Monmouthshire years. It was not refound in the Tutshill to Ban- y-gor rocks area (3 visits), Brockweir Common (3 visits), Bigsweir (2 visits, including Shorn Cliff from below only), Piercefield Park (3 visits) or the Great Doward (7 visits). Hieracium vagicola was refound in one population of 126 plants at Woodcroft Quarry. It was not be refound at Pen Moel (1 visit), Lancaut (2 visits) or Cockshoot, Ban-y-gor (1 visit). The records are mapped in Figure 3. ( ie Coldwell Rocks OO _@ Symonds Yat V.c. 34 West Gloucestershire Bigsweir St Briavels O C Brockweir Tintern ea? Cockshoot OQ Ban-y-gor (of Lancaut Piercefield Cliffs ~ 4@—wooderoft Pen Moel Chepstow Severn Estuary FIGURE 3. Distribution of Hieracium pachyphylloides and H. vagicola in the Wye Valley. H. pachyphylloides: L], old records; @, 2006. H. vagicola: O, old records; @, 2006. 114 J. SAWTSCHUK AND T. C. G. RICH HABITATS Hieracium pachyphylloides occurred in crevices and ledges on vertical Carboniferous Limestone rock faces and on the flatter tops of rocky bluffs. Typical associated herbs included Asplenium trichomanes L., A. ruta-muraria L., Geranium robertianum L., Melica uniflora Retz., Scabiosa columbaria L. and_ other Hieracium species in scrappy, open, herbaceous vegetation in open woodland. Overall the vegetation is probably best attributed to the W8g Fraxinus excelsior — Acer campestre — Mercurialis perennis woodland Teucrium scorodonia subcommunity of Rodwell (1991). Although this is a woodland community, the hawkweed is a light-demanding species which does not tolerate shade and is only found in the more open areas. Two soil pHs measured were pH 6:5 and 7-2. Hieracium vagicola occurred in crevices and ledges on vertical Carboniferous Limestone rocks and on quarry spoil heaps. Associated species included Bromopsis erecta (Huds.) Fourr., Festuca ovina L. and Scabiosa columbaria L. The more natural crevice vegetation is probably best referred to the OV39 Asplenium trichomanes — A. ruta-muraria community of Rodwell (2000) but we have been unable to access it safely to record it in detail. On the quarry spoils heaps H. vagicola occurs in a disturbed, open, scrappy scrub community of Fraxinus excelsior L., Ligustrum vulgare L., Rosa spp. and Buddleja davidii Franch., probably best included in the W21d Crataegus monogyna-Hedera helix scrub, Viburnum lantana subcommunity of Rodwell (1991). The soil from a spoil heap was pH 7:1. DISCUSSION The field survey showed that Hieracium pachyphylloides has declined from six sites to one site, and H. vagicola from two sites to one site. Both species could occur elsewhere on the vast areas of cliff in the Wye Valley which we have not surveyed as there are no historical records. Both species can be classified I.U.C.N. (2001) threat criteria as Endangered’ due to their small, declining populations and clear, continuing threats to their survival. Although much of the Wye Valley has numerous nature conservation designations at national (S.S.S.I.) and international (S.A.C.) level and large-scale under the ‘Critically conservation projects such as the Ravine WoodLIFE project, the designations will do nothing to protect either species; specific conservation action is urgently required for both. The probable major reason for loss or decline of both Hieracium species from many of the wooded cliffs, quarries and railway lines is shading due to closure of canopy by trees or by scrub invasion. Historically the woodlands in the Wye Valley would have been managed, with the coppicing and clearance regularly creating suitable open habitat for many light- demanding species such as H. pachyphylloides and H. vagicola. Old photographs of the Wye valley (Helme 1989; Rainsbury 1989) show some of the rocks and cliffs, especially around Symonds Yat and at Pen Moel, to have been much more open in the past when the areas were quarried or managed. Some _ scenic pathways giving access to areas such as Piercefield or Coldwell Rocks are no longer maintained for the views and have shaded over. To what extent both hawkweeds have been lost directly due to quarrying of their limestone rock habitat is unclear, but both have benefited at least temporarily from the open conditions created by quarrying. The cessation of quarrying has allowed scrub and woodland to reinvade and has resulted in loss. Hieracium pachyphylloides was also lost from the Great Doward quarry by infilling after quarrying had ceased. The closure of the Wye Valley railway has allowed scrub invasion of the cuttings (previously kept open to minimise fire risk) where H. pachyphylloides once occurred. The reason for loss from the Brockweir Common area is not known but it could have been a transient population. There are two immediate threats to survival of H. pachyphylloides. First, the continuing closure of the canopy which will cause further loss by shading at Coldwell Rocks. Second, two subpopulations at Symonds Yat are seriously endangered by, yet partly maintained by, rock climbing. Climbing is very popular at Symonds Yat (Willson 1999) and during the good weather there are regularly large groups of school children on adventure activities during the week, and literally hundreds of recreational climbers may be present at the weekends. On the one hand the climbers trample the vegetation (including UH. pachyphylloides) underfoot and sometimes very locally remove plants from the cliff faces (McMillan & Larson 2002; Miiller et al. 2004; HIERACIUM PACHYPHYLLOIDES AND H. VAGICOLA 115 Kuntz & Larson 2006), but on the other hand they help to maintain the cliff bases and tops open by their access paths. The balance between damage and benefit is very fine. Stopping the climbing activities completely would probably result in the plants being shaded out within a few years. The best way to protect H. pachyphylloides in this area is to regularly monitor the populations and work with the rock climbing community to control access to specific climbs when threats are perceived. There are three main threats to H. vagicola. First, climbing, which is not as popular at Woodcroft Quarry as at Symonds Yat, and does not include school groups. At this quarry, the forth-coming fourth edition of Willson’s (1997) climber’s guide will specifically mention H. vagicola with a request for its conservation and advise climbers of its occurrence including re-routing of some climbs. Second, further development of the quarry, which still has extant extraction permissions, could result in further loss. The quarry is excluded from the Lower Wye Gorge S.5.5.1. and thus H. vagicola is — very vulnerable. Third, spread of alien plants such as Buddleja or Cotoneaster microphyllus Wallr. which are both already well-established in the quarry. Further work is required to ensure ex-situ conservation of both these hawkweeds. A small batch of seed material of H. pachyphylloides has been deposited in the Millennium Seed Bank. One plant of H. vagicola is currently held in cultivation. ACKNOWLEDGMENTS Collation of the historical data for H. pachy- phylloides was funded by the Countryside Council for Wales. We would like to David McCosh and Peter Sell for their considerable help and advice, Trevor Evans, Mark and Clare Kitchen, George Peterken and Stephanie Thompson for information, Claire Jusseau for help with field work, Deborah Spillards, Gina Murrell and the Keepers of the herbaria for access to libraries and specimens, and Libby Houston, John Willson and Paul Wood for discussions on climbing. REFERENCES HELME, A. (1989). Monmouth and the River Wye in old photographs from Goodrich to Brockweir: from the collections of Monmouth Museum. Alan Sutton Publishing, Gloucester. HIERACIA STUDY GROUP (1985). Reports of Meetings. Chepstow, Monmouthshire, VC35, Sunday 10th June 1984. Hieracia Group Notes 8: 34. B.S.B.1. 1.U.C.N. (2001). LU.C.N. Red list categories and criteria. Version 3.1. International Union for Conservation of Nature, Gland. KERY, M., SPILLMANN, J. H., TRUONG, C. & HOLDEREGGAR, R. (2006). How biased are estimates of extinction probability in revisitation studies? Journal of Ecology 94: 980-986. KUNTTZ, K. L. & LARSON, D. W. (2006). Influences of microhabitat constraints and rock-climbing disturbance on cliff-face vegetation communities. Conservation Biology 20: 821-832. MCMILLAN, M. A. & LARSON, D. W. (2002). Effects of rock climbing on the vegetation of the Niagara Escarpment in southern Ontario, Canada. Conservation Biology 16: 389-398. MULLER, S. W., RUSTERHOLZ, H. & BAUR, B. (2004). Rock climbing alters the vegetation of limestone cliffs in the northern Swiss Jura Mountains. Canadian Journal of Botany 82: 862-870. PETERKEN, G. (2005). Flowers in the fields. Parish Grasslands Project, Hewelsfield and St Briavels. PURCHAS, W. H. & LEY, A. (1889). A flora of Herefordshire. Jakeman & Carver, Hereford. PURCHAS, W. H. (1895). Hieracium murorum var. pachyphyllum, n. var. Journal of Botany 33: 114-115. RAINSBURY, A. (1989). Chepstow and the River Wye in old photographs from Severn to Tintern: from the collections of Chepstow Museum. Alan Sutton Publishing, Gloucester. RIDDELSDELL, H. J., HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. T. Buncle & Co., Arbroath. RODWELL, J. S., ed. (1991). British plant communities. Volume 1. Woodlands and scrub. Cambridge University Press, Cambridge. RODWELL, J. S., ed. (2000). British plant communities. Volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge. ROFFEY, J. (1925). Hieracium, in F. J. HANBURY ed., The London catalogue of British plants. Edition 11. G. Bell & Sons, London. SAWTSCHUK, J. (2006). Conservation of endemic Hieracium species in the British Isles and assessment of four Welsh species: Hieracium pachyphylloides, Hieracium pseudoleyi, Hieracium rectulum and Hieracium cambricogothicum. ESEB Masters Thesis, Université de Rouen, France. 116 UE SAQWiTS CHIGIKGAIN DD ai G7 GaRiGh SECRETARIAT FOR THE CONSERVATION OF BIODIVERSITY (2002). Global strategy for plant conservation. Secretariat for the Conservation of Biodiversity, Montreal. SELL, P. D. & MURRELL, G. (2006). Flora of Great Britain and Ireland. Volume 4. Cambridge University Press, Cambridge. SHOOLBRED, W. A. (1886). Recent additions to the flora of West Gloucester and Monmouth. Journal of Botany 32: 263-271. SHOOLBRED, W. A. (1920). The flora of Chepstow. Taylor & Francis, London. WADE, A. E. (1970). The flora of Monmouthshire. National Museum of Wales, Cardiff. WILLIAMS, F. N. (1902). Prodromus Florae Brittanicae. C. Stutter, Brentford. WILLSON, J., ed. (1997). Wye Valley. 3rd edition. Climbers’ Club Guides, Leicester. WILLSON, J., ed. (1999). Symonds Yat. Climbers’ Club Guides, Leicester. ZAHN, K. H. (1921). Compositae — Hieracium, in A. ENGLER ed., Das Pflanzenreich IV. 280. 76: 450. Engelmann, Berlin. (Accepted May 2008) HIERACIUM PACHYPHYLLOIDES AND H. VAGICOLA 117 APPENDIX 1. HISTORICAL RECORDS OF HIERACIUM PACHYPHYLLOIDES V.C. 34 WEST GLOUCESTERSHIRE Near Chepstow (ST59). Limestone cliffs and quarry base, Gloucestershire side of Wye, near Chepstow, 18 September 1893, W. A. Shoolbred (NMW). Brockweir (SO50). Brockweir Common, wood, 20 June 1894, W. A. Shoolbred (NMW). Bigsweir (SOS0). Bigsweir, opposite Tintern, July 1911, H. J. Riddelsdell (BM). Symonds Yat (SOS51). Symonds Yat, 7 June 1889, A. Ley (BM, CGE). Symonds Yat, 26 May 1896, A. Ley (BM, Set of British Hieracia no. 64). Symonds Yat, limestone rocks above the station, 12 May 1894, A. Ley (BM, NMW,CGE). Symonds Yat, May 1894, Captain Steuart (NMW). Symonds Yat, May 1894, A. Ley (BM). Symonds Yat, 12 June 1901, E. S. Marshall (CGE). Symonds Yat, 30 May 1904, A. Ley (LIV, NMW). Symonds Yat, railway bank, 30 May 1904, S. H. Bickham & A. Ley (BM, CGE). Symonds Yat, 17 June 1904, E. F. Linton (Set of British Hieracia no. 64. LIV, NMW). Symonds Yat, railway bank, 22 May 1905, S. H. Bickham (BM, LIV). Symonds Yat, bank the station, 22 June 1908, A. Ley (BM). Symonds Yat, 28 May 1913, H. J. Riddelsdell (BM, NMW, CGE, Riddelsdell et al. 1948). Symonds Yat, railway bank, August 1922, H. S. Redgrove (BM). Symonds Yat, near station, 2 June 1953, J. E. Raven (CGE, herb. McCosh). Symonds Yat, railway bank near station, 14 June 1953, C. A. E. Andrews (BIRM). Symonds Yat, north side of railway, east end of tunnel, May 1956, C. A. E. Andrews (BIRM). Symonds Yat, 2 June 1956, F. R. Browning (E). Symonds Yat, 17 June 1962, cutting at E. end of Symonds Yat Railway tunnel, B. A. Miles (CGE). Symonds Yat, bank of disused railway (cultivated), 5 July 1964, J. N. Mills (MANCH). Coldwell Rocks (SO5616). Coldwell Rocks, 1847, W. H. Purchas (Purchas & Ley 1889). Coldwell Rocks, 23 June 1868, on the Wye, W. H. Purchas (CGE). Coldwell, near Lydbrook, 9 June 1886, A. Ley (CGE). Coldwell, limestone rocks, 5 June 1890, A. Ley (BM). Coldwell, quarry, 13 June 1899, A. Ley (BM). V.C. 35 MONMOUTHSIRE Piercefield (ST59). Piercefield, limestone cliffs above the Wye, 5 May and 3 September 1893, W. A. Shoolbred (NMW; Wade 1970). Piercefield Cliffs, June 1894, W. A. Shoolbred (NMW). Piercefield, limestone cliff, 4 June 1897, W. A. Shoolbred (BM). V.C. 36 HEREFORDSHIRE Great Doward (S051). Great Doward, limestone debris, B. Watkins, 1868 (Purchas & Ley 1889). Great Doward Quarries, on the limestone debris in the Great Quarry, 20 June 1881, 28 June 1881, and lime point near the Quarry Limestone, 8 June 1886 A. Ley (CGE). Great Doward Hill, 2 June 1882, A. Ley (BM). Great Doward Quarries, 2 June 1882, A. Ley (BIRM). Great Doward Hill, 10 June 1889, F. J. Hanbury (BM). Great Doward Quarries, 9 June 1886, A. Ley (CGE). Great Doward Hill, rocks about the caves, 7 June 1889, A. Ley (CGE). Great Doward Hill, limestone, 12 July 1889, F. J. Hanbury (BM). Great Doward, rocks near the King Arthur’s cave, 9 October 1893, A. Ley (CGE). Great Doward, in the quarry, 12 May 1894, A. Ley (CGE, BM, LIV, NMW;; Purchas & Ley 1889). Great Doward, 31 May 1897, E. F. & W. R. Linton (CGE, LIV). Great Doward, limestone quarries, 31 May 1897, A. Ley (BM, LIV, NMW;; Set of British Hieracia no. 64). Great Doward, May 1889, E. Armitage (NMW). Great Doward Hill, 21 May 1902, A. Ley (BM, LIV). Great Doward Hill, 10 June 1902, A. Ley (NMW). Great Doward Hill, August 1903, H. J. Riddelsdell (BM). Doward, May 1904, E. Armitage (NMW). Great Doward, 22 May 1905, A. Ley (LIV). Great Doward, 1 June 1910, H. J. Riddelsdell (BEL, BM, CGE, NMW). DOUBTFUL RECORDS: Stroud, banks near, v.c. 34 West Gloucester (Riddelsdell et al. 1948). Pwll Byffre, v.c. 42 Breconshire, 10 July 1929, H. J. Riddelsdell (BM); possibly a confused label. 118 J. SAWTSCHUK AND T. C. G. RICH APPENDIX 2. HISTORICAL RECORDS OF HIERACIUM VAGICOLA V.C. 34 WEST GLOUCESTERSHIRE Cockshoot, Ban-y-gor (ST59). Beaucliffe (Shoolbred 1920). Lancaut (ST59). Lancaut cliffs (Shoolbred 1886, 1920). One plant (as H. subbritannicum), 10 June 1984, Hieracia Study Group (Hieracia Study Group 1985). Pen Moel (ST59). Pen Moel, limestone cliffs and debris under quarry, 25 June 1891, 8 July 1891, 5 May 1893, 16 May 1894 and 27 August 1895, W. A. Shoolbred (BM, CGE, NMW; Shoolbred 1886, 1920). Pen Moel cliffs, 26 June 1895, E. S. Marshall (CGE). Pen Moel rocks, 6 September 1900, A. Ley (BM, CGE, NMW;; Set of British Hieracia no. 136). Pen Moel, 12 June 1913, W. C. Barton (BM). Pen Moel, rocky bank by road, 31 July 1916, L. B. Hall (BM). Pen Moel, rocks by roadside, 30 July 1920, C. E. Salmon (BM). Pen Moel cliffs, below, 24 June 1932, H. W. Pugsley (BM). Pen Moel, 1958, C. West (cultivated as H.105, CGE). Woodcroft (ST59). Woodcroft (Shoolbred 1920). Woodcroft, August 1922, H. S. Redgrove (BM). Woodcroft, bank at, 24 June 1932, H. W. Pugsley (BM). Woodcroft, ST540960, 1958, P. D. Sell (CGE). DOUBTFUL RECORDS: Symonds Yat, date ‘forgotten’ with | June 1895 added later, A. Ley (BM; sheet has a series of mixed and confusing labels). Watsonia 27: 119-125 (2008) 119 The status of Hieracitum arranense and H. sannoxense (Asteraceae), two endemic hawkweeds from the Isle of Arran, Scotland Ce Gs RICH: Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP and D. J. McCOSH Baconsthorpe Old Rectory, Holt, Norfolk NR25 6LU ABSTRACT Hieracium arranense and H. sannoxense are very rare Scottish endemics confined to the Isle of Arran. A review of the records and field work was carried out in 2007. There are about 225 plants of H. arranensé in 81x sites; it was not refound in two sites. About 27 plants of H. sannoxense were found in two Siicsmmenderetie: IUCN. Whreat Criteria, #7: arranense is ‘Endangered’ and H. sannoxense is ‘Critically Endangered’. KEYWORDS: Arran Hawkweed, Criteria, Sannox Hawkweed. ELUCN, Threat INTRODUCTION Sell & Murrell (2006) described two new hawkweeds endemic to the Isle of Arran (v.c. 100 Clyde Islands), Scotland: Hieracium arranense P. D. Sell, Arran Hawkweed and H. sannoxense P. D. Sell, Sannox Hawkweed (Asteraceae). The only information known about them was that cited by Sell & Murrell (2006), namely that H. arranense occurred in several places on the west coast of Arran and that H. sannoxense was only known at Sannox Bridge. To assess their needs for conservation, and to provide an I.U.C.N. Threat Category as required under the Global Plant Conservation Strategy (Secretariat for the Conservation of Biodiversity 2002), field surveys of the known sites were carried out in June 2007 and are reported in summary herein. Full details are given by Rich & McCosh (2008) and have been deposited with Scottish Natural Heritage, Scottish Wildlife iimsteminicw bss: ba. V-C. Recorder and the B.S.B.I. Threatened Plants Database. *e-mail: tim.rich@museumwales.ac.uk Both species belong to Hieracium section Subalpina Pugsley, whose members typically have dark heads, often few, on long, straight peduncles, ligules often pilose-tipped and leaves often with some microglands (Sell & Murrell 2006). Hieracium arranense (Fig. 1) has mostly ovate, spotted or blotched or unmarked, weakly glaucous to green, entire to acuminate- toothed rosette leaves with numerous simple eglandular hairs abaxially, O-—2 reduced stem leaves with stellate hairs abaxially, involucral bracts with numerous stellate hairs (especially on margins), frequent simple eglandular hairs and sparse glandular hairs, glabrous-tipped ligules, discoloured styles (discoloration more apparent in dried material), and reddish-black achenes (Sell & Murrell 2006). Our observation of plants in the field show it also has sparse stellate hairs on the midrib, yellow rather than discoloured styles in fresh plants, and blackish rather than reddish achenes. It varies in size and leaf toothing and coloration, possibly depending on habitat. Hieracium sannoxense (Fig. 2) has ovate to elliptic, finely denticulate, unspotted, green leaves in a basal rosette, with no or with 1-2 small, reduced stem leaves; involucral bracts acute with many, long simple eglandular hairs, some glandular hairs and a few stellate hairs; and yellow styles when fresh (discoloured with age). Hieracium sannoxense looks like some species of Hieracium section Cerinthoidea, and could well have originated through hybrid- ization of a member of section Cerinthoidea with a species from section Subalpina or section Alpina (Sell & Murrell 2006). 120 T€-G: RICH AND DP IOMceOSH FIGURE 1. Hieracium arranense. A, whole plant. B, vegetative rosette. C—E, rosette leaves (upper sides C, E; lower side D). F, stem leaf, upper side. G, involucral bract. Scale bars 1 cm. PR EO A ns cece NB HIERACIUM ARRANENSE AND H. SANNOXENSE 121 FIGURE 2. Hieracium sannoxense. A, flowering plant. B, plant after flowering. C—E, rosette leaves (upper sides C, E; lower side D). F, involucral bract. Scale bars 1 cm. 122 T. C. G. RICH AND D. J. McCOSH ye Lochranza Blackwaterfoot Ww 5 km 9 annox pss Brodick mal on) 0 FIGURE 3. Distribution of Hieracium arranense (®, 2007; O, pre-2000) and H. sannoxense (™, 2007) on Arran, Scotland. Sites are numbered in clockwise order around the island separately for each species. The Isle of Arran is situated in the Firth of Clyde in South-west Scotland. It is about 32 km long by 7 km wide and has a varied geology (Geological Survey 1928). The core of the northern half of the island is dominated by mountains formed from Tertiary igneous rocks, and rises to 874 m (Goat Fell). The southern half of the island is lowland, and is formed from Devonian and Carboniferous sedimentary rocks cut by igneous intrusions. The rocks and soils in the north are generally acidic, whilst those in the south are locally more calcareous. Igneous intrusions in the form of sills and dykes are widespread in the south and around the coast in the north, and locally may be base- rich. The climate is relatively mild due to the North Atlantic Drift, with mild, wet winters and cool, wet summers. METHODS Existing information was compiled from herbaria and D. McCosh’s Hieracium database; this database contains about 29000 records compiled over the last 20 years from numerous British and Irish herbaria (BEL, BIRM, BM, HIERACIUM ARRANENSE AND H. SANNOXENSE 123 CGE, DBN, E, GL, HTN, LANC, LCN, LTN, MANCH, NMW and PTH) and other determinations. The only material traced in public herbaria was 5 sheets of H. arranense and 3 sheets of H. sannoxense in Cambridge (CGE), on which the original descriptions were based, and one specimen in E. Sites were surveyed on four days between 19-26 June 2007. Plants were counted as far as practical within the time available, and binoculars were used to count plants on cliffs. Voucher specimens from selected sites have been deposited in BM, E and NMW. Soil pH was measured with a calibrated pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water of dried soil samples collected from around the roots. I.U.C.N. Threat Categories were derived using the criteria in I.U.C.N. (2001). RESULTS Figure 3 shows the sites for both Hieracium species on the Isle of Arran. HIERACIUM ARRANENSE 1. DUN FIONN, SOUTH CORRIEGILLS Hieracium arranense was first recorded at this site by A. R. Church on 17 June 1988 (herb. D. McCosh). The site was visited on 25 June 2007. Hieracium caesiomurorum Lindeb. (one plant) and H. cyclicum P. D. Sell (frequent) were scattered on the north-facing, basalt cliffs of the hill fort at NS046336, but H. arranense was not found. Hieracium caledonicum sensu lato has been recorded on the cliffs c. | km to the south. 2. RUBHA GARBHARD, BROWN HEAD This site was first recorded by A. G. Kenneth on 24 May 1974 (CGE). One plant occurred in a rock cleft in low quartz-porphyry sea cliffs at the back of a raised beach at NR/899261 on 21 June 2007, with Festuca rubra L., Holcus lanatus L., Hypochaeris radicata L., Succisa_ pratensis Moench and Teucrium scorodonia L. No other hawkweeds were seen on the cliffs, which generally appear too acid and dry for Hieracium. ) Two populations occur at the top of the raised beach on flat, vegetated shingle. Four plants occurred in a sward of Pilosella officinarum F. W. Schultz & Sch. Bip. with Agrostis stolonifera L., Anthoxanthum odoratum L., Deschampsia flexuosa (L.) Trin., Galium saxatile L. and Sedum anglicum Huds. at NR900263 (soil pH 6:0), and 16 plants in a more grassy sward of Anthoxanthum odoratum and Festuca rubra with Digitalis purpurea L. and Senecio jacobaea L. at NR899264. Seeds from eight plants were collected for the Millennium Seed Bank. 3. THE DOON, BLACKWATERFOOT The Doon is an impressive hill above Kilbrannan Sound capped by massive quartz- feldspar porphyry columns. The underlying horizontal beds of sandstone are exposed at the north end. This site was first recorded in 1987 by A. McG. Stirling as ‘Drumadoon’ (herb. D. McCosh). Twenty seven H. arranense plants occurred scattered over the sandstone rocks in a small area at NR886294 with Dactylis glomerata L., Festuca rubra, Koeleria macrantha (Lebed.) Schult., Plantago maritima L., Primula vulgaris Huds. and Preridium aquilinum (L.) Kuhn. Two probable plants of H. arranense occurred out of reach nearby on the otherwise very species-poor, quartz-feldspar porphyry cliffs at the north end of the hill. 4. BOULDER NORTH OF THE DOON, BLACKWATERFOOT An isolated erratic boulder of igneous origin in the middle of the broad raised beach at NR886296 was covered with 62 plants of H. arranense, mostly as very small rosettes rooted in the rock crevices, associated with Armeria MAGviNG = Wilde estwea. sovinay ln. Hypochaeris radicata, Rumex acetosa L. and Sedum anglicum. The soil was a_ blackish, organic rendzina of pH 4:4. Plants were only flowering on the sheltered eastern side. Seed from two plants was collected for the Millennium Seed Bank. 5. KING’S CAVE, TORR RIGH MOR, BLACKWATERFOOT This site was first recorded by A. Somerville on 10 July 1895 (E), and again by A. G. Kenneth and P. D. Sell on 2 June 1974 (CGE). It is the type locality. Hieracium arranense was frequent on the sandstone cliffs and basalt about King’s Cave on 21 June 2007. About 50 plants occurred out of reach on the sandstone cliffs from c. 200 m south of (c. NR885307), to 30 m north of (NR884309), King’s Cave, but only a few accessible plants could be checked. 35 plants 124 TG: RICEVAND DE Jn Mce€OSH occurred on the basalt dyke where the footpath cuts up the cliff at NR884312, associated with Brachypodium sylvaticum (Huds.) P. Beauv., Calluna vulgaris (L.) Hull, Erica cinerea L., Festuca rubra, Luzula_ sylvatica (Huds.) Gaudin, Polypodium sp., Sedum anglicum and Teucrium scorodonia on a blackish organic soil of pH 5-8. Plants flowered freely on the sheltered eastern side of the dyke away from the sea, where seed was collected from one plant for the Millennium Seed Bank. A small population of c. 5 plants occurred on sandstone rocks which differed from typical H. arranense with its weakly glaucous leaves with yellowish styles in having green foliage and discolored styles; in all other respects they are similar to H. arranense and are assumed to be local variants. One large H. arranense plant occurred on the newly made bank of a ditch by the new footpath to the car park behind and above King’s Cave at NR885307; this plant is unlikely to survive in the long term as the ditch sides colonize with vegetation. 6. CNOCAN CUALLAICH CLIFFS, DOUGARIE This site was first recorded on 24 May 1974 by A. G. Kenneth and again a week later on | June 1974 by A. G. Kenneth and A. McG. Stirling (CGE). A small population of H. arranense occurred scattered on west-facing Devonian sandstone and conglomerate cliff outcrops above bracken behind improved pasture, though most plants were inaccessible on the cliffs. 14 plants occurred in crevices at NR890353 with Asplenium marinun L., Festuca — rubra, Plantago maritima and Rosa sp. One plant occurred on a_ south-west-facing rock at NR891354, and six plants occurred on wet cliffs immediately south-east of the tiny graveyard, with Agrostis stolonifera, Carex flacca Schreb., Eupatorium cannabinum L., Succisa pratensis and sheets of bryophytes dripping with calcareous water; this is an unusual habitat. 7. IMACHAR Three plants were found on a basalt dyke in a field below the road, with H. deganwyense Pugsley (3 plants) and H. vulgatum Fr. (5 plants) at NR864402 on 22 June 2007. 8. ABHAINN BHEAG, CATACOL This site was first recorded on 24 May 1974 by A. G. Kenneth at NR919490 (CGE). The ravine was searched on 22 June 2007 in rain without success, though there were a number of vegetative rosettes which could have been H. arranense in the area indicated by the grid reference at the bottom of the gorge, but none with spotted leaves. Hieracium coniops Norrl. occurred on the slate on the northern side, and H. cyclicum and H. rubicundiforme (Zahn) Roffey on the basalt dyke on the south side, and one plant of H. eucallum P.D. Sell & C. West was found by the lower dam. Hieracium boswellii E. F. Linton, 4. glandulidens P. D. Sell & C. West and H. argenteum Fr. have also been recorded previously but were not refound. Above the main lower waterfall, only H. vulgatum agg. was seen on the shaded north bank near Sorbus pseudofennica E. F. Warb. Above the second main waterfall the ravine opens out and no Hieracium plants could be seen’ with binoculars. HIERACIUM SANNOXENSE 1. GLEN SANNOX Hieracium sannoxense was first collected at Sannox Bridge by A. G. Kenneth and A. McG. Stirling on | June 1974 (CGE, the grid reference cited on the specimen is erroneous). It was collected again on 14 June 1976 by A. G. Kenneth (CGE). A small population totalling 17 plants was found in Glen Sannox scattered up and down the stream on sandstone rocks around the footbridge at NSO10453 on 19 June 2007 (it appears that the type locality is this footbridge and not the A841 bridge at NS016455). At NS011453, - two flowering plants, one vegetative plant and two seedlings were found on low, stream-washed rocks in the river, associated with Campanula rotundifolia L., Hieracium argenteum, Nardus_ stricta L., Solidago virgaurea L. and Viola riviniana Rchb., in very open, sparse vegetation. Four flowering and four vegetative plants occurred at NSO10453 on the southern river bank on bryophyte-rich rocks with Blechnum spicant (L.) Roth., Sanicula europaea L. and Solidago virgaurea, shaded under Betula pubescens Ehrh. and Rosa canina L. Four vegetative plants occurred in rocks in the stream at NS009450. Hieracium argenteum and H. — vulgatum were both more frequent and more widely scattered along the river. The sandstone ge A SS I LILO ALLL: SE A a ae r et HIERACIUM ARRANENSE AND H. SANNOXENSE 125 has some mildly basic characteristics, and barytes used to be mined in this area. Downstream the burn becomes heavily shaded in the area managed by Corrie Golf Club. Upstream the woodland ceases at about the small weir at NS008452, where the glen opens out and the river only has small boulders. 2. SANNOX A new population of c. 10 H. sannoxense plants was found on 19 June 2007 on a large, conglomerate, erratic boulder beside the A841 road at the southern end of Sannox (NSO19448), associated with Deschampsia flexuosa, Festuca rubra and Rubus sp. Seeds from one plant were collected for the Millennium Seed Bank. OTHER SITES SEARCHED AROUND SANNOX The A841 bridge at NS0O16455 had no Hieracium and looked unsuitable. The North Sannox Burn had’ H. argenteum and H. vulgatum c. 100 m below the bridge at Lag nan Sasunnach (NS013465) and A. vulgatum/H. coniops at North Sannox Bridge (NR993468). No Hieracium was seen on waterfalls for c. 1-5 km above North Sannox Bridge. DISCUSSION Our survey indicates that both Arran endemic hawkweeds are very rare. Hieracium arranense has about 225 plants in six sites and was not refound in two sites; it is I.U.C.N. (2001) Threat Category ‘Endangered’. Further searches of the two sites where it was not refound would be worthwhile. Hieracium sannoxense has about 27 plants in two sites, and is [.U.C.N. (2001) Threat Category ‘Critically Endangered’. It is possible that there are other sites for both species as we could not search the whole island. Hieracium sannoxense is the rarest and most threatened, and is a priority for in situ conservation action. The margins of the burn in Glen Sannox are heavily shaded by Alnus glutinosa (L.) Gaertn., Betula pubescens, Fagus sylvatica L., Quercus petraea (Matt.) Liebl., Salix aurita L. and S. cinerea L., and the hawkweeds would clearly benefit from opening up the canopy, and reduction of the sheep and deer grazing. The conglomerate boulder at Sannox is under threat of scrub encroachment and spread of Rubus. Hieracium arranense is less threatened, mainly occurring on open rocks kept open by their exposed situation, and probably does not require active conservation at this stage. Both species should be monitored every five years. To judge from the records in the Hieracium database and our field work, in general Arran is not rich in Hieracium species. 23 Hieracium species have been recorded, though most are infrequent and very local, and include two other species of interest. Hieracium cyclicum is a very locally distributed species of the west coast of Scotland and a few isolated sites in Ireland; it is scattered around Arran on con- glomerate and basalt rocks. Hieracium coniops, another uncommon species in SW Scotland (Sell & Murrell 2006), seems to be as frequent on Arran as its close relative H. vulgatum. ACKNOWLEDGMENT We are grateful to Gina Murrell for images of the specimens in CGE. REFERENCES I.U.C.N. (2001). .U.C.N. Red list categories and criteria. Version 3.1. International Union for Conservation of Nature, Gland. GEOLOGICAL SURVEY (1928). The geology of Arran. Geological Society memoirs — Scotland. H. M. S. O., London. RicH, T. C. G. & McCosH, D. J. (2008). The status of Hieracium arranense and Hieracium sannoxense, two endemic hawkweeds from Arran. Unpublished report, National Museum of Wales. SELL, P. D. & MURRELL, G. (2006). Flora of Great Britain and Ireland. Volume 4. Cambridge University Press, Cambridge. SECRETARIAT FOR THE CONSERVATION OF BIODIVERSITY (2002). Global strategy for plant conservation. Secretariat for the Conservation of Biodiversity, Montreal. (Accepted April 2008) Sid! i - re W if yi 1 y q mT ; cm i 5 ; 7 2 ’ Men i? 7 . . ~ . 7 Me - > Watsonia 27: 127—130 (2008) Red Duckweed (Lemna turionifera Landolt) new to Britain R. V. LANSDOWN* 45 The Bridle, Stroud, Glos. GLS5 4SQ ABSTRACT In May 2007, red duckweed (Lemna turionifera) was found in a ditch at Stoborough in Dorset. In July 2007, a second population was found in the South Forty-Foot Drain in Lincolnshire. L. turionifera is native in North America and northern Asia. It has also been recorded from Austria, Belgium, the Czech Republic, France, Germany, the Netherlands, Norway, Poland and Sweden where it has usually been considered to be an alien. Its occurrence in two very widely separated sites suggests that it may have been in Britain for some time but has been overlooked. Notes are provided to aid identification of L. turionifera, particularly in relation to L. minor and L. gibba and some information is provided on the habitats and species associated with it at its two British sites. KEYWORDS: Lemnaceae, distribution. INTRODUCTION In May 2007, John Bruinsma, Klaus van de Weyer and [| visited a number of sites in southern England to enable Klaus to prepare guidance on the identification of some aquatic plant taxa which occur or have occurred in Germany with which he was not familiar. One of these was Potamogeton x sudermanicus Hagstr. at its only British location near Stoborough in Dorset (vc. 9). Whilst searching for the pondweed, L. turionifera Landolt was found growing with L. gibba L. in populations embayed by a plank bridge and among marginal vegetation. On 19 July 2007 another population of L. turionifera was found in the course of a survey of aquatic vegetation in the South Forty-Foot Drain, between Billingborough Fen and Quadring Low Fen, Lincolnshire (vc. 53). Material from both populations was sent for determination by Professor Dr. E. Landolt at the Institut fiir Integrative Biologie, Ziirich. The material from Stoborough was too degraded for determination by the time it arrived but Prof. Landolt has been able to confirm that the material from the South Forty-Foot Drain was *e-mail: rlansdown @ardeola.demon.co.uk L. turionifera. The two populations were sufficiently similar that I am convinced that they are both of the same species and therefore the collection from Stoborough represents the first record of this species in Britain. Material from this collection has been deposited at BM. DISTRIBUTION L. turionifera is native in North America and northern Asia. In North America it occurs from Mexico north to Alaska and through much of Canada and the United States east to Nova Scotia, it is largely absent from the south- eastern United States. In Asia it occurs in a broad band from Turkey, north and east across Russia to Kamchatka and Sakhalin Island (Landolt 1986). In Eurasia, the western limits of its distribution have been obscured by apparently non-native populations (Muller 2004), where it has been recorded from Austria, Belgium (Hoste & Bruinsma 2007; van Landuyt 2007), the Czech Republic, France (Muller 2004), Germany (Landolt 1986, Wolff and Ortschiedt 1993), Poland (Muller 2004), the Netherlands (Wolff & Bruinsma 2005), Sweden and Norway (GBIF 2007). Its occurrence in Britain may be part of a west- ward range expansion. IDENTIFICATION A total of seven species of Lemnaceae are known to occur in the wild in Britain. In addition to species of Lemna, these include Spirodela polyrhiza (L.) Schleiden and Wolffia arrhiza (L.) Horkel & Wimmer. Another species Landoltia_ (formerly Spirodela) punctata (G. Mey) Les & D. J. Crawford has been recorded as a contaminant in garden centres (Rumsey 2006), but not yet in the wild. The species of Lemna known to occur in the wild in Britain fall into three sections: Section Hydrophylla most clearly distinguished by submerged growth and the fronds tapering to a stalk at the base, represented by L. trisulca L.; 128 R. V. LANSDOWN Section Uninerves, best distinguished by having only one nerve, represented by L. minuta Kunth.; and Section Lemna best distinguished by having floating growth and more than one nerve, represented by L. gibba, L. minor and L. turionifera (Landolt 1986). L. gibba can sometimes be identified by its growth form, Where the cells of the aerenchyma expand so that the whole frond becomes swollen and can be up to 4 mm thick. However, not only can L. gibba occur in populations where no fronds are swollen, but L. minor can show some expansion, although never as much as L. gibba (Landolt 1986). When not expanded, L. gibba and L. minor can usually be separated by the shape of fronds, although this is not always clear and some populations may need to be grown in cultivation until characteristic features are sufficiently well developed (E. Landolt, pers. comm. 2007). L. turionifera can usually be identified by the presence of papules along the median line, of which the apical is not significantly larger than the others and by reddish coloration comm- encing around the point at which the root is attached on the underside and developing as brownish-red discoloration of the bases of fronds on the upper side, making fronds look “dirty”. This latter feature is the best method of recognising plants of L. furionifera among mixed floating Lemna populations. Papules are small raised bumps which occur on the upper surface of fronds and are the same colour as the surrounding tissue. The following key provides guidance on identification of those species of Lemna known to occur in the wild in Britain. KEY I. Fronds denticulate at the apex; narrowed to a stalk-like portion at the base.......... L. trisulca Fronds entire and with at most only a minute stalk-like portion...................:ccccceceseeseeeeeees ig 2(1) At least some fronds greatly swollen to more than 2 mm, with a firm, whitish or translucent swelling, below the green upper Cells esr ne eeereee eee ee see L. gibba Fronds + flat, clearly bifacial and not or at most only slightly swollen ....................::ccccceee: 3 3(2) Fronds with reddish coloration ...22......c.chcccccse-co-25esc57 stesso soe ee ee 4 Fronds lacking reddish coloration, entirely ereen!or clawcous 222---.---eee eee 6 4(3) Reddish coloration developing initially at the margins then spreading inwards on both Surfaces or as scattered dots/over the upper Suiace:..:..-...9 pense eee L. gibba Reddish coloration either mainly on the upper surface -or developing initially at point of root attachment (mode) underneath and! spreading Outward \-2.-----ccce--ceseseesseee ee eee 5 5(4) Reddish coloration + restricted to the upper side of fronds or markedly less intense on the underside; papules near the tip and at the node larger than the ones in between ...... L. minor Reddish coloration either restricted to or more intense on the underside; papules of uniform SIZE s,iccstawate asians nace ai oaae docalsdianiasiie dete oes eae Reece 5 aE SEE SCE Ne L. turionifera 6(5) Fronds evidently glaucous (although this is very difficult to confirm in small or sparse populations); fronds small (<4 mm x 2:5 mm); vein solitary and indistinct............ L. minuta Fronds green, not glaucous; fronds small to large (<10 x 7 mm); veins usually more than one and usually evident as mdges alone the fromd 7 eon vi 7(6) Fronds with distinct papules along the median line of the upper surface, the papule at the tip not distinctly lareer than the proximal papulese.-.ecseceeeeee see L. turionifera Fronds without distinct papules or with papules near the tip distinctly larger than proximal ONES 00, .50-seselecinweaivndpeeadebace cds ducudleasadteens cee deascees denseenedenoeseehenseece StReee et ee 8 8(7) | Fronds widest toward the often broad apex; largest fronds with five veins all arising from the Saime POrint 222.5500 os eekecce case ses scneaes «desea cusses see aeseec oOo cake ese Ae ee L. gibba Fronds widest at the middle or toward the base, apex usually fairly narrow; largest fronds with 4—5 veins; with the outer veins arising fromthe imme ser.c.cessseaeeeeeeeeseeeee eee L. minor LEMNA TURIONIFERA 129 DISTRIBUTION AND HABITAT IN BRITAIN DORSET (V.C. 9) The population at Stoborough was found in a broad ditch, approximately 10 m wide, with a width of about 7m of open water. The margins were densely vegetated mainly with Glyceria maxima (Hartmann) O. Holumb. L. turionifera was mainly found in a mixed population with L. gibba embayed by railway sleeper crossing the ditch near its mid-point. Scattered plants were also found in areas of open water on the landward side of stands of G. maxima. Other plants recorded in the ditch included, Potamogeton x sudermanicus Hagst., P. acuti- folius Link and Lemna gibba; Lemna minuta occurs in a very well-developed population in a nearby ditch. The ditch where the L. turionifera was found is shown running just north of due west of Redcliffe Farm on a map published by Preston & Pearman (1998). LINCOLNSHIRE (V.C. 53) At the point where L. turionifera was found, the South Forty-Foot Drain is a wide, deep channel with very slow-flowing or still water running between high, steep embankments through fenland. The population in the South Forty-Foot Drain occurred with a very rich fenland aquatic plant community including Ceratophyllum demersum L., Chara vulgaris L., C. virgata Kitz., Hottonia palustris L., ennammciova iL. minor, LL. trisulca, Myriophyllum verticillatum L., Nymphoides peltata Kuntze, Potamogeton lucens L., LEAs meee. specinams \., P. trichoides Cham. & Schltdl., Ranunculus circinatus Sibth., Riccia fluitans L., Sagittaria sagittifolia L. and Tolypella prolifera (Desv.) Leonh. Landolt (1986) recognises’ only five phytosociological associations characterised by i unioniverae One ol theses Telemed) to as “Association of L. turionifera and Riccia fluitans” is reported to include L. turionifera, L. trisulca, Spirodella polyrhiza and Wolffia columbiana Karsten with occasional Lemna valdiviana Phil., Wolffia borealis (Engelm.) Landolt and Ceratophyllum demersum. Clearly many of these species are unlikely to occur in Britain however it would seem reasonable to attribute the population in the South Forty-Foot Drain to this association, because it shares Riccia fluitans, Lemna trisulca and Cerato- phyllum demersum. Muller (2004) suggests that in Europe L. turionifera is associated with still, eutrophic water with a pH of 7-9 and this is likely to agree with the conditions in both the British sites. The discovery of this species at two widely separated sites in Britain suggests that it may well be widely established and has simply been overlooked. Lemna species are not necessarily well recorded in Britain and it is hoped that information presented here will help to improve recording. ACKNOWLEDGMENTS I am deeply indebted to Prof. Dr. E. Landolt for the time and patience with which he cultivated Lemna material for determination. I am grateful to Fred Rumsey and Arthur Chater for reading through and commenting on earlier drafts of this text. I am very grateful to John Bruinsma and Klaus van de Weyer both for the pleasure of their company in the field and for showing me L. turionifera many years ago, I am also grateful to Gillan McCoy and Emily Greenall of Jacobs Aquatic for the pleasure of their company in the field. I would like to thank the Environment Agency for permission to publish data collected from the South Forty- foot Drain. REFERENCES G.B.1.F. (2007). Biodiversity occurrence data provided by S, UPS and O. HOSTE, I. & BRUINSMA, J. (2007). Na Noord-Frankrijk en Nederland: Lemna _turionifera nu ook in Belgié ontdekt. Dumortiera 91: 20-22. LANDOLT, E. (1986). Biosystematic investigations in the family of duckweeds (Lemnaceae), 2. The family of Lemnaceae — a monographic study. Volume 1. Veroffebtlichungen des Geobotanschen Instituts der Eidg. Techn. Hochschule, Stiftung Riibel, in Ziirich 71. MULLER, S. (Coord.) (2004). Plantes invasives en France. Muséum national d’ Histoire naturelle, Paris. PRESTON, C. D. & PEARMAN, D. A. (1998). Dandy & G. Taylor’s unpublished study of Potamogeton x sudermanicus Hagstr. in Britain, with an account of the current distribution of the hybrid. Watsonia 22: 163-172. RUMSEY, F. J. (2006). Spirodela punctata — the next invasive duckweed? Poster presented at the 2005 Exhibition Meeting of the Botanical Society of the British Isles. 130 R. V. LANSDOWN VAN LANDUYT, W. (2007). Herkenning van de vier in Belgié voorkomende drijvende Lemna-soorten, in: Dumortiera, 91: 16—20. WOLFF, P. & BRUINSMA, J. (2005). Lemna turionifera Landolt, Red Duckweed, new to the Netherlands. Gorteria 31(1): 18-26. WOLFF, P. & ORSCHIEDT, O. (1993). Lemna turionifera Landolt — eine neue Wasserlinse fiir Stiddeutschland, mit den Erstnachweisen fiir Europa. Carolinea 51: 9-26. (Accepted April 2008) Watsonia 27: 131—141 (2008) Fy John Fitz-Roberts: A little-known seventeenth century botanist VER ENCE OEEN Faraday Building, Department of Biological Sciences, University of Lancaster, Lancaster LAI 4YA* ABSTRACT Information is provided relating to the work and collections of the little-known botanist, John Fitz- Roberts of Kendal, who was active around the end of the seventeenth-century. Records, traceable from the literature of the day, preserved specimens, and letters sent to him by James Petiver, are discussed. Some of his records are cited as obscure polynomials and where possible these have been identified through reference to the early literature. Where localised, the current status of the recorded plant is_ briefly commented upon. Fitz-Roberts appears to have been one of the first, if not the first, to record Sesleria caerulea in Britain. KEYWORDS: John Robinson, Kendal, Nicolson, Petiver, Plukenet, Ray, Sesleria caerulea, Wilson. INTRODUCTION Details of the life of the late seventeenth Kendal botanist John Robinson alias Fitz- Roberts are largely shrouded in mystery. He appears to have been active well before the 1690s and also later but published nothing himself. One of his contemporaries was the well-known Cumbrian botanist and Quaker Thomas Lawson (1630-1691) who mentioned Fitz-Roberts in his correspondence. What little is known of Fitz-Roberts originates mainly from those who knew him or with whom he corresponded directly. These included the Cumbrian botanist, William Nicolson (1655— 1727) who was Archdeacon (later Bishop) of Carlisle, John Ray (1627-1705), James Petiver (1633/4-1718), and the Queen’s Botanist, Leonard Plukenet (1642-1706). Nevertheless, James Newton (1639-1718) and _ Richard Richardson (1663-1741), both of whom sometimes botanised in Westmorland, appear not to have been known to him. In more recent times there have been cursory references to him by Martindale (1888-1889), Dandy (1958), *email: m.foley @lancaster.ac.uk Desmond (1977) and Whittaker (1981, 1986). However, he is not mentioned by Raven (1947) in his review of the naturalists of around that period and has been completely overlooked by Flora writers covering the Kendal area, such as Baker (1885), Wilson (1938) and Halliday (1997). No reason has yet been found for his change of name. Although Nicolson, in his diary entry for June 1690, refers to him as “Joh [John] Robinson’, later he attributes two of his records to “J.F.R” (an abbreviation for John Fitz- Roberts). However, his name as Fitz-Roberts seems to have been established by the mid- 1690s as evidenced by letters written to him around then by Petiver and records quoted by Ray (1696 passim). Petiver, in Philosophical Transactions (1710-1712), confirms his name change when describing him as “Mr John Robinson, alias Fitz-Roberts, a Curious Naturalist, sent up several of these Plants [Osmunda Westmorlandica = Cryptogramma crispa| from Kendal”. Similarly, the relatively few surviving herbarium specimens he sent to Petiver (now in the Sloane herbarium, BM) are, except for one (a lichen), labelled in Petiver’s hand in the name of Fitz-Roberts, not Robinson; similarly, Lawson (prior to his death in 1691) in a letter to Ray had already referred to him as Fitz-Roberts. All this contrasts with some of Nicolson’s 1690 diary entries and it is possible he may have used both names for a while before Fitz-Roberts became finally accepted. His reason for the change may have been an attempt to avoid confusion with another contemporary Cumbrian naturalist, also a correspondent of Ray, Thomas Robinson, who was Rector of Ousby from 1672-1719 and author of Natural History of Westmorland and Cumberland published in 1709. So far, the dates of Fitz-Roberts’ birth and death are unsubstantiated, the Parish register for the Kendal area being lost around the time of his birth whilst that covering later years is 1132 VEE Ne EORENG still un-transcribed. However, a parish record confirms a marriage to have taken place in Kendal on 2 January 1682 between a Mr John Robinson and Anne King of Strickland Roger [Kendal Parish Register transcripts] which could refer to him. Prior to his death in 1691 Thomas Lawson had sent to Ray a drawing made by Fitz-Roberts of Andromeda polifolia (see below) which appears to place Fitz- Roberts as an experienced botanist by the 1680s, if not earlier. Nicolson in his diary for 1690 referred to him flatteringly as “Joh. Robinson (sutoris ultra crepidam docti)” again suggesting that he was well-established and well-known by that date. The botanist John Wilson (1696-1751), author of A Synopsis of British Plants, in Mr Ray’s Method (Wilson 1744), also knew Fitz-Roberts and probably gained much of his early botanical experience from him (see Gentianella amarella below). Known as “Black Jack”, Wilson was originally a Kendal shoe-maker, later becoming a baker and finally, a lecturing botanist and author as well as something of a local character (Nicholson 1832). The connection with Wilson suggests that Fitz-Roberts was botanising until at least the 1720s and probably later. If the recorded Kendal wedding of 1682 was indeed that of Fitz-Roberts, it would seem that his life span approximated to the years c.1655—c.1730. He was known to live at the Ghyll (Nicholson 1832) [SD50.92] which is just to the west of Kendal close to the Brigsteer and Levens roads and is ideally placed for access to the botanically-rich limestone countryside nearby. Although possibly not his actual home, Gill Cottage is shown at this location on a map published in 1859, with Gill Well only a field’s distance away. THE TOPOGRAPHY OF THE AREA AROUND FITZ-ROBERTS’ HOME John Fitz-Roberts appears to have been active mainly in the south of the old county of Westmorland (v.c. 69) and occasionally in “Lake Lancashire”. Both areas are now within administrative Cumbria and lie to the north of the estuary of the River Kent. Even today, the area within a 10 km radius of his home is botanically-rich and it would have been even richer at the end of the seventeenth century. This is especially the case to the west and south, since much of the substrate is limestone, the major features being the long, steep outcrops of Cunswick and Scout Scars and their associated open grassland. Lying below the scars is limestone woodland and the small Cunswick Tarn with its surrounding fen margin. Drained mossland separates this area from another major feature lying to the west, the limestone scar of Whitbarrow, whilst still further away to the west and south are the outcrops at Meathop, Kirkhead and Humphrey Head. By contrast, to the north and east of his home the substrate is largely acidic and so has a different, more restricted flora. Fitz-Roberts is known to have botanised especially on the nearby escarpments as well as on the lower, wetter mossland, much of it nowadays given over to agriculture. Other records of his for more montane plants suggest visits to nearby acidic Silurian areas such as the valleys of Kentmere and Sleddale to the north- west of Kendal. So far there is nothing to indicate that he travelled much further afield and, surprisingly, there are no localised records of his for the Whitbarrow area. Even today the latter is still botanically-rich and it is possible that access over the intervening, now-drained, moss land would have been difficult and involved a long detour round higher ground (the so-called “landway’). However, he did visit Humphrey Head at least once, so he probably used this route on that occasion. FITZ-ROBERTS’ BOTANICAL RECORDS A note in Nicolson’s diary states that on 10 June 1690, he received a specimen of Ophrys insectifera from Fitz-Roberts “missa est mihi a Do. Robinson de Kendal, planta nostras perelegans Orchis myodes”. In the following week Nicholson paid a visit to Kendal, perhaps to preach, and on 18 June visited Fitz-Roberts, noting in his garden a range of plants “horto Joh. Robinson (sutoris ultra crepidam docti) planta occurrebant sequentes Brittanicae”. These were described under their contemporary names, some of which are identifiable: Acetosa Rotundifolia West{morland] (Oxyria digyna), Acorus verus, Betonica aquatica (Scrophularia aquatica), Conyza major (Inula conyza), Conyza media, Cotyledon Hirsuta, Geranium nodosum, Geranium fuscum, Geraniu longius radicatii, Marrubium Aquaticum, Pentaphylla erectii fol. Argenteo (Potentilla argentea), Polygonatum (Polygonatum multiflorum or P. odoratum, both scarce locally), Ptarmica vulgaris (Achillea ptarmica), Serrantula foliis o dissectis (Serratula tinctoria) and Tapsus Barbatus (Verbascum thapsus). JOHN FITZ-ROBERTS 133 On the following day (19 June) Nicholson and Fitz-Roberts together with two others made a small local botanical expedition. They went through the woods by the River Kent to the waterfall at Levens (i.e. downstream from Kendal) noting fossils in the dam on the river and found the scarce Meum athamanticum in fields near Kendal Castle and Euonymus Theophrasti (Euonymus europaeus) plentiful in hedgerows around the villages of Natland and Sedgwick. [“Do. Sutch M.B. medicus Kendaliensis, A Farrington Affinis meus peerdilectis, Joh. Robinson antedictus, et W.N. Inter sylvas et fl. Cantij cataractus, aestivo calare obrutio? Occurrebant Lapides Entrochi in una massam concreti. In ye mill-damm, above ye Force near Levens.....Meum. In ye fields near Kendal-Castle.... and Euonymus Theophrasti. Exceedingly plentiful in ye Hedges about Natland and Sedgwick’’]. Working largely within the Kendal area, Fitz-Roberts was the first to report the presence of a number of interesting plants, specimens of some of which he had collected are now preserved in the Sloane herbarium (BM). Several of his records are also acknowledged in the works of Petiver and Ray. One, Sesleria caerulea, which, although classified as a nationally scarce plant (Halliday 1994), is exceptionally abundant on _ the northern limestones. It was first recorded by Fitz- Roberts (Ray 1696) as: “Gramen parvum montanum spica crassiore purpuro-coerulea [ORS Wlocaonas Ab amico optimo D. Fitz-Roberts accepi, qui alicubi in Cumberlandia collegit”’. This was a well-documented find since there is an excellent specimen (HS 150 f.118) in the Sloane herbarium (BM) collected by Fitz- Roberts (Fig. 1), Petiver’s label stating “Mr Fitz-Roberts gathered this elegant grass at Hols-Fell Nabb % a mile from Kendall”. Accompanying it is also the description taken from Leonard Plukenet’s Almagestum Botanicum (1696). Hols-Fell Nabb (Helsfell Nab) [SD5093] is only | km to the north of where Fitz-Roberts lived at the Ghyll and a little to the east of the long west-facing limestone escarpment of Cunswick Scar [SD4993-94] where Sesleria caerulea is still exceptionally abundant and Where other much scarcer plants are found today, such as Carex ericetorum and Helianthemum oelandicum. RECORDS IN CONTEMPORARY LITERATURE AND DIARIES ACTAEA SPICATA L.: This was recorded by William Nicolson in his manuscript as having been seen on the outskirts of present-day Kendal on 9 August 1691. The locality was Shaw Wood, Stainbank Green (SD5091) where it was described as abundant at that date and presumably in fruit. Whilst there is only circumstantial evidence that the record was originally Fitz-Roberts’, this wood, nowadays no longer present, was only a few hundred metres from his home at The Ghyll and the Actaea was recorded at a period when Nicolson was known to be visiting him (see above). It seems inconceivable that such an interesting plant had not first been found there by Fitz-Roberts who then later showed it to Nicolson. A. spicata has not been recorded for this general area of south Westmorland for some considerable time, the closest being at two sites near Arnside where it was last seen in 2002. At Shaw Wood, Nicolson (cf. manu- script) also recorded Filipendula vulgaris which again would surely have been previously known to Fitz-Roberts and again, presumably shown to Nicolson at the same time as the Actaea. No Fitz-Roberts specimen of either plant has been traced. ANDROMEDA POLIFOLIA L.: Quoted by Ray (1696: 202) as “Ledum palustre nostras Arbuti flore. Rosmarinus sylvestre minus nostras Park...... Marsh Cistus, or wild Rosemary ..... Flos, ut ex icone ejus ad vivum depicta a D. Joanne Fitz-Roberts; & a D. Tho. Lawson a me transmissa apparet.....”. This illustration referred to as drawn by Fitz-Roberts will have been of a plant collected from either Brigsteer Moss (near to Fitz-Roberts’ home) or one of the others close by which nowadays are mostly drained and “improved”. However, the plant is still present on some of the mosses within the area. CRYPTOGRAMMA CRISPA (L.) R. BR. EX HOOK:: Petiver (1710-1712: 376) stated “Osmunda Westmorlandica, foliis tenuissimme dissectis, Mus. Petiver. “Mr John Robinson, alias Fitz Roberts, a Curious Naturalist, sent up several of these plants from Kendal, which were set in Chelsea Garden and elsewhere, but after a 134 MSE Ya OREN winter or two, they often go off’. From the polynomial and illustrations quoted by Petiver: ie: Plukenet (6915 t3s 162) and! Monson [=Bobart] (1699. sect 14, t.4), the plant can be identified as C. crispa. This is still frequent on acidic scree and stone walls in the Kendal area, especially off the limestone to the north. GENTIANELLA AMARELLA (L.) BOERNER: Ray (1696: 156) recorded “Gentianella fugax verna seu praecox. Vernal dwarf Gentian. Found by Mr Fitz-Roberts on the Backside of Halse-fellnab [SD5093] near Kendal; as also in the Parks on the other side of Kendal, on the back of Birk-hag [SD5391]. Aprili mense floret ad Junium usque”. The first site is where Fitz- Roberts also collected Sesleria caerulea (see above), that at Birk Hagg, nowadays farmland, is a little way to the east of Kendal. Both populations appear to have been flowering much earlier than normal. At the time, Ray considered it to be a new species but Wilson, who also knew the plant from where Fitz- Roberts had seen it, stated “Mr Fitz-Roberts told me he never gave any intelligence about it, only that he informed Mr Petever [sic], he had seen the Autumnal Gentian (which grows very plentifully in those places) flower sooner than common; and this probably caused the mistake” (Wilson 1744: 135). G. amarella still occurs on grazed limestone grassland around Kendal. HELIANTHEMUM OELANDICUM (L.) DC. SUBSP. CANUM (L.) BONNIER: Again, given by Ray (1696: 203) as: “Chamaecistus seu Helianthemum alpinum folio Pilosellae minoris Fuchsij J.B.....Hoary dwarf Mountain-Cistus with Coltsfoot Leaves..... At Buck-Barrow Bank Scar, betwixt Brigsteer and Conswick; also on the Rocks about Cartmel-Wells in Lancashire _ plentifully. Observed by Mr Fitz Roberts”. A sheet in the Sloane herbarium (HS 152 f.277) on which an unidentified specimen bears Fitz-Roberts’ name also contains specimens of what appears to be this plant which are very likely to have originated from him. The first locality mentioned is the southern part of the cliff now known as Scout Scar (SD4890-SD4892), the long limestone outcrop lying just to the south of Cunswick Scar, where both Helianthemeum oelandicum (H. canum) and A. nummularium are frequent to this day. The second locality is the coastal limestone ridge of Humphrey Head (SD3973-SD3974) where the plant also survives together with other rarities. MELICA NUTANS L.: Ray (1696: 262) stated: “Gramen Aven. locustis rubris montanum C.B. locustis rubra J.B. Park. Mountain Oat-Grass with red husks. This was sent [to] Mr Petiver out of the north, and by him communicated to me”. Later in the same publication he (Ray 1696: 325) specifically mentioned that it was sent to him by Fitz-Roberts from near Kendal. Also that he had also received what he described as ‘most elegant’ specimens of the same plant from Jacob Breyne (1637-1697, the well-known Polish botanist): “Hoc a D. Fitz-Roberts circa Kendalium Westmorlandiae oppidum primarium collectum & ad me transmissum est, cujus etiam elegantissima specimena accepi a celebratissimo Jacobo Breynio”. Fitz-Roberts’ plant is identified through the reference to John Parkinson (Park.) given above. In his Theatrum Botanicum, Parkinson (1640) provided an illustration of this plant, the central example clearly depicting the rather one-sided inflorescence of M. nutans. In his edited and posthumously published “Concordia” of grasses etc. Petiver (1767) re-confirms Fitz- Roberts as the original collector: “Mr Fitz- Roberts first observed this, about Kendale in Westmoreland”. The plant is still quite frequent in light woodland over limestone on both sides of the Kent estuary. OXYRIA DIGYNA (L.) HILL: Petiver (1710-1712: 377) stated that Welsh Sorrell “Acetosa Cambrobritannica Montana” had been seen “by Mr Tho. Lawson and Mr John Fitz-Roberts, two other expert Botanists, in Westmorland...”. Amongst various synonyms given by Petiver, it can be identified from the illustrations in Parkinson (1640: 745, fig 12) and Morison (1680: vol.2 sect.5, t.28, fig.10), both clearly of Oxyria digyna. A plant of damp mountain rocks and gullies, it still occurs in the central Lake District area from where they most likely saw it. PULMONARIA OFFICINALIS L.: Nicolson, in his manuscript, recorded “Pulmonaria folis Echi, flore rubro. Bugloss- Cowslips; or long-leav’'d Sage of Jerusalem. Nigh Burnishead-Chapple [= Burneside, SD 50.95]. J.F.R. [Fitz-Roberts]”; this is also noted in Whittaker (1981). The plant had been grown in British gardens for centuries and became widely naturalised in scrub and woodland. Thomas Johnson (Gerard 1633) described and illustrated it, stating that it was first found by John Goodyer in the New Forest in 1620. Ww Nn JOHN FITZ-ROBERTS | £ CShbrtpi 3 prvertorr ug p3/ as fons hin etiachar ined CBye ‘lt $37. Pega = x Epps es Se Sch Se Blood SS 4 Be zi Ete}. ils ypir (iy ees 12.9,.92-6. Be {alban l s ilps Ta 405 I Horn. ats ctl. CCB OO. reat Wl PAT GAY. es ba! af! s fretle Cr€. aye = £. Y Loe ty : Buren Pxfl mark ¢lrtrhavid (ve ta ak rl = ‘ FIGURE 1. Fitz-Roberts’ specimens: Sesleria caerulea (top left); Pseudorchis albida (top right); Epipactis palustris (bottom left); the Fitz-Roberts-Bobart specimen of Thlaspi perfoliatum (bottom right). 136 ME Ne OREN TRAGOPOGON PORRIFOLIUS L.: A reference to the vegetable known as salsify is given by Ray (1696: 77): “Tragopogon purpureum in Anglia sponte provenire nobis affirmavit D. Fitz-Roberts”’ and had _ been introduced and become naturalised in Britain by this time. It seems that Fitz-Roberts considered it to be native contrary to Ray’s earlier opinion that it was naturalised; however, Ray was correct. No precise locality was recorded by Fitz-Roberts although there is a much earlier (pre-1597) record from near Whalley (v.c. 59) probably originally assignable to Thomas Hesketh (1560-1613) (or his mother, Lady Hesketh) which is given in Gerard’s “Herball’ (Gerard 1597). From Nicolson’s manuscript it 1s clear that he also knew the plant “in the fields about Carlile (Canisters VERONICA SPICATA L.: Nicolson, in his manuscript, also quoted a record for V. spicata at Humphrey Head which he attributed to Fitz-Roberts: “Veronica spicata. Quae vel — Major. On the rock above Cartmell-Well. J.F.R. [Fitz-Roberts]”. This is also noted in Whittaker (1981). Even in those days Cartmel Well (or Holy Well) was a well- known spring under the west-facing limestone cliff of Humphrey Head and was reputed to have medicinal healing qualities. The spring and V. spicata are still there, the latter surviving as a small scattered population at several points along the cliff along with other local rarities. Fitz-Roberts appears to have been the first to record V. spicata in northern England. It can also be found a little to the east near Arnside (v.c. 69) in two very small populations. FITZ-ROBERTS’ HERBARIUM SPECIMENS Further contributions made by Fitz-Roberts to the flora of his local area are represented by the small series of specimens which he sent to Petiver. These are usually mounted amongst others of different provenance on mixed sheets, randomly distributed in three folio volumes in Petiver’s herbarium (Sloane, BM). Due partly to their age, several are poorly preserved and often difficult to identify. Petiver signified the Fitz-Roberts origin by placing small hand- written labels over, or adjacent to them, variously worded “Fitz-Roberts”, “Fitz- Robarts”, “Fitz-R”, etc. Unfortunately, not all the labels and specimens are directly attached to each other so that in some instances there is doubt about a genuine Fitz-Roberts connection and most of these have been excluded here. Altogether, seventeen specimens attributable to Fitz-Roberts are briefly described below. They are amongst some of the oldest surviving within the British Flora and in some cases may represent the earliest for Westmorland since, despite Lawson sending material to Bobart at Oxford and to Ray about this time, not all of these will have survived. On some of the sheets, Petiver has written “R.H.” followed by a number, this a reference to what he considered to be that particular plant on that page in Ray’s Historia (Ray 1686-1688). The specimen references given below (as HS) are those of the volume and page number in Sloane (BM). CYPRIPEDIUM CALCEOLUS L.: This is represented by specimen(s) at HS 152 f. 171. Long since extinct in the wild in the Kendal area and virtually so in Britain, the most recent records for this area are from the nineteenth century. These are Whitbarrow, where it was recorded by F. Clowes (Martineau (1855), “The Lyth” (which may be the same locality, since Whitbarrow lies immediately to the west of the Lyth valley) known from a fir plantation in the 1860s (Anon, BM!), and Barrowfield [Wood], Scout Scar, a specimen from there having been transplanted into the garden at Levens Halll (i8i5yvhenb sear Linton, BM!). This last locality, close to Fitz- Roberts’ home is the most likely source for his specimens. Unfortunately, one of them (lower left on the sheet) lacks a flower. Plants from the Kendal area were also transplanted into the grounds of nearby Sizergh Castle. EPIPACTIS PALUSTRIS (L.) CRANTZ: This sheet (HS 152 f.174) comprises three specimens (Fig. 1) with the lower left having the inflorescence broken off. The others (both labelled “Fitz-Roberts” by Petiver) have been referred to E. palustris and would appear to comply. It was described by Wilson (1744) as being “In several marshy places about Kendal” but the locality from where Fitz-Roberts collected it would probably be the fen margins of Cunswick Tarn (SD4893) where it was later seen by Gough (Nicholson 1832) and thought to be last recorded there in 1952 (Halliday JOHN FITZ-ROBERTS 37) 1997). There are also records at two other nearby sites where Fitz-Roberts may have collected it: on Whitbarrow, known to W. Foggitt (Wilson 1938), and in Barrowfield wood [SD4891—SD4892] in 1908 (about 2 km to the south of Cunswick Tarn) (Martindale, KDL!). Today, the plant appears to be extinct in this general area (Halliday 1997). Two other species of Epipactis: E. helleborine and E. atrorubens occur on the wooded limestones of Cunswick and Scout Scars, and may well have been known to Fitz-Roberts. ERIOPHORUM VAGINATUM L.: A plant of boggy ground on moorland and mountains, frequent in areas to the north of Kendal. However, it is also known from coastal mosses around the Kent from where Fitz- Roberts may have collected it (HS 150 f.183). LOBELIA DORTMANNA L.: Frequent at the shallow stony borders of the more acidic lakes and tarns away from the limestone areas and still fairly common to this day. Fitz-Roberts closest site would probably have been to the west of Kendal towards Windermere. This specimen is on sheet HS 152 3. PERSICARIA VIVIPARA (L.) RONSE: Fitz-Roberts’ specimen is at HS 151 f.28. This plant is currently found in hill pastures and mountain areas of Cumbria where it can be locally frequent. His closest locality was probably to the north on the limestone near Orton, a distance of c. 25 km from his home and somewhat outside his normal ambit. PSEUDORCHIS ALBIDA (L.) A. & D. LOEVE: Now very scarce in northern England this occurs only in upland pastures, usually off the limestone. However, there is an old record by Gough (Nicholson 1832) for Barrowfield Wood near Kendal. Although limestone woodland is a most unusual locality for P. albida, the site is only 2 km from Fitz-Roberts’ home and is a place he would certainly have visited and from where he would surely have collected a specimen if he had found it. His Specimen (Fig. 1) is on sheet HS 152 f.170. Whilst it is clearly P. albida (Fig. 1) there is further confusion in that Petiver had labelled it “An Nidus Avis...a D. Fitz Roberts collecta”. The area close to Barrowfield Wood is where Neottia nidus-avis is still present to this day but P. albida 1s unknown. Nowadays not present in the Kendal area, P. albida was collected at nearby Spy Crag, Staveley (Martindale & Martindale, KDL!) towards the end of the nineteenth century and also at Bannerrigg, Windermere (1886, Martindale (KDL!). SAXIFRAGA AIZOIDES L..: hesspecimentmSnonmshecta tis. Js st Ol and labelled “Kendall Fitz-R-” and “Sedum montant luted minus nostras. R.C: An Sedu illecebra folijs oblongis. Mer. Pin. [Merrett’s “Pinax” (Merrett 1666)|”. Nowadays, the plant is frequent on wet rocks and flushes in the Lake District hills. The likeliest locality for Fitz-Roberts to have collected it would have been either to the north of Staveley or in Longsleddale, both places near Kendal and where it still occurs. At the latter site it was described as being “in the greatest plenty” by William Curtis in 1782 (Curtis 1855-1856: 111). As indicated on Fitz-Roberts’ label (see above) it had been recorded some years earlier by both Ray and by Merrett from Ingle- borough, both these records almost certainly attributable to Thomas Willisell. Lawson also sent a Westmorland specimen to Bobart at Oxford (Morison 1699) probably predating that of Fitz-Roberts. [Another plant of a similar habitat 1s Sedum villosum. This also has extant localities in the area north of Staveley, and it was from nearby Sleddale that Thomas Lawson recorded it in the late seventeenth century. Surprisingly, Fitz-Roberts doesn’t appear to have known it]. SESLERIA CAERULEA (L.) ARD.: This well-preserved specimen (Fig. 1) (HS 150 f.118) has already been discussed in some detail above. It is one of the few for which a precise locality 1s given. THLASPI PERFOLIATUM L..: The significance of this specimen (Fig. 1) in relation to Fitz-Roberts is uncertain but a label on the sheet suggests that it might at one time have come into his possession. A directly attached label states it to have been found by Jacob Bobart among stone pits between Whitney and Burford in Oxfordshire, an area towards the centre of its very restricted British range. Fitz- Roberts seems unlikely to have visited there himself but may have received the specimen from Jacob Bobart (the younger) who by then would have been in charge of the Oxford Botanic Garden. How it came into Petiver’s possession is uncertain but the reference to Fitz-Roberts must have some significance. 138 Mie IRONY Other specimens in Petiver’s herbarium were sent to him by Fitz-Roberts but, because of their inadequacy or poor quality, they are almost impossible to identify with certainty. These are: ALCHEMILLA SP.: A poor fragmentary specimen (HS152 f.255), again difficult to identify, may be the ubiquitous A. xanthochlora. On the other hand Fitz-Roberts would have had easy access to other members of the genus and so by sending the specimen to Petiver, he may have regarded it as something more unusual. COCHLEARIA SP.: Well-labelled but poorly preserved, this specimen (HS 152 f.27) appears to represent the upper part of an inflorescence of a Cochlearia but, in the absence of the lower leaves, it is not possible to identify it further. Nowadays, the mainly coastal C. anglica, C. danica and C. officinalis are all known fairly close to Fitz-Roberts’ home area whilst C. pyrenaica occurs inland on the nearby hills. GERANIUM SP.: Fitz-Roberts’ specimen (HS 32 f.75) is towards the bottom left side of the sheet and is poorly preserved. It may be G. pratense or G. sylvaticum, both frequent in the meadows and woods of his local area. Alternatively, it may have been a non-native Geranium from his well-stocked garden. NYMPHAEA L. CF. ALBA: This is represented (HS152 f.182) by only a single leaf and fruit of what is probably N. alba, frequent in lakes, ponds and backwaters thereabouts. It was recorded at Cunswick Tarn, close to Fitz-Roberts’ home by Gough in the early 1800s (Nicholson 1832) and is probably still there. SAXIFRAGA OPPOSITIFOLIA L.: A specimen of this is on the same mixed sheet (HS 152 f.101) as the clearly labelled Fitz- Roberts’ specimen of Saxifraga aizoides. However, another label, “Kendal FitzR”’, is so placed as to make it uncertain whether it refers to another nearby undetermined specimen or to that of S. oppositifolia (or even is a second label for the S. aizoides). If S. oppositifolia was intended, the closest localities for Fitz-Roberts would have been in mountains of the central Lake District or on the Craven limestones. VACCINIUM VITIS-IDAEA L.: This specimen (HS 152 f.233) is merely a small fragment but is probably of this plant. Nowadays it is largely upland but in the past was frequent on the lowland mosses around the Kent estuary and still survives at Meathop Moss (Halliday 1997). There is also a specimen of another eric- aceous plant on the same sheet as the Helianthemum oelandicum (discussed above, HS 152 f.277). Unfortunately, it comprises only a small twiglet with few leaves and no flowers and is very difficult to identify. V. oxycoccos would have been frequent in the damp heathy ground west of Kendal and Arctostaphylos uva-ursi present further into the central Lake District hills. It is possible that one of these is represented. Finally, there is a specimen (HS 150 f.66) of a poorly-preserved lichen. The label, difficult to transcribe, appears to state “Lichen cinereus [indeciph] subterus/subtius nigricans Nob. Kendall. J. Robinson”. It is allotted the reference number R.H. 117 relating to the “Lichen terrestris cinereus. Ash-coloured ground Liverwort” of Ray (1686-1688). This is the only specimen traced where Petiver uses the name Robinson rather than Fitz-Roberts. The lichen appears to be a species of Parmotrema, possibly the fairly widespread P. perlatum (Huds.) Choisy (Parmelia perlata (Huds.) Ach.). One other Fitz-Roberts specimen, so far untraced, is in OXF. LETTERS FROM JAMES PETIVER TO FITZ-ROBERTS Other than the records shown above, there is little else to throw light onto Fitz-Roberts as either a person or as a botanist. Fortunately however, there are two surviving copies of letters to him written by Petiver which provide at least a modest insight. These are preserved in the Sloane manuscripts section of the British Library in a bound volume (ms. 3332) and are randomly distributed amongst a large number of other miscellaneous correspondence between Petiver and contemporary naturalists. Both letters are difficult to transcribe in places but Petiver’s admiration for Fitz-Roberts expertise is clear to see. In one, undated but written in the Spring of the early 1690s (Appendix 1), he refers to Fitz- Roberts’ “moss cropping” [plant collecting] ability. (This term seems to have had local JOHN FITZ-ROBERTS 139 origins since, Thomas Lawson, in a letter to John Ray, stated: “After it [Eriophorum vaginatum] turns white sheep are greedy after it; so it is called Mboss-crops about... [Westmorland]” (Lankester 1848). Petiver’s letter suggests that Fitz-Roberts was _ well- known to members of Petiver’s botanical circle. This was especially so to Petiver’s friend, the eminent Samuel Doody (1656- 1706), curator of the Chelsea Physic Garden, who wanted Fitz-Roberts to collect for him. Although, Fitz-Roberts appears to rarely have travelled far outside his native Westmorland, Petiver’s comment “I should be glad to hear that we might suddenly expect to see you in London.....” suggests he might have visited the capital on at least one occasion. The second letter is dated 17 September 1695 (Appendix 2) and opens rather quaintly with Petiver saying that he had sent (via a relation of a friend) a 12 penny token, and that this person would visit him and drink with him. A comment about the imminent publication of the second edition of Ray’s “Synopsis” follows and then a further request is made for Fitz-Roberts to collect plant specimens. Petiver also asks for those of insects and suggests how they should be collected and preserved; he also encourages him to recruit others for the same purpose. In the postscript he reminds him of the shells and insects which he had sent “as a pattern for you to instruct others by, not doubting but you will very much add to them having far more opportunities than we have” (i.e. Petiver living in London). This is the only instance traced which suggests that Fitz-Roberts was in any way an ento- mologist but in those day anyone interested in botany would often be an all-round naturalist as well. Petiver also states that his “first century of Plants Insects Shells is not yet quite published” but nevertheless sends him uncorrected proofs. This is the first pamphlet in a series of ten “centuries”, the first of which he published towards the end of that year (1695) in his Musei Petiveriana (Petiver 1695-1703). In the letter, Petiver indicates that a second one is to follow shortly and that he is willing to acknowledge any specimens Fitz-Roberts might send. However, an examination of these which were also posthumously re-published in Opera Historiam naturalem spectantia (Petiver 1767) provide nothing of further significance in relation to Fitz- Roberts. The reason why he didn’t take up Petiver’s offer is strange since they were apparently in close contact at this time. There is no doubt more to be learnt about John Fitz-Roberts (alias Robinson) of Kendal but unless relevant, so-far undiscovered, manuscripts or letters are found, a further search of the old herbaria may provide the only opportunity to gain additional information about him. Unfortunately though, in the late seventeenth century, herbarium specimens were rarely accurately labelled as to collector and locality so that whilst Petiver, Plukenet and Ray were quite ready in their published works to acknowledge Fitz-Roberts’ contributions, they were less meticulous (Petiver being the exception) in noting the same in their private herbaria. It is quite likely that additional Fitz- Roberts’ specimens are hidden anonymously in the herbaria of Doody, Ray and others. This situation also arose with Thomas Willisel, probably the most prolific plant collector of the early period, who provided a vast number of records and specimens for Christopher Merrett’s “Pinax” (Merrett 1666; 1667). Merrett merely gave Willisell a cursory mention in his Introduction and abjectly failed to acknowledge any of the records for the many plants he had provided. Again, Willisell’s specimens similarly lie unattributed and dispersed amidst Merrett’s large personal collection (BM). John Fitz-Roberts emerges as a_ highly respected and competent field botanist, held in high esteem by his colleagues. Mainly specialising in his local area of south Westmorland, it is very likely that he showed many interesting plants to Thomas Lawson, and that some of these, as well as others recorded by William Nicolson, were his original finds. Herbarium acronyms given are as indicated by Holmgren ef al. (1990); all the specimens originating from Fitz-Roberts (Sloane, BM) have been seen; the symbol (!) indicates that herbarium specimens from other sources have also been examined. ACKNOWLEDGMENTS I thank the herbarium curators V. Papworth (BM) and C. Davies (KDL) for their help with the examination of specimens and B. Coppins, J, ee IML SS. leone gmc! I (Ce Gy Iino itee discussions relating to some of the plants mentioned. I am also most grateful to the authorities at the British Library for allowing me access to James Petiver’s letters, to the library staff at the Royal Botanic Garden, Edinburgh for permitting access to the pre- Linnaen literature, and to the custodian of the Rare Books Archive, University of Lancaster. 140 MAE Ye EOREY. REFERENCES BAKER, J. G. (1885). A Flora of the English Lake District. George Bell & Son, London. CuRTIS, W. (1855-1856). A Catalogue of certain Plants growing wild, chiefly in the environs of Settle, in Yorkshire. The Phytologist, New Series, vol. 1. DANDY, J. E. ed. (1958). The Sloane Herbarium: an annotated list of the Horti sicci composing it; with biographical details of the principal contributors. British Museum. London. 195-196. DESMOND, R. (1977). Dictionary of British and Irish Botanists and Horticulturalists. Taylor & Francis Ltd., London. GERARD, J. (1597). The herball or generall historie of Plantes. John Norton, London. GERARD, J. (1633). The herball or generall historie of Plantes...... enlarged by Thomas Johnson. London. HALLIDAY, G. (1997). A Flora of Cumbria. University of Lancaster, Lancaster. HALLIDAY, G. (1994). Sesleria caerulea, in A. STEWART, D. A. PEARMAN & C. D. PRESTON eds. Scarce Plants in Britain. JNCC. Peterborough. HOLMGREN, P. K., HOLMGREN, N. H. & BARNETT, L. C. (1990). Index Herbariorum. Part 1: The herbaria of the World. Bronx, New York. LANKESTER, E. ed. (1848). The correspondence of John Ray: consisting of selections from the philosophical letters published by Dr. Derham, and original letters of John Ray in the collection of the British Museum, p.204. Ray Society, London. MARTINDALE, J. A. (1888-1889). Early Westmorland Plant Records, Part 1, Pre-Linnaean Period 1597-1744. The Westmorland Natural History Record, a Quarterly Magazine, Vol. 1. MERRETT, C. (1666). Pinax rerum naturalium britannicarum. London. MorISON, R. (1680). Plantarum Historie Universalis Oxoniensis pars secunda (--tertia), seu Herbarum distributio nova, per tabulas cognationis & affinitatis ex libro nature observata & detecta. Oxford. MOoRISON, R (1699). Plantarum Historie Universalis Oxoniensis, seu herbarum distributio nova ... Pars tertia ... a J. Bobart. Oxford. NICHOLSON, C. (1832). The Annals of Kendal. Kendal. PARKINSON, J. (1640). Theatrum Botanicum. London. PETIVER, J. (1695-1703). Musei Petiveriani centuria prima [-decima]: rariora Naturae continens. London. PETIVER, J. (1710-1712). An Account of Diverse Rare Plants, Lately Observed in Several Curious Gardens about London, and Particularly the Company of Apothecaries Physick Garden at Chelsey. Philosophical Transactions 27: 376-377. PETIVER, J. (1767). Opera Historiam naturalem spectantia, vol.2. J. Millan, London. PLUKENET, L. (1691). Leonardi Plukenetij Phytographia, sive, Stirpium illustriorum, & minus cognitarum icones microform. Londini. PLUKENET, L. (1696). Almagestum botanicum, sive, Phtyographiae......, NW, RAVEN, C. E. (1947). English Naturalists from Neckam to Ray. Cambridge. RAY, J. (1686-1688). Historia Plantarum, species hactenus editas aliasque insuper multas noviter inventas et descriptas complectens, &c. Londini. RAy, J. (1696). Synopsis methodical Stirpium Britannicarum....., Edition 2. Londini. WHITTAKER, E. J. (1981). A Seventeenth Century Flora of Cumbria. Northumberland Press Ltd., Gateshead. WHITTAKER, E. J. (1986). Thomas Lawson North Country Botanist, Quaker and Schoolmaster. Sessions Book Trust, York. , WILSON, A. (1938). The Flora of Westmorland. T. Buncle & Co., Arbroath. WILSON, J. (1744). A Synopsis of British Plants, in Mr Ray’s Method. John Gooding, Newcastle-upon-Tyne. (Accepted October 2007) JOHN FITZ-ROBERTS 14] APPENDIX 1 LETTER PETIVER TO FITZ-ROBERTS (UNDATED) (SLOANE MS 3332, F.27) “Mr Fitz-Roberts, Being now entered upon the New year and find the verdant spring trying to show its early product, this is therefore to put you in mind of moss cropping as also to collect fair quantity to supply....Moss cropper you have formerly hinted to me ye have observed near ye in prime abt [about] this time I mean ye Schonoglaurus. Dear Friend Our Botanical Society have you in frequent remembrance but especially my own singular friend Mr Doody who particularly desiresst@ have his ...... hundred to ye and he beggs ye would be mindfull of him in relation to wi roots ye could procure of ye choicest plants you have in ye Parts. I should be glad to hear that we might suddenly expect to see you in London if we cannot be so happy in ye interim let me enjoy a letter. All correspondence met will be very acceptable to him who is sincerely your Friend. J Retiver APPENDIX 2 LETTER PETIVER TO FITZ-ROBERTS (LOND., SEPT 17, 1695) (SLOANE MS 3332, F.171V/F.172) “Mr Fitz-Robarts, My good ffriend. I take this opportunity of letting you know that I have you, which let this 12 penny token to follow, which I have sent by a relation of a friend of mine to drink with you, as also that I have taken care to .... ye new Edition of Mr Ray’s Synopsis which will now be very suddenly published. I hope ..... this you have got plenty of specimens of ye Adiantum floridum prolongans ye Polium and some other plants that I formerly gave you a catalogue of, which are rare with us. As also a collection of Insects formed glorious etc. Pray what plants you have by you as ye first return or if you prefer to give ye to this Gent he will send them to me or his Bro: which is all and I will gladly defray ye charges. I am very zealous in ye discovery of Insects which I desire yr assistance in and begg that will gather or collect so most of them shall come in your way or you can from others procure, I would have you in the Spring oblige some therefore to collect all ye Butterflies, Moths, Beetles and Boogs etc that he can catch, which he may stick through the back with a Pin having always a Pincushion in his pocket, for that purpose and those when caught he may stick within ye crown of his hat till he comes home, or into a small Pill box which he may carry in his Pocket. My first century of Plants Insects Shells is not yet quite published, but too let you see it is a doing I have inclosed sent you an uncorrected part of it. I do design to print another century before Christmas and would gladly have some communications from you insert therein and therefore pray fail not of something though never so small, which shall be justly acknowledged by kind friend... Yr most humble servant James Petiver. P.S. I have sent you of Shells and Insects 3 or 4 of each as a pattern for you to instruct others by, not doubting but you will very much add to them having far more opportunities than we have... vale. N.B. The Shells and Insects are Numbred according to my Century to wch I sd for you”. ah faa) - ¥., ; oh pau j mi i i , 7 € oy ‘ an i i z - = vA ‘ 1 Watsonia 27: 143-148 (2008) 143 The status of Coral-necklace Illecebrum verticillatum L. (Caryophyllaceae) in Great Britain D. A. PEARMAN* Algiers, Feock, Truro, Cornwall, TR3 6RA ABSTRACT The native range of Jllecebrum verticillatum in England is re-considered in the light of newly available historical records and nineteenth and twentieth century forestry and recreational activities. It is concluded that is probably a nineteenth century arrival outside of Cornwall, but that it is now a benign and intimate part of the flora _ that conservationists want to preserve. Milder winters seem to aiding the spread and survival of new populations. Nonetheless if only the Cornish populations were to be taken into account, then it would have a much higher Threat Status and thus attract more funding. KEYWORDS Native and alien ranges; conservation criteria; forestry and recreation; climate. DESCRIPTION, HABITATS, ECOLOGY AND PHYTOGEOGRAPHY Illecebrum verticillatum is a small glabrous annual with reddish stems which elongates to form a many-stranded plant over summer. It is often scarcely visible before late July, flowers late in the season and is usually killed by frosts. However, well-grown plants, presumably belonging to the previous year, have been noticed in West Cornwall and the New Forest as late as January; C. Chatters (pers. comm.) on a visit in late January 2008, noticed much material including plants in active growth in a New Forest pond kept open by mechanical cutting, arson and mountain bikes. He found. The submerged material, in about 20 cm of water, had infrequent green growing points from the apical tip of the last season’s ‘branches’. The terrestrial material, on a slight slope above the pond’s current water-level, in a zone that might be inundated in extreme circumstances, was much greener with well developed leaves and fresh growth, with the base of stems and the central root quite woody. The plants had weathered the hard frosts of *e-mail: dpearman4 @aol.com early winter with the terrestrial plants in stronger growth than those buffered from the hardest frosts by being submerged. None had the appearance of recently germinated plants Over-wintering to bloom the following year. Another West Cornwall site visited by DAP in mid-February 2008, revealed good growth with green, not red, stalks. It grows in or by pools or on seasonally wet sandy, or gravelly, tracks (including rides in forestry plantations) and heathy lawns on acid soils. It also occurs on railway clinker, possibly spreading from nearby forestry rides, and as a casual elsewhere. It spreads by seed though it has been known to root from stem fragments. In West Cornwall associates in 2005 included Hydrocotyle vulgaris, Hypericum _ elodes, Isolepis setacea, Juncus bulbosus, Lythrum portula, Myosotis secunda, Potamogeton polygonifolius, and Ranunculus flammula. On Bodmin Moor associates in 2007 included Hypericum elodes, Juncus articulatus, J. bulbosus, Lythrum portula, Myosotis secunda, Potamogeton polygonifolius, and Ranunculus flammula. Murphy (1994) gives _ track associates as Cicendia filiformis, Gnaphalium uliginosum and Radiola linoides whilst for the New Forest pools she lists Galium constrictum, Hydrocotyle vulgaris and Littorella uniflora, with rarer Anagallis minima and Cicendia filiformis. Illecebrum_ verticillatum is a_suboceanic southern-temperate species which appears to be widespread in N.W. Europe, from Poland to southern Spain, with very isolated scattered outliers in Italy, Greece and the Azores (Jalas & Suominen 1976). Hulten & Fries (1986) show many additional outliers, together with a large part of N.W. Africa. Little is known as to its current status in Europe; the Atlas for West Germany (Haupler & Schonfelder 1988) shows it to have declined severely in that area. The first record in the British Isles was made by John Ray on a trip to Cornwall in 1662, 144 D. A. PEARMAN possibly near to St Columb near Newquay; there have been many other records from that county during the next two centuries. It was not until 1846 that it was found elsewhere (in Kent), with later records from Berkshire (1891), Hampshire (c. 1920) and Dorset (1981). In the British Isles most recent floras and atlases have treated it as a native plant in Cornwall, Hants and Berkshire and as an alien elsewhere, though with no rationale for those decisions. HISTORY IN BRITAIN CORNWALL (V.CC 1 & 2) Rose Murphy (pers. comm.) writes that “old and recent records of /llecebrum verticillatum appear to be concentrated on the granite areas of Cornwall, or on the metamorphic aureoles that surround the granite where most of the mining and tin-streaming (and disturbance) took place”. Davey (1909) gives about 40 sites, dating back to thes ayes, uinste necord Ol woo, describing it as widely distributed but rather rare; Thurston & Vigurs (1922) add another six sites. There are currently records from 63 tetrads, past and present (Colin French pers. comm.) Not until the flora of Margetts & David (1981) is there any mention of a decline, where the caption merely states that ‘this very local plant has declined in the last fifty years for reasons that are not easy to find, as there are still plenty of suitable habitats’. Whatever their reservations, it is clear that lack of manage- ment, with no grazing and no disturbance, led to a massive decline in both IJllecebrum verticillatum and in its associated rare species such as Anagallis minima and Radiola linoides. Stewart et al. (1994) gave recent records from seven 10 km squares in Cornwall. The latest Cornish Flora, French et al. (1999) correctly attributed the serious decline to loss or change of habitat, but still showed recent records from five 10 km squares. The situation in 2008 is that is known in West Cornwall (v.c. 1) from three small sites in one 10 km square (perhaps around twenty plants in all) and in East Cornwall (v.c. 2) as less than five plants in one site and good populations on Bodmin Moor — a total of three 10 km squares and five tetrads for the county. Further details of the decline there will be covered in a forthcoming article (Pearman, in press). OTHER COUNTIES SOUTH HAMPSHIRE (V.C. 11) Illecebrum verticillatum was first recorded in v.c. 11 at Titchfield Haven in 1920 (Rayner 1929), across the Solent Estuary from its current stronghold in the New Forest. There are no further details of this record. Rayner gives the first record from the New Forest ‘young spruce plantation, Wilverley Inclosure, in great abundance, 1925’ and describes its gradual disappearance there over the next three years as it was displaced by native grasses. He relates the experience of a Mr Hall, who grew it in cultivation, and noted its ability to very quickly cover cleared ground, but that it was unable to maintain its hold in competition with other plants. Rayner also reports that a specimen, labelled as from this locality (presumably he meant Wilverley) was brought anonymously to the Southampton Wild Flower Exhibition in 1920. Brewis ef al. (1996) noted that it had steadily spread over some 50 years, with records there from some 30 tetrads. There was no comment in this work on its likely origin, other than a comment on the original Wilverley record, where it was noted as “perhaps introduced with imported conifers’. It is not clear if the increase in records is the result of genuine spread or due to better recording. Factors leading to a rapid spread might include heavy grazing pressures (though precise figures are hard to interpret, cattle numbers are higher than in the past), recreation (lots of riders), together with the disturbances associated with gravel extraction; there has also been much forestry work. The newly- established -Forestry Commission took over responsibility for the management of the New Forest in 1923. Much conversion of broad- leaved woodland to conifers took place from that date, though some had already occurred following the New Forest Deer Removal Act of 1851 (Tubbs 1986). Work on forestry is still ongoing, including removal of trees planted as late as the 1970s. All these activities could have physically spread the plant from site to site and created suitable, often seasonally flooded, open ground. Warmer winters and different rainfall patterns may also have aided its spread. Clive Chatters (conservation officer for the Hampshire Wildlife Trust and the new chairman of the New Forest National Park) and the current V.C. Recorder, Martin Rand (pers. ILLECEBRUM VERTICILLATUM 145 comm.) have pointed out that many of the older and the currently known sites for //lecebrum are at well-known botanising locations. It is therefore unlikely that it was overlooked although under-recording in the past certainly accounts for the difficulty in tracking the history of its spread and the dynamics of individual populations. Furthermore, they feel it is still spreading, though not just within the core of the Forest itself. Rand adds that it behaves as a coloniser and he cannot think of an instance where it occurs in a_ well documented and localised site that has not been recently disturbed. He also drew attention to a comment on a herbarium specimen in BM, collected at Shave Green Enclosure, Cadnam, 1930, Mrs E. Rothwell, “so abundant that a cartload was taken away and destroyed”. Illecebrum verticillatum 1s so _ often associated with other native flora characteristic of these temporary, disturbed sites that it would be difficult for a stranger to the New Forest to judge that it was other than native there. It grows around ponds with a suite of plants including Pulicaria vulgaris (first recorded 1835), Mentha pulegium (1839), Galium constrictum (1924), Ludwigia palustris (1645), Gnaphalium uliginosum, and Lythrum portula (Tubbs 1986). Note that these dates are for the first records for the whole of Hampshire, rather than from the New Forest — all, with the exception of Galium constrictum, were first recorded from v.c. 12, North Hampshire. It seems that botanical exploration for the New Forest was quite late compared to many parts of Britian (see Rand (2005) for pertinent comments on Gladiolus illyricus — another late discovery). There are also many more records and sites known for Galium constrictum and Ludwigia palustris over the last 40 years, though Chatters (1999) suggests that this may be due to greater recording effort starting in the 1950s and 1960s rather than any general increase in the numbers of sites or populations. Galium constrictum, though it is a critical plant, was also not discovered in Hampshire until the 1920s when it was recorded (as G. debile) from Lyndhurst and Hatchet Pond (Druce 1924). There was a similar increase in records over the next half-century. Druce makes rather a meal of its native status, concluding ‘the European distribution is not antagonistic to debile being a native of the Channel Isles and south-western England.’ The map in Meusel & Jager (1992) only tenuously Supports this view, showing a core distribution around the Mediterranean with a narrow extension north through France. In Britain it is almost confined to the New Forest. There was a relatively short-lived occurrence in Devon, at a site subsequently lost to ball-clay extraction and two unusual records from South Wiltshire (v.c. 8), where it was recorded in 1984 in water-meadows at North Charlton (SU12), and with imported scalpings lining a new pond at the Shire Horse Centre, Teffont (S193) — this latter site was wrongly mapped as native in Preston et al. (2002). Elsewhere in South Hampshire there are records from the Avon Valley at Somerley (SUIO) (1980s), and Southleigh Forest (SU70), where it was recorded in 1957 and destroyed in 1982. Recently it has been found at Ringwood Forest, west of the R. Avon, on the borders of Hampshire and Dorset, found by Neil Sanderson in 2006. WEST KENT (V.C. 16) All Kent records come from the area now known as Bedgebury Forest. It is the site of the National Pinetum, developed on an area of acid soils over gravels and clay which is surrounded by a larger forested area. Bedgebury was owned by Viscount Lord Beresford, who was a General (later Field Marshal) under the Duke of Wellington in the Spanish Peninsula wars. He bought the estate in 1836, and planted many trees in the earlier years of his ownership. The main estate was purchased from his heirs by the Crown in 1919 for forestry purposes and was developed jointly by the Forestry Commission and the Royal Botanic Gardens, Kew from 1923 until 1965, When the Pinetum reverted solely to the Forestry Commission. The first trees for the Pinetum were raised at Kew in 1921. There is an mysterious record in the Flora of Kent (Hanbury & Marshall 1899), referring to herbarium specimens that were found by a Mr Duthie (presumably Mr J. F. Duthie, who was born in Kent, botanising there until the 1870s, but then spent 30 years in India and Ceylon before returning to Kew in 1903) at Edinburgh University (now at E) labelled ‘boggy places four miles SW from Dover’. He noted that there was no likely station in the Dover neighbourhood. Hanbury & Marshall wondered whether there is a village of that name in the south-western counties. But Druce (1920) includes a record from near Doves, Kent (indeed, his correspondent pointed out that her site was about four miles away), and on the 146 D. A. PEARMAN current OS map there is a Doves Farm at TQ7834, four miles east of the Bedgebury site. The Edinburgh specimen was examined by DAP in October 2007. There are two sheets, both from the Kent site: one from _ the herbarium of W. G. Mclvor, collected by him in 1846, and another from the herbarium of a Mr Murchison, though this second sheet lacks a date and source. MclIvor’s specimen has been annotated by David McClintock as referring to Doves rather than Dover. Druce’s 1920 record seems to have become well-known, as Miss Vachell and her friends visited Sophia Wood, near Hawkhurst, in October 1923 (Rich & Forty 2005). George Hutchinson located Vachell’s herbarium specimen at Cardiff Museum (NMW), and plausibly suggested that Sophia Wood was the estate name for what is now known as Frith Wood (TQ7332). In 1944 Francis Rose reported it from rides in Bedgebury Forest (Anon 1946), where it still OCCULS) dt a 1Cs se hO7232 a inene seanes cotnen specimens at NMW from the same area collected by J. E. Lousley (1944) and R.A. Boniface (1947). It is very likely that all of these records relate to the same site, which extends the history of the plant in the Bedgebury area back to 1846. BERKSHIRE (V.C. 22) First discovered in the county by A. W. S. Fisher, a scholar at Winchester College, in 1891 (Druce 1891) and described by Druce (1897) as ‘found by a pool; it grows on the damp sandy margin with Hydrocotyle, Ranunculus, Millegrana (Radiola_linoides), Drosera, Veronica scutellata, Juncus bulbosus and others. No other introduced vegetation is near, except that it is in a district in which Pines and Rhododendrons are planted... Subsequently I found it in_ considerable abundance, and it appeared as if the //lecebrum had been covered with sand brought from the neighbourhood in order to raise the road...’. Druce did not give details of the location, but it was at the western end of Nine Mile Ride near Wellington College (SU86) (Crawley 2005). Though it was last seen there in 1956, it has been found in at least seven nearby sites, in a strip about 8 km in length from west to east. It still persists at Heath Lake (SU8265), and possibly on private land that is inaccessible nearby (Crawley 2005). Wellington College was opened as a monument to the great Duke in 1859, on a site described at the time as ‘a wild and cheerless spot on a _ bleak, inhospitable-looking moor (Wellington College website). It lies about five miles to the east of the Stratfield Saye, the house presented to the Duke of Wellington by the nation in 1817. The Kent and Berkshire sites are both on estates in which forestry plantings were started in the 19th century by military aristocrats. The possibility that they introduced Jllecebrum from a common source needs investigation. DORSET (V.C. 9) The first definite, and for many years the only, record was from the Royal Army Ordnance Corps base at West Moors (SUOO), where it was found on the edges of a pool in 1981 and later spread on the cinder base of nearby disused railway sidings. In 2002 Jonathan Spencer, Senior Ecologist for England with the Forestry Commission, reported that he had seen it, some years before, in the rides of a Commission nursery just to the north. This was probably the origin of the plants at West Moors Army site. It still survives in small quantity there. Odd plants have also been found on Army land at Barnsfield Heath (SUI1O) (political Dorset, but actually in v.c. 11) and at Woolbridge, on army training ranges (SY88). There is an intriguing earlier record, from the beach at Ferrybridge, Weymouth, in 1792, from a recorder, William Sole, who apparently knew the plant from Cornwall (Barrett 1912), which is assumed to be only a casual (see the records from Braunton and Anglesey, below). NORTH HANTS (V.C. 12) Plants were first recorded in 2005, growing in profusion along some of the wet sandy tracks used for military training near Aldershot (SU82) (Mundell 2006). Since then there have been records from at least six sites within five 10 km squares, all within the same type of habitat and use. OTHER SITES It is likely that other British records are casual occurrences, though there are a few of interest. NORTH DEVON (V.C. 4) There is a specimen in BM collected at Braunton Burrows, 1862. ANGLESEY (V.C. 52) There is a specimen in BM annotated “rocks by the sea, Anglesea’, 1921. SOUTH NORTHUMBERLAND (V.C. 67) J. P. M. Brenan (Anon 1947) reported records from a nursery near Falstone, noting that it was ILLECEBRUM VERTICILLATUM 147 suggested that the ///lecebrum was introduced with bundles of trees obtained from various nurseries. He reports that it was said to be plentiful in certain tree nurseries in the New Forest. MID LOTHIAN (V.C. 83) Another specimen in the herbarium at Edinburgh (E), labelled ‘Dickson’s garden in Edinburgh’, 19 August, 1796. OUTER HEBRIDES (V.C. 100) Specimens in E from the Isle of Barra collected by J. W. Heslop Harrison and his colleague W. A. Clark. Details are given by Clark in a cover letter to the director of the Royal Botanic Garden in Edinburgh (dated 19 July 1939) ‘thinly scattered around the margin of a loch near Borve’.). He also enclosed a specimen of Juncus capitatus, gathered the same day, noting that this latter was only found in one spot. Pankhurst & Mullin (1991) remarked that the Jllecebrum record required confirmation and it was widely dismissed, along with the Juncus, as another Heslop Harrison fraud. However, there is an earlier record from Eriksay (Goodrich-Freer 1902), and Whittington & Edwards (2000) report the finding of subfossil pollen of I/lecebrum there. STATUS AND CONCLUSIONS There is little reason to doubt the nativeness of the records from Cornwall. It has been known continuously for nearly 350 years and, whilst it is almost totally dependant on heavy stock- grazing and disturbance by animals or humans, this is the case for many other annuals of open habitats. As far as the rest of England is concerned there would be little doubt of its nativeness from a phytogeographical view if we did not know its recent history here. On the criteria used to judge other potential neophytes (see Pearman 2007) the situation would be evenly poised, with spatial coherence, trends in frequency and to some extent first recorded date everywhere other than Cornwall, being balanced by occurrence in_ semi-natural habitats, persistence and European range. On balance it is unlikely that it was overlooked for nearly 200 years after its discovery in Cornwall in the well-botanised south-east of England and it was probably introduced with various forestry operations there. In the New Forest the evidence is more equivocal; it is an intimate part of the flora of the wet lawns and pool edges and associated with a suite of native species and is a valued part of a community that conservationists wish to preserve (vide Pearman 2007). But it is still increasing its range (there has been a veritable surge in records recently); the J/lecebrum population appears to have fluctuated with changing stocking practices, leisure pursuits, the amount of tree-planting and _ gravel extraction. The source of trees planted in the early and mid-nineteenth century in Kent and Berkshire, and that of those planted in the New Forest in the twentieth century is unknown. The reasons for the delayed spread from the Forest need investigating; it is possible that the increase there coincided with the collapse of grazing elsewhere in the south of England, which thus limited the opportunities to spread. Furthermore, as //lecebrum is frost-sensitive, recent warmer winters may have helped it spread into areas where it was previously checked by hard frosts. In the latest Red List (Cheffings & Farrell 2005) I/lecebrum is categorised as Vulnerable. The heavy losses in Cornwall have, to some extent, been offset by consolidation and new sites in central-southern England. If those latter sites were not treated as native, then it would probably be in the most threatened category, Critically Endangered. It is now on the new BAP list which should stimulate considerable conservation effort in Cornwall, where currently there is none. ACKNOWLEDGMENTS I would like to sincerely thank Andy Byfield, Clive Chatters and Martin Rand for really helpful comments on a draft and particularly for details of the New Forest populations; Tony Mundell, for records and information from North Hampshire and for historical background on J. F. Duthie; George Hutchinson and Douglas McKean for information from and access to herbaria at Cardiff and Edinburgh, and Chris Preston and Kevin Walker for other assistance and comments. 148 D. A. PEARMAN REFERENCES ANON. (1946). Plant records. Botanical Exchange Club 12: 695-778. ANON. (1947). Plant records. Botanical Exchange Club 13: 43-76. BARRETT, W. B. (1912). Contributions to a Flora of Portland. Proceedings of the Dorset Natural History and Antiquarian Field Club 33: 96-143. BREWIS, A., BOWMAN, R. P. & ROSE, F. (1996). The Flora of Hampshire. Harley Books, Colchester. CHATTERS, C. (1999). Galium constrictum in M. J. Wigginton (comp. & ed.), 1999. British Red Data Books. 1. Vascular plants, edn 3. Joint Nature Conservation Committee, Peterborough. CHEFFINGS, C. M. & FARRELL, L. (2005). The Vascular Plant Red Data List for Great Britain. Joint Nature Conservation Committee, Peterborough. CRAWLEY, M. J. (2005). The Flora of Berkshire. Brambleby Books, Harpenden. DAVEY, F. H. (1909). Flora of Cornwall. F. Chegwidden, Penryn. DRUCE, G. C. (1891). New Berks Plants. Journal of Botany 29: 307. DRUCE, G. C. (1920). New County and other records. Botanical Exchange Club 6: 108-163. DRUCE, G. C. (1924). Plant notes etc., for 1924. Botanical Exchange Club 6: 438. FRENCH, C. N., MURPHY, R. J. & ATKINSON, M. G. C. (1999). Flora of Cornwall. Wheal Seton Press, Camborne. GOODRICH-FREER, A. (1902). Outer Isles. Archibald Constable & Co (Westminster). HANBURY, F. J. & MARSHALL, E. S. (1899). Flora of Kent. F. J. Hanbury, London. HAUPLER, H. & SCHONFELDER, P., eds. (1988). Atlas der Farn- und Bliitenpflanzen der Bundersrepublik Deutschland. Eugen Ulmer, Stuttgart. HULTEN, E. & FRIES, M. (1986). Atlas of north European vascular plants: north of the Tropic of Cancer. 3 vols. Koeltz Scientific Books, K6nigstein. JALAS, J. & SUOMINEN, J., eds. (1972-1994). Atlas Florae Europaeae. Volume 3 (1976). Committee for Mapping the Flora of Europe and Societas Biologica Fennica Vanamo, Helsinki. MARGETTS, L. J. & DAVID, R. W. (1981). A review of the Cornish flora. Institute of Cornish Studies, Redruth. MEUSEL, H. & JAGER, E. J. (1992). Vergleichende Chorologie der zentraleuropdischen Flora. Volume 3. 2 vols. Gustav Fischer, Jena. ) MUNDELL, T. (A. R. G.). (2006). Ladies, Roses and Coral Necklaces in v.c. 12. BSBI News 101:26—27. MURPHY, R. J. (1994). Ilecebrum verticillatum, in A. STEWART, D. A. PEARMAN & C. D. PRESTON (comps & eds.), (1994). Scarce plants in Britain. Joint Nature Conservation Committee, Peterborough. PEARMAN, D. A. (2007). “Far from any house’ —assessing the status of doubtfully native species in the flora of the British Isles. Watsonia 26: 271-290. PEARMAN D. A. (in press). The decline of I/lecebrum verticillatum in Cornwall. Botanical Cornwall. PERRING, F. H. & WALTERS, S. M. eds. (1962). Atlas of the British Flora. Thomas Nelson & Sons, London. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D., eds. (2002). New Atlas of the British and Irish flora. Oxford University Press, Oxford. RAND, M. (2005). Gladiolus illyricus in Hampshire. BSBI News 98: 21-23. RAYNER, J. F. (1929). A supplement to F. Townsend’s Flora of Hampshire and the Isle of Wight. privately published, Southampton. Ric, T. C. G. & FoRTY, M. eds. (2005). The Botanist. - the botanical diary of Eleanor Vachell (1879-1948). National Museum of Wales, Cardiff. , THURSTON, E. & VIGURS, C. C. (1922). A supplement to F. Hamilton Davey’s Flora of Cornwall. Royal Institution of Cornwall, Truro. TUBBS, C. R. (1986). The New Forest. Collins, London. WHITTINGTON, G. & EDWARDS, K. T. (2000). Illecebrum verticillatum in the Outer Hebrides. Botanical Journal of Scotland 52: 101-104. (Accepted May 2008) Watsonia 27: 149-157 (2008) 149 Corynephorus canescens (L.) P. Beauv. (Grey Hair-grass) on the Sefton Coast, Merseyside (v.c. 59) PSII Jal snes 9 Hayward Court, Watchyard Lane, Formby, Liverpool L37 3QP ABSTRACT This paper describes the history, current status and habitat of the nationally rare Corynephorus canescens on the Sefton Coast, some 250 km north-west of its nearest native localities in East Anglia. The grass is now largely confined to acidic fixed-dunes on the Southport & Ainsdale Golf Course where it has apparently increased over the past decade to a population comprising nearly 10,000 plants occupying c. 0-86 ha. KEYWORDS: Climate, conservation, fixed-dunes, golf course, habitat, population. INTRODUCTION Corynephorus canescens (L.) P. Beauv. (Grey Hair-grass) is a nationally rare, perennial grass which occurs on consolidated sand-dunes, on sandy shingle and on open sand either in coastal locations or inland. The New Atlas of the British & Irish Flora (Leach 2002) shows its recent occurrence in 14 hectads as a British native and states that the plant’s distribution has probably stabilised following losses in the 1930s C. canescens occurs mainly on or near the coasts of East Anglia, specifically East Suffolk (v.c. 25) and East and West Norfolk (v.cc. 27, 28). It is also rare inland in West Suffolk (v.c. 26) and is present in two hectads in Jersey (v.c. S(113)). Its only other “native” populations are on the Sefton Coast, Merseyside, South Lancashire (v.c. 59), some 250 km north-west of the nearest East Anglian sites. There has been some debate about the origin of C. canescens on the Sefton Coast, far from its main centre of distribution in East Anglia. Savidge et al. (1963) regard it as native, as does Leach (2002), but Hubbard (1954) and Sell & Murrell (1996) treat it as probably introduced. Occurrences in Glamorgan (v.c. 41), Worcestershire (v.c. 37), Staffordshire (v.c. 39), Moray (v.c. 95), Westerness (v.c. 97), and Co. Antrim (v.c. H39) are shown as “non- native’ in the New Atlas (Leach 2002). However, Ryves ef al. (1996) suggest that the plant’s recent arrival in Scotland may be an expansion of its natural range, while Trist (1998) also argues that the Staffordshire and Worcestershire sites could be relicts of ancient heathlands and therefore also native localities. As it is known from only 27 localities, the plant’s conservation status is described as “near threatened” by Cheffings & Farrell (2005) whilst, on the Sefton coast, concern for its correct management has resulted in the prepar- ation of a “Species Action Plan” (Merseyside Biodiversity Group 2001). According to Stace (1991) the main habitats of C. canescens are leached fixed-dunes and inland sandy heathlands on acid soils. Marshall (1967) reports that the plant can also grow in calcareous dunes but the substrate must be very low in mineral nutrients. Ellenberg Indicator Values for this species in Hill et al. (2004) are listed below: e Light: 9 e Moisture: | Full light. Extreme dryness; restricted to soils that often dry out for some time. Mainly on acid soils but exceptionally also on near-neutral ones. Extremely infertile sites. e Salinity: 0 .- Absent from saline sites. Marshall (1965, 1967, 1968) states that individual plants are relatively short-lived (2-6 years), so the grass relies on high reproductive capacity and seed viability to maintain its populations, seed successfully re-sowing itself in stable sand for several years. C. canescens grows most vigorously where there is up to 10 cm of sand accumulation per year, and it is a potentially long-lived perennial providing sand accretion is taking place. This seems to be due, in part, to the fact that the efficiency of the root e Reaction: 3 e Nutrients: | 150 P. H. SMITH system decreases with age, good growth being dependent on the production of new roots positioned higher on the morphological axis and which is stimulated by sand accumulation. This is a particularly distinctive and attract- ive grass, densely-tufted with stiff, bristle-like blue-grey blades and purplish-pink sheaths. Trist (1998) describes how, in early July, plants can be seen at a distance of 30-40 metres when the panicles are spreading at anthesis and the colony is a mass of silver and pale pink. In autumn, the bleached inflorescences also stand out when seen from a considerable distance (personal observations). OCCURRENCE ON THE SEFTON COAST As earlier recorders did not have access to modern satellite positioning technology, details of localities are often imprecise. Travis (1928) states that the plant was thought to have been discovered in 1928 by Sir G. Talbot but was actually first recorded, though not identified, by F. W. Holder on 4 May 1919 “among the rough and high dunes west of the pine woods near the Formby golf links”. This record is also included in Savidge et al. (1963) who state that it apparently became extinct at one of its two known sites in 1937. Roberts & Smith (2007) include a photograph taken on | August 1937 by Holder, showing J. N. Frankland and J. E. Lousley with C. canescens “on dunes near Ainsdale”. This seems likely to have been Holder’s original locality. There are herbarium specimens of C. canescens in LIV collected by Holder, Frankland and J. D. Massey between 1929 and 1931 but, frustratingly, the localities given are extremely vague: “sandy slopes Formby dunes”; “Freshfield sandhills’; “dry dunes Formby”; “near Southport.” Similarly, a specimen collected by V. Gordon on 16 September 1986 is simply labelled “Ainsdale”’. The database of the New Flora of South Lancashire Project lists records made in 1982 by, BS Doody, Ra hicldeand iy CG akicht son the edge of the golf links” (G.R. $D318128), while P. Lockwood recorded the plant at G.R. SD318127 on 22 August 1987. This location appears to lie in a transitional zone between an Ainsdale recreation ground and the Southport & Ainsdale Golf Course and is presumably the site described by Trist (1998) as “a recent find” (i.e. about 1980) at G.R. SD318127. He states that “...at Ainsdale, accretion is now active and rabbits are contributing to the sand accretion requirements of Corynephorus canescens. Here the site is c. 100 x 4 m where our grass is dominant on two fixed dunes and has only Hypochaeris radicata and Ornithopus perpusillus as constant species.” However, Trist’s description does not accord with the playing area of the golf course, where bare sand lying outside bunkers and_ rabbit (Oryctolagus cuniculus) activity are not tolerated by course managers. The first National Vegetation Classification (NVC) survey of the Sefton Coast (Edmondson et al. 1988-1989) records extensive stands of C. canescens on Southport & Ainsdale Golf Course. It occurred in “a possible variant of SD13b”, later reclassified as SDlla (Carex arenaria — Cornicularia aculeata_ dune, Ammophila_ arenaria sub-community), with much Agrostis capillaris and mosses, or as a co-dominant with Festuca ovina. Other associates included Ammophila arenaria, Anthyllis vulneraria, Carex arenaria, Hypochaeris radicata, Sedum acre, the hair- moss Polytrichum juniperinum, and _ lichens (Cladonia). This site is at least 2 km from its earliest known locations, being in an area of low dunes, rather than the high dune landscape where the grass was first reported. The plant was also found by D. A. Nissenbaum growing abundantly on railway track ballast near Ainsdale (Atkinson & Houston 1993). In his survey of Sefton Coast dune-heath in 1993-1994, Gateley (1995) records C. canescens in two quadrats on Southport & Ainsdale Golf Course. He states that the Polytrichum Juniperinum-dominated low dunes, with frequent C. canescens, found here are unique locally. Soils are generally acidic with pH 3:43-4:44 at the surface and 3:82-4:62 at 10 cm depth. The Sefton Coast Life Project (co-financed by the European Commission through the community LIFE-Nature fund) mapped eight large and seven small colonies on Southport & Ainsdale Golf Course in 1996-1998. The surveyors also located two patches of the grass on the eastern edge of Hillside Golf Course, across the railway line from one of the Southport & Ainsdale sites. A second NVC survey of the dune system in the years 2003 and 2004 (Gateley & Michell 2004) again found large populations of C. canescens at Southport & Ainsdale Golf Course, on the low eastern dunes with a very short acid sward, rich in the moss Polytrichum juniperinum. The CORYNEPHORUS CANESCENS ON THE SEFTON COAST 15] TABLE 1. SUMMARY OF CORYNEPHORUS CANESCENS DATA FOR 23 GOLF COURSE, RAILWAY TRACK AND RECREATION GROUND SITES Site no. Grid References Ne. SD3164412837—-SD3 18612849 Ds SD3161912825—SD3 164412780 3L Not found 4L Not found SU, SD3155312758—SD3 153512828 6L $D3 156812798 TG S$D3168513255 SD3 170413247 8L SD3195613100—SD3 199113130 OL Not found OI $D3177612970—-SD3 179312970 IUD Not recorded WAL $D3182112848-SD3 185412855 3 $D3187312855 14S $D3177613411 Ss) Not found 16S Not found LS Not found RSSz $D3 187212950 19S* $D3 180312914 20S -§D3183512883 21S $D3170012787 22S $D3 162012745 DBS $D3 150012782 R $D317132 Rec $D31911276—SD3 1931278 Total Dimensions Area(m’) — Aspect Estimated no. (m) of plants ISHS Zl sz] S 5000 45 x 15 675 - 600 70 x 22 1540 W 500 IS XO 90 - 60 x3 47 W 80 IlOx 2 60 x 22 1320 SE 1500 24 x 20 480 - 250 WDE 66 - 100 40 x 2 80 - 230 ISX) 135 W 300 «3 15 S 20 DD + - AO 5X5 ES W LS 5x3 IES) W 10 - - - Ih Sx 90 W 200 3x38 9 - IS . 20 SIlxes 93 - 110 8611 9951 * = new site not mapped in 1997; L = large; S = small; R = Railway track; Rec. = Ainsdale recreation ground authors attribute the vegetation to Ul (Festuca ovina — Agrostis capillaris — Rumex acetosella grassland) which is characteristic of grazed swards on acid substrates in the pastoral upland fringe. The Ul community has similarities to SD12 (Carex arenaria — Festuca ovina — Agrostis capillaris dune grassland), and grades into it, but generally lacks typical coastal dune plants. THE 2007 SURVEY — RATIONALE The Species Action Plan for C. canescens recommends surveys of the golf course colonies at five-year intervals (Merseyside Biodiversity Group 2001). Therefore, ten years on from the Sefton Coast Life Project survey, it was thought desirable to revisit the main sites for the plant on Southport & Ainsdale Golf Course. Accordingly, permission for a survey was obtained in mid-September 2007, a time of year when the plants are particularly noticeable. METHODS All colonies mapped by the Sefton Coast Life Project and other sites recommended by the golf course manager were visited on 18 and 20 September 2007. The positions of the colonies were determined using a hand-held Garmin Etrex Global Positioning System and the areas occupied by C. canescens were estimated by pacing. The associated vascular plant taxa were also listed and an indication of their relative frequency was estimated using the DAFOR 152 P. H. SMITH Corynephorus canescens 100m Grid Small colonies FIGURE |. Location of Corynephorus canescens colonies 1996-1998, Southport and Ainsdale golf course scale. Notes were made on _ habitat characteristics, including aspect. In addition, visits were made to the Liverpool Road Cemetery, Ainsdale, the nearby Ainsdale recreation ground and the trackside for a distance of | km north of Ainsdale Station. Hillside Golf Course was not surveyed. Distribution maps were prepared by staff of the Technical Services Department of Sefton Metropolitan Borough Council, using the Sefton Coast Geographic Information System (Figs | & 2). CORYNEPHORUS CANESCENS ON THE SEFTON COAST 153 cl] Oe Corynephorus canescens 100m Grid ‘® Small colonies Ws Large colonies FIGURE 2. Locations of Corynephorus canescens colonies 2007, Southport and Ainsdale golf course and Ainsdale recreation ground RESULTS Although a large area of ostensibly suitable acidic fixed-dune habitat is present in Liverpool Road Cemetery, Ainsdale, no C. canescens was found, neither was a search of the railway north of Ainsdale Station successful. However, a population of 110 plants was discovered on a low ridge 31 m long x 3 m and c. 0:5 m high near the western boundary of the Ainsdale recreation ground at G.R. SD319128 (Table 1, Fig. 2). Vascular plant associates were Agrostis 154 P. H. SMITH capillaris, _Ammophila_ arenaria, Bellis perennis, Carex arenaria, Festuca ovina, Hypochaeris radicata, Plantago coronopus and Senecio jacobaea which suggests a moderately acidic substrate. There is also a high frequency of mosses and some patchy bare ground caused by regular mowing. No evidence of rabbit- grazing could be seen. Six of the golf course sites mapped by the Sefton Coast Life Project (three large and three small) could not be refound but eight new sites (five large and three small) were located and mapped (Figs | & 2). In all cases, the habitat of extant sites was described as a short, open, sparse community of grasses (mainly Agrostis capillaris and Festuca ovina) and mosses with frequent rosette-forming composites (espec- lally Hypochaeris radicata) and often some Rumex acetosella. Many sites are on fixed- dune crests or on steep slopes, usually west- or south-facing, although some were found on relatively level or slightly undulating terrain; none occur on north-facing slopes. There was no sign of rabbit-grazing but many sites are affected by human activity, most of it light to moderate trampling by golfers. Some populations of C. canescens are situated on grassy footpaths and one turf-stripped area (site 18) has been recently colonised by the plant. A seven-metre diameter bunker at site 1, filled in three years ago, was planted up with C. canescens by course managers (M. Mercer pers. comm. 2007) and now supports a vigorous sward of the grass. Adjacent to site 7, c. 20 plants of C. canescens could be seen (through binoculars) within the railway fence on a disturbed, sandy strip next to track ballast. A small site mapped on the edge of Hillside Golf Course, opposite site 7, appears now to be too heavily vegetated to be suitable for C. canescens. The largest population (c. 5000 plants) was found at site 1, where densities of c. 20 tussocks per square metre occur on a south-facing fixed-dune ridge. A list of associates of C. canescens is given in Table 2; only 35 vascular taxa (three non- native) were noted, most being characteristic of fixed-dunes with a somewhat acidic substrate. In nearly all cases, the vegetation of extinct sites (e.g. 3, 4, 7 and 9) was deemed to be now unsuitable for C. canescens, due to coarsening of the sward. The total golf course population of C. canescens is estimated at nearly 10,000 plants, occupying an area of c. 8600 m° (0-86ha) (Table 1). DISCUSSION On the Sefton Coast, C. canescens was first recorded on high dunes near the sea between Formby and Freshfield. Although the exact localities were not specified, it is clear that the plant has been long extinct in this area, its recent distribution being confined to an area on, and adjacent to, the Southport & Ainsdale Golf Course, which is 2 km inland. While the 1996-1998 Sefton Coast Life Project survey did not estimate the area of the sites or the population sizes, it is reasonable to infer that C. canescens has consolidated its status on the golf course. It has disappeared from six sites, but eight new ones were found; these supported c. 1500 additional plants (Table 1). Also, two of the largest sites (1 and 8) now cover consider-ably larger areas than in 1997 (Figs 1 & 2). Losses are probably attributable to habitat change, specifically the development of taller and denser grass swards. The recreation ground population is evid- ently a relict of that recorded by Doody, Field and Rich in 1982 and Lockwood in 1987, subsequently described by Trist (1998), alth- ough it seems now to be much smaller in area. The vegetation of the golf course sites accords well with the Gateley (1995) and Gateley & Michell (2004) descriptions, though the latters’ determination of the U1 grassland community may be unsatisfactory in view of the many fixed-dune plants recorded as associates of C. canescens (Table 2). Although no NVC samples were taken for analysis, the species composition seems closer to the rather similar SD12 Carex arenaria — Festuca ovina — Agrostis capillaris dune grassland than to U1. SD12 is characteristic of fixed, acidic sands around the coasts of Britain, especially towards the north and west, where calcareous wind- blown material has been leached over time (Rodwell 2000). However, in places, the vegetation may be closer to the SD11 (Carex arenaria — Cornicularia aculeata dune comm- unity), a lichen-rich type associated with Corynephorus canescens on the East Anglian coast (Rodwell 2000) and which Edmondson et al. (1988/89) considered best to represent the golf-course vegetation. Gateley & Michell (2004) mapped only 3-3 ha of SDI1 on the Sefton Coast, mostly associated with pine plantations and the oldest, most leached, eastern soil of the dune system. However, the fixed-dune communities on Southport & Ainsdale Golf Course were poor statistical fits TABLE 2. VASCULAR ASSOCIATES OF CORYNEPHORUS CANESCENS AT SOUTHPORT & AINSDALE GOLF COURSE CORYNEPHORUS CANESCENS ON THE SEFTON COAST Taxon Agrostis capillaris Aira praecox Ammophila arenaria Anthoxanthum odoratum Anthyllis vulneraria Arenaria serpyllifolia Asparagus officinalis* Calluna vulgaris Campanula rotundifolia Carex arenaria Centaurium erythraea Conyza canadensis * Cytisus scoparius Deschampsia flexuosa Equisetum arvense Festuca ovina Festuca rubra Galium verum Hypochaeris radicata Leontodon saxatilis Lotus corniculatus Oenothera xfallax* Ononis repens Ornithopus perpusillus Pilosella officinarum Plantago coronopus Plantago lanceolata Polypodium vulgare Rubus caesius Rumex acetosa Salix repens Sedum acre Senecio jacobaea Trifolium arvense Viola canina Total 35 taxa (3 alien) English name Frequency Common Bent Early Hair-grass Marram Sweet Vernal-grass Kidney-vetch Thyme-leaved Sandwort Garden Asparagus Heather Hare-bell Sand Sedge Common Centaury Canadian Fleabane Broom Wavy Hair-grass Field Horsetail Sheep’ s-fescue Red Fescue Lady’s Bedstraw Cat’ s-ear Lesser Hawkbit Bird’s-foot-trefoil Intermediate Evening-primrose Common Restharrow Bird’ s-foot Mouse-ear Hawkweed Buck’s-horn Plantain Ribwort Plantain Se a in OO oy SEN © Si) i Oa © © © =: Common Polypody lo Dewberry If Sheep’ s-sorrel f Creeping Willow lo Biting Stonecrop lo Common Ragwort ip Hare’ s-foot Clover O Heath Dog-violet N * = non-native or introduced native taxon; a = abundant; f = frequent; o-= occasional; r = rare; v = very; 1 = locally. to known NVC communities (P.S. Gateley, pers. comm. 2007). C. canescens was often found on fixed-dune crests as well as on quite steep south- or west- facing slopes. Such sites will be well-drained and susceptible to summer dryness, which presumably gives a competitive advantage to this relatively drought-tolerant species. Marshall’s (1967) findings that the plant requires sand accretion for vigorous growth and a long life-span do not explain the plant’s success at the golf course where there appears to be little sand movement. Although the grass has colonized a turf-stripped site and has grown well on an infilled bunker at site 1, management to create greater instability seems unnecessary at the golf course. At the Ainsdale recreation ground, the C. canescens colony has persisted for at least 25 years, apparently assisted by regular mowing which creates bare ground for _ seedling establishment. Although small, this population 156 125 Jal, SIMIOUa! is of particular interest, being the only one readily accessible without special permission. In Merseyside, C. canescens is near the northern limit of its European distribution, which closely matches the 15°C July isotherm. The climate in northern Britain is cold enough to have a cumulative effect on flowering date and seed germination, sufficient to postpone the emergence date of seedlings beyond the critical time for survival (Marshall 1968). Therefore, recent increases in summer temperatures may well have benefited repro- duction and survival of the Sefton Coast populations. Most of the existing populations on the golf course appear healthy with plenty of young plants. There also seem to be extensive areas of potentially suitable habitat in the golfers’ roughs to allow for further spread. Although no evidence of rabbit-grazing was found and the grassy roughs are not mown (M. Mercer pers. comm. 2007), the extreme infertility of the old, heavily leached dune soils here, coupled with small-scale disturbance from course activity, seem capable of maintaining C. canescens populations at present. However, the long-term future of the plant may be threatened by nitrogen deposition from air pollution which is reaching critical levels in northwest England, perhaps sufficient to cause vegetation change (Gateley & Michell 2004). The Sefton Coast Life Project included objectives for C. canescens conservation in a Site Management Plan for Southport & Ainsdale Golf Course (1998). Since about 1999, the Southport & Ainsdale Club, with the support of Countryside Stewardship funding, has pursued a programme of “rough” management designed to restore and conserve the dune and dune heath character of the course (Gill 2004). As this is a nationally rare plant, the abundance of C. canescens at Southport & Ainsdale Golf Course has great conservation significance. Indeed, based on Trist’s (1998) data, this may be the largest population in Britain outside North Norfolk and the Channel Isles. The presence of this species is also listed as a reason for the notification of the Sefton Coast Site of Special Scientific Interest, which was extended in 2000 to include the Southport & Ainsdale Golf Course. ACKNOWLEDGMENTS I am grateful to the Southport & Ainsdale Golf Club for giving me permission to carry out the survey and to the Course Manager, Mike Mercer, for information and guidance. Pat Lockwood very kindly assisted with much of the field work and Steve Cross was extremely helpful in searching LIV and the Liverpool Museum library for records and early literature sources. Dave Earl provided records from the New Flora of South Lancashire Project and Michelle Newton and Graham Lymbery assisted with preparation of distribution maps. REFERENCES ATKINSON, D. & HOUSTON, J., eds. (1993). The Sand Dunes of the Sefton Coast. National Museums & Galleries on Merseyside, Liverpool. CHEFFINGS, C. M. & FARRELL, L., eds. (2005). The Vascular Plant Red Data List for Great Britain. Joint Nature Conservation Committee, Peterborough. EDMONDSON, S. E., GATELEY, P. S. & NISSENBAUM, D. A. (1988-1989). National Sand Dune Vegetation Survey: the Sefton Coast. Nature Conservancy Council, report no. 917. Peterborough. GATELEY, P. S. (1995). Sefton Coast Dune Heath Survey 1993-1994. Report to Sefton Metropolitan Borough Council, Bootle. GATELEY, P. S. & MICHELL, P. E. (2004). Sand Dune Survey of the Sefton Coast. Report by TEP, Warrington to Sefton Metropolitan Borough Council. GILL, B. (2004). Conservation management at Southport and Ainsdale Golf Club. Coastlines, summer 2004: Wee HILL, M. O., PRESTON, C. D. & Roy, D. B. (2004). PLANTATT. Attributes of British and Irish Plants: Status, Size, Life History, Geography and Habitats. Centre for Ecology & Hydrology. Monks Wood, Cambridge. HUBBARD, C. E. (1954). Grasses. Penguin Books, Harmondsworth. LEACH, S. J. (2002). Corynephorus canescens Grey Hair-grass, in C. D. PRESTON, D. A. PEARMAN & T. D. DINES, eds. New Atlas of the British & Irish Flora, p. 771. Oxford University Press, Oxford. MARSHALL, J. K. (1965). Corynephorus canescens (L.) P. Beauv. as a model for the Ammophila problem. Journal of Ecology 53: 447-463. MARSHALL, J. K. (1967). Biological Flora of the British Isles, no. 105. Corynephorus canescens (L.) Beauv. Journal of Ecology 55: 207-220. CORYNEPHORUS CANESCENS ON THE SEFTON COAST 157 MARSHALL, J. K. (1968). Factors limiting the survival of Corynephorus canescens (L.) Beauy. in Great Britain at the northern edge of its distribution. Oikos 19: 206-216. MERSEYSIDE BIODIVERSITY GROUP (2001). North Merseyside Biodiversity Action Plan. Environmental Advisory Service, Maghull. ROBERTS, F. J. & SMITH, M. E. (2007). Joseph Norman Frankland (1904-1995): a tribute. BSB News 106: 13-15. RODWELL, J. S., ed. (2000). British Plant Communities volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge. RYVES, T. B., CLEMENT, E. J. & FOSTER, M. C. (1996). Alien grasses of the British Isles. B.S.B.1., London. SAVIDGE, J. P., HEYWOOD, V. H. & GORDON, V., eds. (1963). Travis’s Flora of South Lancashire. Liverpool Botanical Society, Liverpool. SELL, P. D. & MURRELL, G. (1996). Flora of Great Britain and Ireland. 5: Butomaceae — Orchidaceae. Cambridge University Press, Cambridge. STACE, C. A. (1991). New Flora of the British Isles. Cambridge University Press, Cambridge. TRAVIS, W. G. (1929). Scirpus americanus Pers. and Weingaertneria canescens Bernh. in Lancashire. North Western Naturalist 4: 175-177. TRIST, P. J. O. (1998). The distribution of Corynephorus canescens (L.) P. Beauy. (Poaceae) in Britain the Channel Islands with particular reference to its conservation. Watsonia 22: 41-47. (Accepted February 2008) oa ee pe IGA a ge ms ne came: Watsonia 27: 159-166 (2008) 159 Population explosion of Hypochaeris glabra L. on the Sefton Coast, Merseyside in 2007 PHILIP H. SMITH 9 Hayward Court, Watchyard Lane, Formby, Liverpool L37 3QP ABSTRACT Hypochaeris glabra has been historically scarce and recently rare in South Lancashire (v.c. 59) and is poorly represented in North-west England. In 2007, 28 populations supporting over 5200 plants and occupying about 2:4 ha were found on the Sefton Coast sand-dunes. The typical habitat is rabbit- grazed fixed-dune with a short, open sward, often with a substantial cover of mosses and lichens. The composition of associates suggests a neutral to somewhat acidic substrate at most sites. It is surmised that unusual weather conditions contributed to this apparent population explosion. KEYWORDS: Climate, grazing, habitat, Hypochaeris glabra, population, sand-dunes, Sefton Coast, v.c. 59. INTRODUCTION Hypochaeris glabra L. (Smooth Cat’s-ear) is a native annual of open, summer-parched grasslands and heathy pastures, on usually acidic, nutrient-poor, sandy or gravely soils, also occurring in dune grassland or on sandy shingle. It was formerly widespread as a weed of arable fields and as a wool-shoddy alien (Preston et al. 2002). Rodwell (2000) includes H. glabra as a component of the SD19 Phleum arenarium — Arenaria serpyllifolia dune annual community. This plant is of European southern temperate origin but is widely naturalised outside its native range (Preston et al. 2002). Thus, in parts of North America and Australia it is classed as a “problem weed” on sandy soils. STATUS IN THE BRITISH ISLES H. glabra is scattered throughout Britain from the Channel Islands to Scotland where, however, there are only four recent records. It is mainly concentrated in the East Anglian Breckland and Sandlings, the Surrey heaths and the Welsh Borders (Wilson & King 2003). The plant is declining in semi-natural habitats, having disappeared in many areas as a result of agricultural improvement or loss of grazing. However, it is easily overlooked, particularly as the flowers close in the afternoon (Preston et al. 2002). The species has a Change Index (1930-1999) of -1.01, its conservation status is given as “vulnerable” and it is a U.K. Species of Conservation Concern (Cheffings & Farrell 2005). Confined to Ulster in Ireland, H. glabra is Red Listed and protected under the Wildlife (N.I.) Order, 1985. In North-west England, the plant is notified as a Species of Conservation Importance (Regional Biodiversity Steering Group 1999). It is poorly represented in most vice-counties in this region. There are no modern records for Cheshire (Ween 58) huss ords der Mabley, (1899) states that H. glabra has been found in three stations in fields and on banks on a sandy soil but Newton (1971) lists the plant as apparently extinct in the county. It has never been found in West Lancashire (v.c. 60), perhaps because the only likely habitat on the coastal golf courses is difficult of access (E. F. Greenwood pers. com.). H. glabra has been recorded from 13 tetrads in Cumbria (v.c. 69, WC, Oo, Wc. JO), WINES Te 1S eI Cm ererel sandy ground and sand-dunes from South Walney to Ravenglass (Halliday 1997). The plant is also known from the Ayres’ dunes (three hectads) in the north of the Isle of Man (O76, WW), HISTORY IN SOUTH LANCASHIRE (V.C. 59) The earliest relevant floras (Hall 1839; Dickinson 1851) do not mention H. glabra. Twenty-six records for South Lancashire before 2007 are listed in Table 1. The first for the vice-county is given in Savidge ef al. (1963) as “Moseley Common” (undated, pre- 1900). For this record, the New Flora of South Lancashire Project cites the Grid Reference SD7201 and the date 1859. Moseley Common still exists, situated between Tyldesley and Worsley in Greater Manchester. There are only two other records away from the Sefton Coast, both in 1975 for Gorse Wood, Westhoughton (SD6506) (New Flora Project). 160 P. H. SMITH TABLE 1. PRE-2007 RECORDS OF H. GLABRA IN SOUTH LANCASHIRE Year Locality Recorder Herbarium sheet 1859 Moseley Common Anon. 1866 Half-a-mile north of Crosby Lord de Tabley 1869 Freshfield G. E. Hunt 1870 Formby H. S. Fisher MANCH 1873 Formby H. S. Fisher 1875 Hightown R. Brown LIV 1876 Hightown J. H. Lewis 1878 Ainsdale J.C. Melvill Melvill 1879 Ainsdale J. H. Lewis 1879 Southport, Ainsdale J. H. Lewis MANCH 1880 Birkdale @eiiiGreen 1881 Southport @GalaGreen 1907 Freshfield W. G. Travis LIV 1909 Freshfield W. G. Travis LIV 1926 Ainsdale J. N. Frankland LIV 1933 Birkdale Cemetery F. W. Holder LIV 1950 SD31 Anon. 1950 SD20 Anon. 1953 Ainsdale F. W. Holder LIV 1953 Freshfield F. W. Holder LIV 1953 Freshfield V. Gordon LIV 1956 Formby M. H. Bigwood 1975 Gorse Wood, Westhoughton M. E. Martin IDS Gorse Wood, Westhoughton E. H. Hancock, E. G. & M. E. Martin 2000 Ainsdale Sand Dunes N.N.R. P. H. Smith 2003 Ainsdale Sand Dunes N.N.R. S. Taylor Sources: New Flora of South Lancashire Project, Liverpool Museum Herbarium (LIV), Manchester Museum Herbarium (MANCH) The first record of H. glabra on the Sefton Coast was by Hon. J. L. Warren (Lord de Tabley) who found the plant in 1866 on sandhills half-a-mile north of Crosby (Brown 1875). The next sighting was by G. E. Hunt: “Sandy ground, Freshfield in 1869” (Brown 1873; Wheldon 1909). There are also records for Formby in 1870 and 1873, Hightown in 1876, Ainsdale in 1878 and 1879 and Birkdale in 1880 (Table 1). Melvill (1909) had a sheet of H. glabra in his herbarium labelled “Very scarce on _ sandhills at Ainsdale, S.W. Lancashire .... August 24" 1878”. Some of these early records are represented by sheets in LIV and MANCH (Table 1). Green (1933) summarises the status of H. glabra in the Liverpool district as: “Sand dunes. Rare. Hightown to Birkdale.” Travis’s Flora of South Lancashire (Savidge er al. 1963) describes the occurrence of the plant as: “Occasional on sand-dunes, rare elsewhere’, and cites records thus: “Sand-dunes from Freshfield to Ainsdale; by Birkdale Cemetery; sand-dunes, Crosby to Hightown.” Prior to the present study, the only Sefton Coast records in the last 50 years are for Ainsdale Sand Dunes National Nature Reserve in 2000 and 2003. The Ainsdale records were from Phase | of the Dune Restoration Area Where pine plantations were felled in 1992, followed by winter sheep-grazing. However, a Wildflower Society meeting on 9 July 1994, led by Mrs Shirley Brown, reports finding H. glabra near the National Trust carpark at Victoria Road, Formby Point, some 2 km to the south. The above observations strongly suggest that H. glabra has historically been a scarce plant in the vice-county, but particularly rare over the last half-century and largely confined to the sand-dunes of the Sefton Coast. Before 2007, I had only seen it once (Ainsdale N.N.R., 2000) in nearly 40 years of recording. HYPOCHAERIS GLABRA ON THE SEFTON COAST TABLE 2. SUMMARY OF H. GLABRA POPULATION DATA Site Locality l Blundell Avenue, Formby Point 2 Blundell Avenue, Formby Point 3 Blundell Avenue, Formby Point 4 Blundell Avenue, Formby Point 5) Blundell Avenue, Formby Point 6 Victoria Road Formby Point di Victoria Road, Formby Point 8 Victoria Road, Formby Point 9 Victoria Road, Formby Point Victoria Road, Formby Point Victoria Road, Formby Point Lifeboat Road, Formby. Point Lifeboat Road, Formby Point Lifeboat Road, Formby Point Lifeboat Road, Formby Point NOR Re Re SCoOmMoNnNANBRWNK OC Formby Golf Course Formby Golf Course Formby Golf Course Formby Golf Course Ainsdale Sand Dunes N.N.R. 21 Ainsdale Sand Dunes N.N.R DD) Ainsdale Sand Dunes N.N.R. 23 Ainsdale Sand Dunes N.N.R. 24 Ainsdale Sand Dunes N.N.R Dy = Ainsdale Sand Dunes N.N.R. 26 Ainsdale Sand Dunes N.N.R. Da, Ainsdale Sand Dunes N.N.R. 28 Ainsdale Sand Dunes N.N.R. Total 16] Grid reference Area No. of No. of vascular (m°) plants associates $D2809707780 | DZ, 14 $D2799707925 1300 c.500 35) $D2794507920 4500 c.100 DS SD2780408094 5 c.50 17 S$D277076 700 c.500 40 SD2791408565 50 30-50 10 SD2785608830 350 c.500 19 $D2786408544 120 c.500 20 $D2785408603 37/5 100+ 21 $D2778 108576 64 100+ 16 $D2789208522 26 c.20 1 SD2782706883 200 c.500 24 $D2778406844 3) c.30 15 S$D277980681 1 4 c.10 10 SD2763706878 10,000 c.100 7 S$D2765309013 850 1000+ Dopp $D2780009020 2850 c.500 DD $D2778909247 1100 C20 20 $D2776009242 300 100+ Dp) $D2914610822 116 c.20 3} S$D2915510820 36 c.50 13 $D2916910817 30 c.20 19 $D2917110769 104 c.50 20 $D2916510752 Di, 200+ 16 $D2917510702 18 c.50 16 $D2913410720 12 c.30 21 $D2913810702 288 c.100 18 SD2905610691 LS c.50 Dil DBI833 5242+ THE 2007 SURVEY — METHODS On 19 June 2007, while surveying a colony of Pyrola minor on National Trust land north of Victoria Road, Formby Point, with Pat Lockwood, I came across two colonies of H. glabra on _ heavily Rabbit-grazed, acid grassland (National Grid Reference SD2708). thenone July 2007, we found more populations on similar habitat near Blundell Avenue, Formby about 500 m south of the first discovery (SD2807, SD2707, SD2708). These initial sightings led to a more systematic search of the dune system during July, August and early September, with the result that additional colonies were found on the National Trust property and also at Sefton Council’s Lifeboat Road site (SD2606, SD2706), at Natural England’s Ainsdale Sand Dunes National Nature Reserve (SD2910) and on Formby Golf Course (SD2709). Visits were also made to potentially suitable habitat at Ravenmeols dunes (SD2705, SD2706), Birkdale Common (SD3114), Liverpool Road Cemetery, Ainsdale (SD3213) and the southern half of Southport & Ainsdale Golf Course (SD3112, SD3113) but no H. glabra was found. For each site with AH. glabra, associated vascular plants were listed, their relative abundance being assessed using the DAFOR scale. Locations were determined using a hand- held Garmin Etrex GPS, the approximate areas of colonies were measured by pacing and an estimate made of the number of H. glabra plants in each. Notes were also made on habitat characteristics. RESULTS A total of 28 populations of H. glabra was recorded, five at Blundell Avenue and six at Victoria Road on National Trust property, four at Lifeboat Road, four at Formby Golf Course and nine at Ainsdale Sand Dunes National Nature Reserve. These cover an area of about 23,800 m° (2:4 ha) and are estimated to contain P. H. SMITH S : Figure |. Locations of Smooth Cat’s-ear Hypochaeris glabra on the Sefton Coast over 5200 plants (Table 2, Map 1). The largest single colony is at site 16 on the western edge of Formby Golf Course, and supports over 1000 plants. This is a large (850 m*) site which has been recently turf-stripped, providing an open habitat that has been colonised by ruderal vegetation, including H. glabra. Other large populations were found at sites 2 and 5 near HYPOCHAERIS GLABRA ON THE SEFTON COAST 163 Blundell Avenue, Formby Point. Site 2 is a transitional zone between a former Asparagus field and a conifer plantation, while site 5 is part of a disused Asparagus field next to the National Trust site office. In recent years, the latter site has been mowed annually but, like site 2, has recently been planted up with young conifers. Both localities support hundreds of H. glabra plants at relatively high density in a plant community dominated by Carex arenaria and Agrostis capillaris. The largest number of colonies (nine) was found on Ainsdale N.N.R. in Phase | of the Dune Restoration Area. Plants were largely confined to an eastern zone of open, more consolidated, dune within about 200 m of the existing conifer woodland. These populations have a relatively low density of H. glabra plants. A scattering of isolated indivi- duals was noted between the main colonies and also on the adjacent access track which is used by Land Rovers. Species-richness of sites is generally low, ranging from a mean of 16-3 at Victoria Road to 26:2 at Blundell Avenue, the overall average being 19-2 vascular taxa per site (Table 2). A total of 83 vascular taxa was recorded as associates of H. glabra (Table 3). Agrostis capillaris and Carex arenaria are each absent from only two sites, while Senecio jacobaea, Erodium lebelii, Crepis capillaris, Sedum acre, Centaurium erythraea, Arenaria serpyllifolia and Lotus corniculatus also have high rates of occurrence. The species composition suggests a neutral to somewhat acidic substrate at most sites. The habitat at almost all sites was described as fixed-dune with a short, open sward, often with a substantial cover of mosses and lichens. Except on Formby Golf Course, there is invariably evidence of heavy rabbit-grazing. The Ainsdale N.N.R. sites are also winter- grazed by sheep. Moderate human trampling is a factor at some localities, but heavy recreational pressure seems to result in an absence of H. glabra. DISCUSSION It is apparent that, on the Sefton Coast, H. glabra is associated with a rather specific habitat type; sparsely vegetated, short open turf, usually rabbit grazed, with a high frequency of mosses and lichens and a low species-richness of vascular plants, many of which are indicative of neutral to moderately acidic soil conditions. The plant seems to be absent from areas heavily disturbed by trampling, or sand-blow, preferring a more consolidated substrate. Few individuals were found in more luxuriant, grassy swards or where fresh sand was being deposited. As Sefton Coast dune sand is initially high in calcium, the plant tends to be found in older parts of the dune system where prolonged leaching has washed out much of the calcium carbonate (Smith 1999). Several colonies are on the edges of pine plantations or in areas cleared of pines, where the presence of needle litter may have increased levels of acidity. The Formby Golf Course colonies are nearer the sea than the others, the substrate appearing less acid than elsewhere. However, these dunes are actually quite old, their proximity to the shore being due to over 400 m of marine erosion since the early twentieth century (Smith 1999). All these golf course sites have recently been turf-stripped, H. glabra being part of ruderal vegetation recolonising the disturbed areas which, in this case, shows no evidence of rabbit grazing. These observations accord with habitat characteristics described in the literature, namely impoverished sandy grasslands and abandoned arable fields on non-calcareous soils (Wilson & King 2003). Although samples were not taken to determine N.V.C. communities, the associated species list (Table 3) suggest that most of the sites accord with the SD12 Carex arenaria- Festuca ovina-Agrostis capillaris dune grassland, though some stands may be closer to, or include patches of, SD19 Phleum arenarium-Arenaria serpyllifolia dune annual community (Rodwell 2000). Of the commoner associates, Agrostis capillaris and Carex arenaria are listed by Rodwell (2000) as major components of SD12, while Arenaria serpyllifolia, Carex arenaria, Centaurium erythraea, Crepis capillaris, Sedum acre and Senecio jacobaea are similarly listed for SD19. In view of the paucity of previous records for the Sefton Coast and the nationally declining status of the species, the abundance of H. glabra on the Sefton Coast sand-dunes in 2007 is extraordinary. It should also be noted that extensive areas of potentially suitable habitat on parts of Ainsdale N.N.R. and several golf courses were not visited. As the plant is small and rather shy of flowering, it could have been overlooked in the past. However, the intensity of botanical recording over many years in this area, and the large size of the populations found, make this unlikely. 164 12, Jet, SIMOOEE! TABLE 3. OCCURRENCE AND FREQUENCY OF H. GLABRA AND VASCULAR ASSOCIATES ON 28 SITES Taxon English name Acer pseudoplatanus Achillea millefolium Aesculus hippocastaneum Agrostis capillaris Aira caryophyllea Aira praecox Ammophila arenaria Anagallis arvensis Anchusa vulgaris Anthoxanthum odoratum Anthyllis vulneraria Arenaria serpyllifolia Arrhenatherum elatius Asparagus officinalis Betula pendula Bromus hordeaceus Carex arenaria Centaurea nigra Centaurium erythraea Cerastium fontanum Chamerion angustifolium Cirsium arvense Conyza canadensis Crataegus monogyna Crepis capillaris Cynoglossum officinale Echium vulgare Epipactis dunensis Equisetum arvense Erigeron acer Erodium cicutarium Erodium lebelii Erodium xanaristatum Euphorbia portlandica Festuca ovina Festuca rubra Fragaria vesca Galium verum Geranium molle Holcus lanatus Hypochaeris glabra Hypochaeris radicata Leontodon saxatilis Lotus corniculatus Luzula campestris Oenothera glazioviana Oenothera xfallax Ononis repens Ornithopus perpusillus Phleum arenarium Pilosella officinarum Sycamore Yarrow Horse-chestnut Common Bent Silver Hair-grass Early Hair-grass Marram Scarlet Pimpernel Bugle Sweet Vernal-grass Kidney Vetch Thyme-leaved Sandwort False Oat-grass Garden Asparagus Silver Birch Soft Brome Sand Sedge Common Knapweed Common Centaury Common Mouse-ear Rosebay Willowherb Creeping Thistle Canadian Fleabane ~ Hawthorn Smooth Hawk’ s-beard Hound’ s-tongue Viper’ s-bugloss Dune Helleborine Field Horsetail Blue Fleabane Common Stork’ s-bill Sticky Stork’ s-bill Hybrid Stork’ s-bill Portland Spurge Sheep’s Fescue Red Fescue Wild Strawberry Lady’s Bedstraw Dove’ s-foot Crane’ s-bill Y orkshire-fog Smooth Cat’ s-ear Cat’ s-ear Lesser Hawkbit Common Bird’ s-foot-trefoil Field Wood-rush Large-flowered Evening-primrose Intermediate Evening-primrose Common Restharrow Bird’ s-foot Sand Cat’ s-tail Mouse-ear-hawkweed No. of occurrences Modal frequency — — ih) —" — — NR — NO Re WrRre ONRYTAA MNO wo BRAY BB BRD WD RB RR KN KF OOF ANNAN ADAWNNRF DK NN CONN AD pm pee, SO im} Fh} =n O CO =O COCO CO Sane O OOO fF Om HOOF A OO eA AM OW eB A OW a A OO oO © & HW J eet heel enema aie TABLE 3 continued Taxon Pinus nigra Plantago coronopus Plantago lanceolata Poa annua Poa humilis Polygala vulgaris Populus alba Prunella vulgaris Quercus sp. Ranunculus bulbosus Rosa canina agg. Rubus caesius Rumex acetosella Sagina apetala Salix repens Sedum acre Senecio jacobaea Senecio vulgaris Solanum dulcamara Sonchus asper Sonchus oleraceus Spergula arvensis Taraxacum officinale agg. Trifolium arvense Trifolium campestre Trifolium dubium Trifolium repens Veronica officinalis Vicia lathyroides Viola canina Vulpia bromoides Vulpia fasciculata 84 taxa HYPOCHAERIS GLABRA ON THE SEFTON COAST 165 English name No. of occurrences Modal frequency Austrian Pine Buck’ s-horn Plantain Ribwort Plantain Annual Meadow-grass Spreading Meadow-grass Common Milkwort White Poplar Selfheal Oak Bulbous Buttercup Dog-rose Dewberry Sheep’ s-sorrel Annual Pearlwort Creeping Willow Biting Stonecrop Common Ragwort Groundsel Bittersweet Prickly Sow-thistle Smooth Sow-thistle Corn Spurrey Dandelion Hare’s-foot Clover i Hop Trefoil Lesser Trefoil White Clover Heath Speedwell Spring Vetch Heath Dog-violet Squirreltail Fescue Dune Fescue aA Bp = © Wane AMWiS Hd HS | ws NR S| PN De BRN NRK VN DK KF OR HK WY CO Im} Im) Im} Im} C) C OC TR) lea la Ja) lei ie) In} Iai CO © [3 © ma © my fl th © 's) teh © C) Ap lan Being an annual, this species may be susceptible to adverse environmental conditions, especially at time of germination and establishment of seedlings. Spring and summer 2007 were characterised by wholly exceptional weather conditions, specifically a warm, dry April followed by the wettest early summer (May to July inclusive) since records began. The habitat occupied by H. glabra is particularly susceptible to spring and summer drought and it is possible that 2007 provided ideal conditions for a population explosion. Thus, Rodwell (2000) points out that, in a number of annual species, both viability and germination rate have been shown to decline markedly with decreasing soil moisture, desiccation often arresting establishment. However, Wilson & King (2003) state that germination of H. glabra occurs mainly in the autumn and that seed is relatively short-lived. The inference is that, despite the paucity of recent records, there must have been sufficient plants on the dunes to provide a seed-bank and that the weather conditions of autumn 2006 were important, as well those of spring and summer 2007. Contrary to statements in the literature (e.g. Wilson & King 2003) that H. glabra closes its flowers in the afternoon, it was noted that fully open flowers were usually to be found from about 11.00 to 15.00 hrs BST in sunny or partly cloudy conditions. At these times, the plants were easy to find but, even when closed, inflorescences could readily be recognised by 166 P. A SME the tapering shape and characteristic pattern of purple-edged involucral bracts. There was, however, a remarkable variation in the height of mature plants, from about 2 cm to 30 cm. The spiky pappus-heads were identifiable from a few metres away and often led to the discovery of a colony, which was also assisted by the characteristic appearance of the preferred habitat. Although two of the largest sites contain recently planted pine trees, most of the habitat of H. glabra on the Sefton Coast dunes appears to be self-sustaining, requiring little active management. At most sites, present levels of rabbit-grazing and human trampling seem sufficient to maintain the open, sparsely vegetated communities favoured by the plant. However, the recently noted trend towards denser swards, perhaps influenced by aerial nitrogen deposition (Gateley & Michell 2004), could represent a long-term threat to this species. Thus, Wilson & King (2003) specify increases in nitrogen applied to nutrient-poor soils and agricultural improvement of sandy arable and pasture land as among the reasons for a national decline of H. glabra. ACKNOWLEDGMENTS I am extremely grateful to Pat Lockwood for assistance in the field and to Alice Kimpton for drawing attention to colonies at Formby Golf Course and Ainsdale N.N.R., which I would not otherwise have found, and for transport to the Dune Restoration Area. Steven Cross very kindly looked up old records and references in the World Museum Liverpool library and LIV, while Dave Earl provided records from the New Flora of South Lancashire Database. Eric Greenwood made helpful comments on an early draft of the manuscript. REFERENCES BROWN, R. (1873). Appendix to the Flora of Liverpool. Notes of additions to the Flora of Liverpool. Liverpool Naturalists Field Club. BROWN, R. (1875). Second Appendix to the Flora of Liverpool. Liverpool Naturalists Field Club. CHEFFINGS, C. M. & FARRELL, L. (eds.) (2005). The Vascular Plant Red Data List for Great Britain. Joint Nature Conservation Committee, Peterborough. DICKINSON, J. (1851). The Flora of Liverpool. Deighton & Laughton, Liverpool. GATELEY, P. S. & MICHELL, P. E. (2004). Sand Dune Survey of the Sefton Coast. Report by TEP, Warrington to Sefton Metropolitan Borough Council, Bootle. GREEN, C. T. (ed.) (1933). The Flora of the Liverpool District. Second edition. Buncle & Co., Arbroath. HALL, T. B. (1839). The Flora of Liverpool. Whitaker & Co., London. HALLIDAY, G. (1997). A Flora of Cumbria. Centre for North-West Regional Studies, University of Lancaster. LORD DE TABLEY (1899). The Flora of Cheshire. Longmans, Green & Co., London. MELVILL, J. C. (1909). Hypochoeris glabra in Lancashire. Journal of-Botany 47: 355. NEWTON, A. (1971). Flora of Cheshire. Cheshire Community Council, Chester. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. (eds.) (2002). New Atlas of the British and Irish Flora. Oxford University Press, Oxford. REGIONAL BIODIVERSITY STEERING GROUP (1999). A Biodiversity Audit of North West England. 2 vols. RODWELL, J. S. (ed.) (2000). British Plant Communities volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge. SAVIDGE, J. P., HEYWoob, V. H. & GORDON, V. (eds.) (1963). Travis’s Flora of South Lancashire. Liverpool Botanical Society, Liverpool. SMITH, P. H. (1999). The Sands of Time. An introduction to the sand-dunes of the Sefton Coast. National Museums & Galleries on Merseyside, Liverpool. WHELDON, J. A. (1909). Hypochoeris glabra L. in Lancashire. Journal of Botany 47: 386. WILSON, P. & KING, M. (2003). Arable Plants — a field guide. Wild Guides, Old Basing, Hants. (Accepted February 2008) Watsonia 27: 167—170 (2008) 167 Notes VIPER’S-GRASS SCORZONERA HUMILIS L. AT WAREHAM MEADOWS, DORSET VIPER’ S-GRASS IN BRITAIN Viper’s-grass Scorzonera humilis is restricted to four sites in Britain, in two Dorset and two in South Wales, and is classified as Vulnerable in the Vascular Plant Red List for Great Britain (Cheffings & Farrell 2005). Historically, another three populations were known from England, one in Warwickshire, and two in Dorset. In Warwickshire, five plants were found in ‘a damp, old meadow on a sandy clay soil’ in the 1950s (Hawkes & Phipps 1954). The meadow was an unimproved, but not particularly species-rich, grassland, a habitat that ‘is not apparently specialised and must be mirrored in thousands of places in the British Isles.’ This small population persisted for less than fifteen years after its discovery and it was extinct by 1965 (Perring & Farrell 1983). The two extinct Dorset sites were on a roadside 1 km to the east of Wareham Meadows, and in a wet, grassy field over 11 km away (Hall 1928). The roadside site contained ‘a single fine plant’ according to John Raven in 1972 (record card in Biological Records Centre, Monks Wood), but this plant has since died. Viper’s-grass is still present at Corfe Common in Dorset, where there were four flowers in 2006 (B. Edwards pers. comm.) Two Welsh sites were discovered in 1996 and 1997 (Preston et al. 2002), and at least one of them contains many thousands of plants in ‘several hectares of marshy pasture’ (Woodman & Walls 1999). The remaining English site is at the Royal Society for the Protection of Birds’ Wareham Meadows nature reserve. The earliest record certainly from this locality is from 1950, but it is assumed that this is the same location where the plant was first discovered in Britain in 1914 (Druce 1915). The ‘Scorzonera field’ is a 1 ha plot surrounded by ditches on three sides, with a rushy pasture to the west into which the Viper’s-grass is spreading into. It has been found in some of the adjacent fields in the past (Druce 1916). CONDITION OF THE WAREHAM MEADOWS SITE If this field is the same one that Noel Sandwith found over 90 years ago, it appears to have become more overgrown and enriched. On his visit in 1916, G. C. Druce found Viper’s-grass ‘In great plenty growing in a very wet pasture among Hydrocotyle, Ranunculus flammula, Carex panicea, Carex echinata, Carex flava var. oedocarpa, Carex [nigra], Molinia, Cirsium [dissectum], Potentilla palustris, Juncus articulatus, Juncus acutiflorus, Festuca rubra, Agrostis [stolonifera], and [Danthonia decumbens]. In his note from 1929 he adds Bulbous Foxtail Alopecurus bulbosus (Druce 1929). Purple Moor-grass Molinia caerulea is still frequent, but Soft Rush Juncus effusus is spreading, there are two stands of Common Reed Phragmites australis, and the northern part is dominated by a stand of Greater Pond- sedge Carex riparia. The reed and pond-sedge both spread after changes were made to the water levels on the rest of Wareham Meadows in 1997-1998. Viper’s-grass is now most frequent in the drier, western half of the field, in the more open grassland. In the late 1990s, although there were still thousands of plants in the field, RSPB staff were concerned that the vegetation was becoming too rank for the Viper’s-grass; this concern was also expressed by the authors of the account in the Red Data Book (Woodman & Walls 1999). A small part of the south- western corner of the field, covering approximately 15 m x 20 m, was cut with a hay mower in mid-August 1999, to see if this would benefit the Viper’s-grass by reducing the growth of the surrounding vegetation. The following year, a larger area, covering most of the western half of the field, was hay cut, and the reed and sedge beds were flailed. Prior to these cuts, the field had been fenced off (by the previous tenant) to protect the plants, and it had not been managed for at least five years. In subsequent years, encroaching reed and sedge were cut and removed, and in August 2006, after the Viper’s-grass had finished flowering, cattle were brought in to the field to graze it. 168 NOTES Watsonia 27 (2008) These measures do seem to have created a more open structure, but more information was needed about how the plants were responding. POPULATION STRUCTURE AT WAREHAM MEADOWS In Belgium, many Viper’s-grass populations contain a large number of old plants with very few young ones, and seedling establishment is rare (Colling et al. 2002). This age structure is particularly prevalent in drier sites with taller vegetation and more nutrients. These aged populations may be in danger of extinction if they are unable to replace dying plants with new seedlings. In contrast, there are other populations, in wetter and more nutrient-poor sites, with a much higher proportion of young plants. These regenerating populations may be more viable than the older ones. In July 2002 attempts were made to assess the condition of the population using similar methods to the Belgian study: the number of plants in each of four size classes was recorded in ten randomly sited 1 m Xx | m quadrats. Size is probably closely correlated with age in Viper’s-grass, and it was measured by the number of basal leaf rosettes on each plant (the four classes were: one rosette; 2—5 rosettes; 6— 25 rosettes; and more than 25 rosettes). The results suggested that the structure of the Wareham Meadows’ population’ was somewhere between that of the regenerating and aged populations found in Belgium, but this was a small sample with only 33 plants in the quadrats and very few young seedlings. This indicated that the conditions in the field could be improved by creating a more open structure to the sward. In August 2002, the bridge across the ditch into the Scorzonera field was repaired and cattle were allowed in to graze the site and open up the vegetation. In April 2003 a larer sample was taken using 50 randomly sited quadrats to assess the age structure and map the distribution of the plants. The results (from 195 plants) were very similar to those from 2002 (Fig. 1), with 86% of the plants in the 2—5 and 6-25 rosettes classes. There were few presumed young plants (those with a single rosette), but there were even fewer very old ones (those with more than 25 rosettes) (Table 1). In 2005, the reed and sedge beds were cut again, and cattle grazed the field in 2006. In April 2007 the 2003 survey was repeated using another 50 random quadrats. This time the results (from 178 plants ) showed 87% of plants were in the 2—5 and 6—25 rosette classes with more (56% of all plants) in the 2—5 class (Table 1). This could suggest that there were more younger plants in 2007, in response to the habitat management work. Difficulties in distinguishing individual plants may have led to overestimates of the number of smaller plants. For example, one plant had three single rosettes at least 10 cm apart from each other, but they were all connected by one rhizome. VIPER’ S-GRASS DISTRIBUTION AND THE VEGETATION AT WAREHAM MEADOWS The map (Fig. 2) shows the number of Viper’ s- grass plants found in each quadrat, the position of each quadrat, and the vegetation in the field in April 2007. In both 2003 and 2007, most of the plants were amongst the more open Red Fescue Festuca rubra and Purple Moor-grass Molinia caerulea on the western side of the field. There were also many plants in the stand of other grasses in the south-west corner of the field, but this was largely excluded from the quadrat survey, which used a north-south and east-west grid and so cut off the north-west and south-west corners of the field. This open grassland was the favoured habitat: there were a few plants in the more tussocky grassland, in the open rush beds, and in the edge of the sedge bed along the northern side, but there were none in the middle of the dense rush, TABLE 1. THE NUMBER OF VIPER’S-GRASS SCORZONERA HUMILIS PLANTS IN DIFFERENT SIZE CLASSES IN SAMPLES FROM 2002, 2003, AND 2007 Year Number of rosettes per plant l 2-5 6-25 25+ N plants N quadrats 3002 4 12 16 | 33 10 12% 36% A8% 3% D 3003 17 22 96 10 195 50 9% 37% 49% 5% 007 19 100 56 8 178 50 31% 2% NOTES Watsonia 27 (2008) 169 06 - 05 | m2002 (33) [2003 (195) [2007 (178) Os 4 O24 Proportion of plants 0.1 1 2-5 6-25 Zor Size class (number of rosettes) FIGURE |. The proportion of Viper’s-grass Scorzonera humilis plants in each of the four size classes in 2002 and 2003. Sample sizes were 33 plants in 2002 and 195 plants in 2003. The number of plants surveyed in each year is shown in brackets. Vegetation Dense clumps of Soft Rush Open rush Other grasses Red Fescue and Purple Moor-grass Reed Sedge beds Tussocks of Soft Rush and Purple Moor-grass z L N Z Number of plants @~ @ 12 * 2. FIGURE 2. Vegetation and the number of Viper’s-grass Scorzonera humilis plants in fifty 1 m x 1m quadrats in the field at Wareham Meadows in April 2007. Squares show the position of quadrats where no plants were found. 170 NOTES Watsonia 27 (2008) reed, or sedge beds. A survey in 1997 produced similar results (Clowes 1998). In 2007, when the vegetation was mapped using GPS equipment, the favoured habitat covered just under a quarter (0:21 ha, 24%) of the field, the unsuitable area of rushes, reeds, and sedge beds covered two-thirds (65%, 0-57 ha), and the remaining 10% (0.09 ha) was tussocky Purple Moor-grass and rushes. Extrapolating the mean number of plants per quadrat (12-3) in the Red Fescue and Purple Moor-grass area over the whole area covered by this favoured habitat, gives a population estimate of about 23,000 plants (bootstrap analysis places confidence limits of 15,000- 29,000 on this estimate). HABITAT MANAGEMENT FOR VIPER’ S-GRASS After a gap of many years, the Scorzonera field is being grazed again. It is expected that cattle grazing will open up the tussocky grassland, create more suitable areas for seedlings to establish, and improve conditions for the Viper’s-grass. Open conditions in the adjacent western field are maintained by cutting or grazing, which should aid the spread of the plant and increase its distribution. Viper’s-grass has already spread at least 27 m into this field. The population will be monitored to make sure that grazing 1s having the desired effect. ACKNOWLEDGEMENTS Thanks to Ian Clowes, former warden at Wareham Meadows, for comments and information about the past management of the field, and Chris Preston for supplying me with the details of John Raven’s record in the Biological Records Centre. Bryony Williams and Amber Rosenthal mapped the distribution of Viper’s-grass in the 1997 survey. Bryan Edwards provided details of the Corfe Common population. REFERENCES CHEFFINGS, C. M. & FARRELL, L. (2005). The Vascular Plant Red Data List for Great Britain. http:// www.jncc.gov.uk CLOWES, I. (1998). Survey into the current status of Scorzonera humilis on Wareham Meadows in 1997. Unpublished report, Royal Society for the Protection of Birds, Wareham. COLLING, G., MATTHIES, D. & RECKINGER, C. (2002). Population structure and establishment of the threatened long-lived perennial Scorzonera humilis in relation to environment. Journal of Applied Ecology 39: 310-320. DRUCE, G. C. (1915). Scorzonera humilis L. Report of the Botanical Exchange Club 1915: 202. DRUCE, G. C. (1916). Scorzonera humilis L. Report of the Botanical Exchange Club 1916: 404. DRUCE, G. C. (1929). Scorzonera humilis L. Journal of Botany 67: 26-27. HALL, L. B. (1928). Scorzonera humilis L. Report of the Botanical Exchange Club 1928: 309. HAWKES, J. G. & PHIPPS, J. B. (1954). Scorzonera humilis L. in Warwickshire. Proceedings of the Botanical Society of the British Isles 1: 152-153. PERRING, F. H. & FARRELL, L. (1983). British Red Data Books: I Vascular Plants, 2nd edn. Royal Society for Nature Conservation, Lincoln. PRESTON, C. D., PEARMAN, D. A., and DINES, T. D. (2002). New atlas of the British flora. Oxford University Press, Oxford. WOODMAN, J. & WALLS, R. M. (1999). Scorzonera humilis L. (Asteraceae), in M. J. WIGGINTON ed. (1999). British Red Data Books: I Vascular Plants, 3rd edn. p 338. Joint Nature Conservation Committee, Peterborough. M. GURNEY Royal Society for the Protection of Birds, The Lodge, Sandy, Beds SG19 2DL Watsonia 27: 171—187 (2008) 17] Plant Records Records for publication must be submitted to the appropriate Vice-county Recorder (see BSBI Year Book 2008), and not to the Editors. Following publication of the New Atlas of the British & Trish Flora and the Vice-county Census Catalogue, new criteria have been drawn up for the inclusion of records in Plant Records. (See BSBI News no. 95, January 2004 pp. 10 & 11). These are outlined below: First records of all taxa (species, subspecies and hybrids) included in the VCCC, designated as native, archaeophyte, neophyte or casual. First record since 1970 of the taxa above, except in the case of Rubus, Hieracium and Taraxacum. Records demonstrating the rediscovery of all taxa published as extinct in the VCCC or subsequently. Newly reported definite extinctions. Deletions from the VCCC (e.g. through the discovery of errors, the redetermination of specimens etc.) NB — only those errors affecting VCCC entry. New 10km square records for Rare and Scarce plants, defined as those species in the New Atlas mapped in the British Isles in 100 10km squares or fewer. (See BSBI News no. 95, January 2004 pp. 36-43). Records for the subdivisions of vice-counties will not be treated separately; they must therefore be records for the vice-county as a whole. However, records will be accepted for the major islands in vaccsO2=104" 1110 and 113. In the following list, records are arranged in the order given in the List of Vascular Plants of the British Isles and its supplements by D. H. Kent (1992), from which the species’ numbers, taxonomy and nomenclature are taken. The Ordnance Survey national grid reference follows the habitat and locality. With the exception of collectors’ initials, herbarium abbreviations are those used in British and Irish Herbaria by D. H. Kent & D. E. Allen (1984). Records are field records if no other source is stated. For all records ‘det.’ or ‘conf.’ appear after the herbarium if the determination was based on material already in an institutional herbarium, otherwise before the herbarium. The following signs are used: * before the vice-county number: to indicate a new vice-county record. + before the species number: to indicate that the plant is an archaeophyte. * before the species number: to indicate that the plant is a neophyte. © before the species number: to indicate that the plant is a casual. The above 3 signs may also used before the vice-county number to indicate the status of the plant in that vice-county. ®) before the vice-county number: to indicate that this is an additional hectad for a Rare or Scarce plant. @ at end of entry: established taxon not in Vice-County Census Catalogue. Name of authority provided. [ ] enclosing a previously published record: to indicate that that-record should be deleted or changed. The following list contains the first set of records up to and including the year 2007. The second set of records for this period will be published in the next issue of Watsonia. Huperzia selago subsp. arctica 1/1.1b. *110, Outer Hebrides: moorland, Grianan, Barra, NF694016, R. J. Pankhurst, 2007, det. F. J. Rumsey, E. +Selaginella kraussiana 2/1.2.._ *38, Warks.: one metre square patches in rock gardens and palm house, Birmingham City Botanic Garden, SP0485, J. W. Partridge, 2007, conf. M. W. Poulton, WARMS. Well naturalised. Birmingham & Black Country County Flora Project. *100 (Great Cumbrae), Clyde Is.: lawn behind former Garrison House, Millport Garrison, NS164551, BSBI, 2007. Doubtless a long established escape. 172 PLANT RECORDS Equisetum xdycei (E. fluviatile x E. palustre) 4/1.4x8. *100 (Arran), Clyde Is.: shoreline south of King’s Cave, NR8830, C. N. Page, 1988. Ist record for Arran and for v.c.100. 100 (Bute), Clyde Is.: raised beach flush, White Port, NS109543, A. C. Hannah, 2007. Ist record for Bute. 110, Outer Hebrides: roadside ditch, Seilebost, NG065968, R. J. Pankhurst, 1996, det. H. McHaffie, E. Ist record since 1970. Equisetum xrothmaleri (E. arvense x E. palustre) 4/1.5x8. _*10, Wight: on slumped coastal cliffs, Brambles Chine, Colwell, SZ3388, P. D. Stanley, 2007, det. P. J. Acock, herb. P. D. Stanley. Pilularia globulifera 9/1.1. 8, S. Wilts.: pool, Plaitford, SU274173, R. Veall, 2005. 1st record since 1892. Had been presumed extinct. Ceterach officinarum 15/3.1. | *67, S. Northumb.: on urban wall, North Shields, NZ348686, G. & A. Young, 2006. +Cyrtomium falcatum 17/2.1. *63, S. W. Yorks.: wall near R. Don, Sheffield, SK372885, J. Westfold, Sheffield City Ecology Unit, 2007. Dryopteris affinis subsp. affinis 17/3.3a. *63, S. W. Yorks.: Glen Howe Woods, Stocksbridge, SK2994, Yorks. Fern Group, 2004. Dryopteris affinis subsp. cambrensis 17/3.3b. *63, S. W. Yorks.: Bradfield Dale, Stocksbridge, SK2491, Yorks. Fern Group, 2007. Dryopteris affinis subsp. borreri 17/3.3c. *63, S. W. Yorks.: Glen Howe Woods, Stocksbridge, SK2994, Yorks. Fern Group, 2004. tAzolla filiculoides 19/1.1. © *©100 (Arran), Clyde Is.: pond at Merkland, NSO20383, A. R. Church, 1993. Did not persist. tGinkgo biloba L. 19A/GIN.bil. *63, S. W. Yorks.: one regenerating seedling, Hickleton Hall, Thurnscoe, SE4805, G. D. T. Wilmore, 2001. @ +Larix xmarschlinsii (L. decidua x L. kaempferi) 20/5.1x2. *7, N. Wilts.: four seedlings in woodland, Savernake Forest, Marlborough, SU225655, J. Oliver, 2005. | +Nymphaea marliacea Lat.-Marl. 26/1.mar. *110, Outer Hebrides, lochan opposite Loch an Duin, Northbay, Barra, NF699033, R. J. Pankhurst, 2007, E. @ tHelleborus argutifolius 28/3.arg. *©39, Staffs.: four seedlings in pavement, parents other side of wall, Wombourne, SO872921, C. B. Westall, 2003. *©63, S. W. Yorks.: limestone exposure near High Melton College, SESO01, C. A. Howes, 2007. ONigella damascena 28/5.1. | 113(S), Channel Is. (Sark): roadside verge at foot of bank, east of Port a la Jument House, WV460761, R. M. Veall, 2007, herb. Société Sercquiaise. Self-seeded from nearby garden. Ist record for Sark. tConsolida ajacis 28/7.1. ©63, S. W. Yorks.: waste ground, Longley Park, SK3591, K. Balkow, 2007. Ist record since 1970. ; +tClematis tangutica 28/12.3. *59, S. Lancs.: 13 plants as pavement weeds originating from a nearby garden plant, Salthill Road, Clitheroe, SD7442, D. P. Earl, 2006. tClematis recta L. 28/12.rec. *29, Cambs.: three self-sown plants in road gutters along Greenfields, Newmarket, TL650632, A. C. Leslie, 2007. @ Ranunculus parviflorus 28/13.7. *+85, Fife: several clumps in waste ground, former harbour, Kirkcaldy, NT285921, G. H. Ballantyne, 2007. Ist record for Scotland since c. 1880. Ranunculus baudotii 28/13.21. *100 (Bute), Clyde Is.: in brackish pond, Cradh ruadh, NS087528, A. C. Hannah, 2005. Aquilegia vulgaris 28/16.1. *©110, Outer Hebrides: pavement weed, Stornoway, NB425332, P. A. Smith & M. Teneva, 2007. tBerberis aggregata 29/1.5. *110, Outer Hebrides: low sea cliff, north shore of Loch Boisdale, NF800193, P. A. Smith & M. Teneva, 2004, det. E. J. Clement, E. E. J. Clement describes this as nearest to B. aggregata. Three visits were needed to obtain sufficient material! tBerberis julianae 29/1.7._ *38, Warks.: four or five seedlings on pavement edge, Corncrake Close, Wylde Green, Birmingham, SP126950, J. W. Partridge, 2007, conf. M.W. Poulton, WARMS. Birmingham & Black Country County Flora Project. +Berberis darwinii 29/1.9. *38, Warks.: roadside, Dunchurch, SP4871, M. D. G. Jones, 1960, WARMS. Original record unearthed by Computer Map Flora Team. Seen since. Bird-sown elsewhere. PLANT RECORDS 73 *+Papaver rhoeas 30/1.4. 100 (Great Cumbrae), Clyde Is.: NS15, NCC Freshwater Loch Survey, 1996. Ist record for v.c.100 but not localised. ©100 (Bute), Clyde Is.: disturbed ground, John Street, Rothesay, NS086644, A. C. Hannah, 2006. Ist localised record for v.c.100 and Ist for Bute. +Chelidonium majus 30/5.1. 100 (Arran), Clyde Is.: spreading on to road verges, Lamlash, Whitehouse Wood, NS0O25307, A. McBain, 2005, det. A. W. Smith. tDicentra spectabilis (L.) Lemaire 31/l.spe. *63,S. W. Yorks.: wetland area, Chellow Dene Woods, Bradford, SE132338, B. A. Tregale & M. Wilcox, 2007. @ +Corydalis cheilanthifolia 31/2.3. | *59, S. Lancs.: bare ground, site of former garden centre, Cable Street, Formby, SD301078, P. H. Smith & M. P. Wilcox, 2006, det. M. P. Wilcox. Fumaria officinalis subsp. wirtgenii 31/5.7b. *7, N. Wilts.: farmland, Bromham, ST968648, J. Newton, 2005. Occasional. *+Fumaria densiflora 31/5.8. ®13, W. Sussex: edge of arable, Goodwood, The Trundle, SU862108, M. M. Shaw, 2007. Ulmus xhollandica (U. glabra x U. minor x U. plotii) 33/1.1x3x4. *62, N. E. Yorks.: three trees in woodland, their provenance uncertain, Deepdale Wood, Loftus, NZ717183, V. Jones, 2004, conf. M. Coleman, herb. V. Jones. Ulmus minor subsp. angustifolia 33/1.3b. *£63, S. W. Yorks.: edge of North Beck, Canal Road, Bradford, SE153362, B. K. Byrne, 2007. tNothofagus obliqua 39/2.1. *©7, N. Wilts.: one young sapling among hundreds of Fraxinus excelsior in woodland, Savernake Forest, Marlborough, SU203680, J. Oliver, 2005. +Quercus ilex 39/4.3. *©100 (Bute), Clyde Is.: two small trees in woodland, not obviously planted, Kerrycroy, NS106615, A. C. Hannah, 2005. Betula pubescens subsp. tortuosa 40/1.2b. *38, Warks.: woodland, Sutton Coldfield, SP0998, E. F. Warburg, 1960s, WARMS. Original record unearthed by Computer Map Flora Team. 110, Outer Hebrides: top of sea cliff, Camas Aspaic near Cromor, NB403217, P. A. Smith & M. Teneva, 2007, det. A. O. Chater, E. Ist record since 1970. Alnus xhybrida (Alnus glutinosa x A. incana) 40/2.1x2. *29, Cambs.: one self-sown sapling on waste ground, Barnwell East LNR, Cambridge, TL479583, A. C. Leslie, 2007, CGE. +Alnus cordata x A. incana 40/2.3x2. *29, Cambs.: one coppiced sapling, apparently having arisen spontaneously on waste ground, Barnwell East LNR, Cambridge, TL479583, A. C. Leslie, 2007, CGE. @ +Alnus rubra 40/2.rub. *59, S. Lancs.: disused railway line, Padiham old railway, SD8032, C. S. Crook, 2006. +Carpobrotus edulis 42/8.2. 113(S), Channel Is. (Sark): bank on east side of road, Rue du Fort, WV466765, R. M. Veall, 2007, herb. Société Sercquiaise. Spread from plants in garden. Ist record for Sark. +Chenopodium glaucum 43/1.5. *©7, N. Wilts: nursery weed, Atworth, ST851654, P. Green, 2005. +Chenopodium polyspermum 43/1.8. | *©100 (Bute), Clyde Is.: demolition site, Mansefield Place, Rothesay, NS088642, A. C. Hannah, 2006, det. D. R. McKean, E. ©Chenopodium quinoa Willd. 43/1.qui. *8, S. Wilts.: in tree nursery, Landford, SU252201, P. Woodruffe, 2005. Introduced, naturalising. *85, Kinross, set-aside field, Burleigh, by Loch Leven, NO128037, L. Lavery, 2006, det. W. Hay. Now occasionally planted for wildlife and occurring as a casual. @ Atriplex xgustafssoniana (A. prostrata x A. longipes) 43/3.2x4. | ®13, W. Sussex, bank of River Adur, river side of flood bank, Beeding Brooks (South), TQ190114, A. Spiers, A. G. Knapp, M. M. Shaw & E. J. Clement, 2007, det. A. G. Knapp & A. Spiers. tAtriplex halimus 43/3.9. *59, S. Lancs.: sand-dunes, Hightown, SD298030, P. H. Smith, 2007, herb. D. P. Earl. +Amaranthus hybridus 44/1.2. *©100 (Bute), Clyde Is.: demolition site, Mansefield Place, Rothesay, NS088642, A. C. Hannah, 2006, det. D. R. McKean, E. ©Amaranthus caudatus 44/1.cau. *38, Warks.: five plants on farm dump, Castle Hills, Elmdon, SP177822, A. Morris, 2007, det. J. W. Partridge, WARMS. Self-sown escape. 174 PLANT RECORDS +Portulaca oleracea 45/1.1. *13, W. Sussex: well established along edges of concrete roads in several places within military base, Thorney Island, SU7602, Sussex Botanical Recording Society, det. A. Spiers. Stace’s Flora reports this as casual only but it is clearly fully established here. Minuartia hybrida 46/4.5. ®8, S. Wilts.: several fine plants by tank track in chalk grassland, near Tilshead, SU018487, S. Pilkington, 2005. Stellaria neglecta 46/5.4. 85, Fife: old railway line west of Balwearie Mill, Kirkcaldy, NT260900, G. H. Ballantyne, 2006. First found here in 1985 but not confirmed. Ist record since es M/Z. +Cerastium tomentosum 46/7.3. 113(S), Channel Is. (Sark): on heap of rubble from demolished building above Derrible Bay north of dew-pond, WV472754, R. M. Veall, 2007, herb. Société Sercquiaise. Ist record for Sark since 1957. Myosoton aquaticum 46/8.1. *67, S. Northumb.: Bolam Low House, NZ083815, C. M. Irvine, 2006, herb. G. A. Swan. tPetrorhagia nanteuilli 46/24.1. *38, Warks.: flower bed in garden, Haddon Road, Leamington Spa, SP328667, R. W. B. Dawson, 1966, det. J. G. Hawkes & R. C. Readett, WARMS. Herbarium specimen seen. ©Gypsophila viscosa Murray 46/GYP.vis. *85, Fife, waste ground, Springfield, NO340118, W. Hay & L. Lavery, 2006, det. R.B.G.E. One plant outside a garden but not in it. @ tPersicaria capitata 47/1.cap. *©63, S. W. Yorks.: waste ground, Saltaire, SE1436, M. Wilcox, 2000. *+Polygonum rurivagum 47/4.6. *110, Outer Hebrides: sandy waste ground near Baghasdail, NF732174, P. A. Smith & M. Teneva, 2007, conf. A. O. Chater. E. Rumex xschulzei (R. crispus x R. conglomeratus) 47/8.13x14. | 29, Cambs.: one plant with both parents, north side of shallow pool by Ely North Junction, Ely, TL557810, A. C. Leslie & A. Stevenson, 2007. Ist record since 1951. : tHypericum ‘Hidcote’ 51/1.hid. *©63,S.W. Yorks.: waste ground near car park, Cusworth Hall, SE5403, G. D. T. Wilmore, 2004. @ tHypericum olympicum L. 51/l.oly. *38, Warks.: seedlings on pavement edge, Maney Hill Road, Wylde Green, Birmingham, SP126942, M. W. Poulton, 2007, conf. J. W. Partridge, WARMS. Birmingham & Black Country County Flora Project. © tTilia ‘Petiolaris’ DC. 52/l.pet. *63,S. W. Yorks.: waste ground, Keresforth Hill, Barnsley, SE3305, P. Middleton, 2007. © ¢Tilia tomentosa Moench 52/l.tom. *63,S.W. Yorks.: roadside verge, Fenwick, SE5916, G. D. T. Wilmore, 2007. @ Malva moschata 53/1.1. +100 (Bute), Clyde Is.: Ardbeg shore, NS081671, A. C. Hannah, 2005. ©Althaea cannabina L. 53/3.can. *29, Cambs.: a single self-sown plant on abandoned allotments, Crane’s Lane, Kingston, TL345549, P. J. Reynolds, 2004. © Drosera xobovata (D. rotundifolia x D. anglica) 55/1.1x2. *67, S. Northumb.: Hummell Knowe, NY704715, A. J. Richards et al., 2007, herb. G. A. Swan. Ist localised record. Viola xbavarica (V. riviniana x V. reichenbachiana) 57/1.4x5. 38, Warks.: seven or eight plants with parents nearby in front garden path, Russell Terrace, Leamington Spa, SP324654, J. W. Partridge, 2007, conf. M. W. C. Hardman, WARMS. Ist record since 1970. See BSBI News no.106 p. 8. *Viola xwittrockiana (V. lutea x V. tricolor x V. altaica) 57/1.11x12xalt. | *38, Warks.: ten plants on soil dump by roadside, Castle Vale, Birmingham, SP142891, J. W. Partridge, 2007, conf. E. J. Clement, WARMS. Self-sown. *©62, N. E. Yorks.: one plant in wasteland, Battersby Junction, NZ588072, V. Jones, 1993. Populus trichocarpa 61/1.4._ *67, S. Northumb.: Scotch Gill, Morpeth, NZ1886, G. & A. Young, 2007. Determined from photos by G. Young and A. J. Richards. *Populus balsamifera 61/1.bal. *©7, N. Wilts.: suckering at field edges, Great Bedwyn, SU280643, J. Oliver, 2005. Salix xfruticosa (S. viminalis x S. aurita) 61/2.9x12. *85, Fife: field edge near old moss, north side of Star of Markinch, NO313035, L. Tucker, 2006. Grown on in 2007 to confirm identity. PLANT RECORDS 175 Salix caprea subsp. sphacelata 61/2.10b. 67, S. Northumb.: Long Rack, NY909731, M. E. Braithwaite, 2007, det. R. D. Meikle. 2nd published record. Salix xstrepida (S. cinerea x S. myrsinifolia) 61/2.11x14. 67, S. Northumb.: River Derwent, Blanchland, NY9650, Q. Groom, 2007, det. R. D. Meikle. Ist record since 1959. tSisymbrium strictissimum 62/1.1. *38, Warks.: one plant in abandoned allotment, St. Mary’s Allotments, Leamington Spa, SP332654, J. W. Partridge, 2006, conf. T. C. G. Rich, WARMS. Originally cultivated? +Sisymbrium irio 62/1.2. *©13, W. Sussex: cracks in concrete at car park & gutters, Churchill Square car park, Brighton, TQ305040, A. Spiers, 2007. tBunias orientalis 62/6.1. ©63, S. W. Yorks.: edge of arable field adjacent to M1, SE2911, P. Middleton, 2007. Ist record since 1970. +Cardamine corymbosa 62/14.cor. *©7, N. Wilts.: nursery weed, Atworth, ST851654, P. Green, 2005. *©59, S. Lancs.: bare ground, Lydiate, SD3604, R. F. Goodchild, 2006. Garden escape from a nearby nursery. Erophila glabrescens 62/22.3. 63, S. W. Yorks.: weed on waste ground, Jacobs Well, Bradford, SE1633, M. Wilcox, 2007. Ist record since 1970. +Thlaspi arvense 62/28.1. 100 (Bute), Clyde Is.: disturbed field edge at The Plan (turning circle), NS092531, A. C. Hannah, 2007. Ist record since 1970. ©Lepidium bonariense 62/30.bon. 29, Cambs.: about a dozen plants on gravely ground used for parking farm machinery, north-east of Clayway Farm, Padnal Fen, Ely, TL575832, A. C. Leslie, 2007, conf. E. J. Clement, CGE. 2nd record and Ist since 1961. +Hirschfeldia incana 62/38.1. 67, S. Northumb.: Skinnerburn Road, Newcastle, NZ242634, G. & A. Young, 2002, herb. G. A. Swan. Ist published record. Crambe maritima 62/41.1. *67, S. Northumb.: Whitley Bay, NZ37, B. W. Fox, 1948, Bolton Museum. Ist record. Not seen since. Raphanus raphanistrum subsp. maritimus 62/42.1b. | *©63, S. W. Yorks.: car park, waste ground, Vicar Lane, Bradford, SE166328, B. A. Tregale, 2007. ©Biscutella auriculata L. 62/BIS.aur. *38, Warks.: allotment, Leamington Spa, SP330657, J. W. Partridge, 1990, conf. T. C. G. Rich. © ©Eruca vesicaria subsp. sativa (L.) Cav. 62/ERC.ves.a. 59, S. Lancs.: pavement, King Street, Clitheroe, SD742419, M. P. Wilcox, 2006. Ist record since 1970. @ +Gaultheria shallon 65/8.1. ©8, S. Wilts.: Black Lane, near Redlynch, SU224194, D. J. Wood, 2005. Ist record since 1970. Pyrola minor 66/1.1. ®13, W. Sussex: large colony at edge of ride and in adjacent woodland, Graffham Common, SU932194, B. Middleton, 2006, conf. A. G. Knapp. tLysimachia ciliata 69/4.4. *60, W. Lancs.: rough ground, Carleton, $D328401, M. Sutcliffe, 2007. Lysimachia thyrsiflora 69/4.7. *t13, W. Sussex: Woodcote Pond, Ifold, TQ024311, T. Greenaway, 2007, det. E. J. Clement, herb. E. J. Clement. Not deliberately introduced. tLysimachia verticillaris Spreng. 69/4.ver. *©63, S. W. Yorks.: lay-by, north side of road near Rossington, SK643971, G. Coles, 2006. See Watsonia 22: 279-281 (1999). @ Anagallis minima 69/6.3. 85, Fife: forest track, Tentsmuir, NO4927, P. Gaff, 2007. Ist record since c.1890 (also at this site). tPhiladelphus xvirginalis (P. coronarius x P. microphyllus x P. _ pubescens) 71/1.1xmicxpub. *38, Warks.: dump, Parkhall Wood, Minworth, SP163908, J. W. Partridge, 2006, conf. E. J. Clement, WARMS. Birmingham & Black Country Flora Project. tEscallonia xlangleyensis (KE. macrantha x E. virgata (Ruiz & Pav.) Pers.) 72/1.1xvir. *38, Warks.: dump, Hook Drive, Sutton Coldfield, SP106994, J. W. Partridge, 2006, conf. E. J. Clement, WARMS. Birmingham & Black Country Flora Project. tRibes sanguineum 72/2.4. ©100 (Bute), Clyde Is.: Cnocnicoll Wood, NS1060, A. C. Hannah, 2004. Probably bird-sown. Crassula tillaea 73/1.1. 13, W. Sussex: disturbed sandy paths, Chapel Common, SU821285, M. Edwards, 2006, conf. A. G. Knapp. Had been considered extinct in county. @®13, W. Sussex: in reasonable quantity on car parking area, Littlehampton, TQ029015, M. Berry, 2007, conf. A. Spiers. 176 PLANT RECORDS +Sedum praealtum 73/5.2. *39, Staffs.: four or five plants in shade on footpath verge, Wordsley, SO884877, C. B. Westall, 2007. «Sedum ‘Herbstfreude’ (S. spectabile x S. telephium) 73/5.4x5. *©62, N. E. Yorks.: base of wall, Swainby, NZ475022, V. Jones & W. A. Thompson, 2006. *©63, S. W. Yorks:.: Ainsbury Avenue Tip, Thackley, SE1738, B. A. Tregale & M. Wilcox, 2006. xSedum nicaeense 73/5.9. *29, Cambs.: two plants on flat garage roof of Wychwood, Crane’s Lane, Kingston, TL345549, P. J. Reynolds, 2007. «Sedum sexangulare 73/5.13. *13, W. Sussex: on graves, Midhurst Cemetery, SU876215, Sussex Botanical Recording Society, 2007, det. A. G. Knapp. Apmale 1917 in Southgate, now in London. He told me that he did not have a very happy childhood, either at home or as a boarder to Brentwood School, which he did not like, not enjoying either academic work or any sports. Most of his holi- days he spent not at home but with his paternal grandparents in Poole. These were, he said, idyllic periods at a time before motor vehicles all, at his retirement in 1977 Wye. t l Peter OBITUARIES 205 were common (though his father had a large car), and fond memories of the place were the reason for Peter and Joan’s final move there in 1994. After school, with few qualifications, he briefly joined his father’s accountancy business, but he did not take to that and in 1935 he took examinations and passed into the Scientific Civil Service as a chemist (chemistry being his favourite school subject). He was stationed at Woolwich Arsenal, specialising in petroleum products, but on the outbreak of war he was moved to an [.C.J. laboratory in Dumfries, returning to Woolwich at the end of the war. In 1946 he was seconded to Risley, Cheshire, where he took charge of the oil laboratory. It was there that he met Joan, a paint specialist, sharing with her a love of rambling in the northern hills and dales. They were posted back to Woolwich a few years later, and they married soon after in 1950, setting up home in Erith. Other secondments to the British Army and N.A.T.O. saw Peter working in Chessington, Paris and Brussels, and visiting further afield in Hong Kong. Joan retired in 1969, and Peter in 1977, aged 60, spending his last three years working in the Admiralty Oil Laboratory in Cobham, Kent. Within two years they moved to Monmouth, where they spent 15 happy years gardening, visiting National Trust properties (especially those with good gardens), and taking part in botanical excursions. Peter was always very interested in words, and they became expert crossword solvers, winning several national competions in the broadsheet newspapers and The Listener. Their library included many dictionaries and related works, including a forbidding series called Difficult Crosswords. Late in life Peter also took up Sudoku. He enjoyed watching sport, especially horse-racing, show-jumping and rugby. Enthusiasm gained from reading J. E. (Ted) Lousley’s Wild Flowers of Chalk and Limestone, and then a chance meeting with the man himself in the field at Box Hill, Surrey, per- suaded them to join the B.S.B.I. in 1952. In the same year they joined the London Natural History Society, in which they became especially friendly with E.B. (Ted, or Bang) Bangerter, with whom they continued in correspondence after his emigration to New Zealand until his death in 2001. Peter became the Botanical Meetings Secretary of the L.N.H.S., taking pride in arranging trips to benefit from the cheap excursions offered by British Rail. One such that I attended was to the Lewes downs and levels in 1961. Peter and Joan were among the founder members of the Kent Bela @iubinr lossy in 1979 W. i. Stearn sponsored Peter for election to Fellowship of the Linnean Society. Peter served as a B.S.B.I. Council member from 1956 to 1959, as a member of Meetings Committee from 1955 to 1968, as a member of Records Committee from 1966 (when it comm- enced) to 1980, and, most importantly, as Honorary Field Secretary from 1956 to 1967. This last period took the Society through its first Atlas project and beyond. The post involved much effort in organising and liasing, two of Peter’s strong points, and he claimed to have taken part in every field excursion of the B.S.B.I. in England and Wales while the Atlas data were being gathered. Additionally, he helped Francis Rose and Paul Jovet to organise a highly successful trip to France in July 1959, centred on Boulogne and Abbeville, our first official foreign excursion. In recognition of all his hard work for the B.S.B.I. he was elected an Honorary Member in 1967. Peter and Joan became expert field botanists without ever forming an herbarium, and many less experienced people, myself included, benefited greatly from their help so willingly and graciously given. As mentioned above, they were very much a pair. Joan was very happy to remain more in the background, but she was undoubtedly the brains and driving force and she derived much pleasure from seeing Peter in the limelight. In the 1970s and 1980s they made trips to many botanical hotspots on the Continent, particularly Spain, the Alps, Jugoslavia and Crete, for which Peter made extremely detailed preparatory notes and Left to right, Brenda Howitt, Leaver Howit and Peter Hall 206 OBITUARIES Joan wrote extensive diary-style reports. Peter’s most tangible contribution to botany was Sussex Plant Atlas (1980), completed just before they left the south-east for Monmouth and summarising the work of the Sussex Flora Committee from 1966 to 1978. His convivial nature and friendly smiling face, however, are more vivid in the minds of many of his friends and companions. Peter was known by all as a jovial character with a keen sense of humour and a thirst for knowledge, and he mixed well with everybody. Two of the many friends that Joan and Peter made during the period of Atlas recording were R. C. L. (Leaver) and Brenda Howitt, authors of A Flora of Nottinghamshire (1963). Their obituaries can be found in Watsonia 15: 417-— 419 and 14: 291 respectively, but there were no accompanying photographs. The larger photo- graph reproduced here shows Brenda, Leaver and Peter at a Scottish Field Meeting in 1955, planning the day’s Atlas recording. The image of Joan (Watsonia 25: 442, 2005) used for her obituary was taken from the same group photo- graph (wrongly dated there as “‘around 1960”). We regret the passing of yet another of the B.S.B.I. ‘old guard’, those who were instru- mental in the Society winning its spurs and moving into its ‘golden age’. Neither Peter nor Joan had any siblings, and they had no children. Peter himself provided me with many of the above details. I am also grateful to David Allen for reading a draft and for supplying a number of additional historical points. CLIVE STACE INSTRUCTIONS TO CONTRIBUTORS Scope: Authors are invited to submit Papers and Notes concerning British and Irish vascular plants, their taxonomy, biosystematics, ecology, distribution and conservation, as well as topics of a more general or historical nature. Authors should consult the Hon. Receiving Editor for advice on suitability or any other matter relating to submission of manuscripts. This is a peer-reviewed journal, each paper is handled by one of the Editorial Panel who will usually seek comments from an appropriate referee before acceptance. A list of referees for each volume will be published in the appropriate index. Papers and Notes must be submitted in duplicate, typewritten on one side of the paper, with wide margins and double-spaced throughout. Pages should be numbered. Submission of final edited copy on computer disc will be requested, but two hard copies of the text are acceptable if computer facilities are not available. Format should follow that used in recent issues of Watsonia. Underline where italics are required. Names of periodicals should be given in full, and herbaria abbreviated as in British and Irish herbaria (Kent & Allen 1984). The Latin names and English names of plants should follow the New Flora of the British Isles (Stace 1997). Further details on format can be obtained from the Hon. Receiving Editor or by viewing the website at: http://www.bsbi.org.uk/new_style_manual.htm Tables, figure legends & appendices should be typed on separate sheets and attached at the end of the typescript. Figures should be drawn in black ink or be laser-printed and identified in pencil on the back with their number and the author’s name. They should be no more than three times final size, bearing in mind they will normally be reduced to occupy the full width of a page. Scale-bars are essential on plant illustrations and maps. Lettering should be of high quality and may be done in pencil and left to the printer. Black and white photographs can be accepted if they assist in the understanding of the article. If you are able to submit figures on disc please contact the Receiving Editor to check they are in a suitable format. Contributors must sign a copyright declaration prior to publication which assigns the copyright of their material to the Botanical Society of the British Isles. Twenty-five offprints are given free to authors of Papers and Notes; further copies may be purchased in multiples of 25 at the current price. The Society takes no responsibility for the views expressed by authors of Papers, Notes, Book Reviews or Obituaries. Submission of manuscripts Papers and Notes: Mr M. N. Sanford, c/o The Museum, High Street, Ipswich, Suffolk, IP] 3QH. Books for Review: John Edmondson, National Museums Liverpool, William Brown Street, Liverpool L3 8EN. Plant Records: the appropriate vice-county recorder, who should then send them to Mike Porter, 5 West Avenue, Wigton, Cumbria CA7 9LG. Obituaries: Mrs M. Briggs, 9 Arun Prospect, Pulborough, West Sussex, RH20 IAL. Back issues of Watsonia Available from BSBI Hon. Membership Secretary, Gwynn Ellis, 41 Marlborough Road, Roath, Cardiff, CF2 5BU. E-mail: rgellis@ntlworld.com Tel & Fax 029 2049 6042 Printed in the United Kingdom by Henry Ling Limited, at the Dorset Press, Dorchester, DT1 1HD on Chromomat and Horizon stock manufactured from fibre sourced from certified, environmentally managed forests, at mills with ISO 14001 certification. Printed using VOC free, vegetable based inks. Henry Ling Ltd is fully accredited with ISO 14001. Watsonia August 2008 Volume twenty seven Part two Contents COWAN, R. S., SMITH, R. J., FAY, M. F. & RICH, T. C. G. Genetic variation in Irish Whitebeam, Sorbus hibernica E. F. Warb. oe and its sein toa Sorbus from the Menai Strait, North Wales SAWTSCHUK, J. & RICH, T. C. G. Conservation of Britain’ S eee status of the two Wye Valley endemics Hieracium pachyphylloides, Carboniferous Hawkweed and H. vagicola, Tutshill Hawkweed (Asteraceae) RICH, T. C. G. AND McCosH, D. J. The status of Hieracium arranense and H. sannoxense ns two endemic hawkweeds from the Isle of oe Scotland... LANSDOWN, R. V. Red Decree ee eT ie andor new to Britain FOLEY, M. J. Y. John Fitz-Roberts: A little-known seventeenth century botanist ... PEARMAN, D. A. The status of Coral-necklace J//ecebrum verticillatum L. (Caryophyllaceae) i in Great Britain SMITH, P. H. Corynephorus canescens (L.) P. Beauv. (Grey Hair-gass) on ée Sefton Coast, Merseyside (v.c. 59) SMITH, P. H. Population ie of Rue slabra L. on the Schon ee Merseyside in 2007 Bae ; gO) een ellie et ae NOTES Gurney, M. Viper’s-grass Scorzonera humilis L. at Wareham Meadows, Dorset... PLANT RECORDS BOOK REVIEWS OBITUARIES SMI ITHSONIA ISSN 0043-1532 Typeset by D.K. & M.N. SANFORD © : Printed in Great Britain by HENRY LING LIMITED, THE DORSET PRESS, DORCHESTER, DORSET DT1 1HD AKAM 99-108 109-118 119-126 127-130 131-141 143-148 149-157 159-166 167-170 171-187 189-198 199-206 AK | W447 ee ON! A Botanical Society of the British Isles Volume 27 Part 3 February 2009 Editors: M. Briggs, J. Edmondson, M. J. Y. Foley, D. L. Kelly, M. S. Porter, M. N. Sanford Botanical Society of the British Isles Registered Charity Number: 212560 Applications for membership should be addressed to the Hon. General Secretary, c/o Department of Botany, The Natural History Museum, Cromwell Road, London, SW7 5BD, from whom copies of the Society’s Prospectus may be obtained. Officers for 2008-2009 President, Mr M. E. Braithwaite Vice-Presidents, Miss L. Farrell, Mr R. G. Woods, Dr F. J. Rumsey, Mrs J. M. Croft Honorary General Secretary, Mr D. A. Pearman Honorary Treasurer, Mr A. M. Nixon Editors of Watsonia Papers and Notes, M. J. Y. Foley, D. L. Kelly, M. N. Sanford* Book Reviews, J. Edmondson Plant Records, M. S. Porter Obituaries, M. Briggs “Receiving editor, to whom all MSS should be sent (see inside back cover). © 2009 Botanical Society of the British Isles The Society takes no responsibility for the views expressed by authors of Papers, Notes, Book Reviews or Obituaries. The cover illustration of Primula scotica Hook. (Scottish Primrose) was drawn by Rosemary Wise. Watsonia 27: 207—216 (2009) 207 Some new British and Irish Sorbus L. taxa (Rosaceae) iC, G, RICE Department of Biodiversity & Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP and M. C. F. PROCTOR School of Biosciences, University of Exeter, Stocker Road, Exeter EX4 4QD ABSTRACT Nine new Sorbus taxa are described from Britain and Ireland: S$. admonitor M. C. F. Proctor, sp. nov., S. cambrensis M. C. F. Proctor, sp. nov., S. cuneifolia ia CuGikichusp: nov. S. eminentiforms LC. G. Rich, sp. nov., S. margaretae M. C. F. Proctor, sp. nove amore: I. @. G, Rich, hybr: nov, ‘S. scannelliana T. C. G. Rich, sp. nov., S. stenophylla M. C. F. Proctor, sp. nov. and S. stirtoniana T. C. G. Rich, sp. nov. KEYWORDS: agamospecies, Britain, Ireland, isozymes, whitebeam. INTRODUCTION Some new Sorbus L. (Rosaceae) names are required for taxa to be included in an identification handbook (T. C. G. Rich ef al., in preparation). Some taxa, such as the form of S. anglica Hedl. from Llangollen with narrower cuneate bases to the leaves or the more deeply lobed form of S. devoniensis E. F. Warb., have been known for many years, whilst other taxa have only recently been discovered. The new taxa are described below. METHODS Broad leaves from the short, vegetative shoots in sunlit situations, excluding the oldest and youngest leaf (Aas et al. 1994), were measured on herbarium material in NMW. Lobing was measured perpendicular to the main vein at the centre of the leaf. The numbers of individual veins or pinnae are described rather than number of pairs as the veins or pinnae are usually alternate or subopposite rather than being strictly opposite. Fruits were measured *e-mail: tim.rich@museumwales.ac.uk on fresh material, and the colours matched against the Royal Horticultural Society colour charts (Royal Horticultural Society 1966). Conservation status assessments were made for the new species following the I.U.C.N. (2001) guidelines. These are largely based on known population sizes or estimates as there are no data on population trends. NEWS SPECIES DESCRIPTIONS Sorbus admonitor M. C. F. Proctor, sp. nov. IOLOINADUS2 laity imee ADoOVve “Nene, Watersmeet, v.c. 4 North Devon, England, SS744490, 10 October 2007, T. Rich & D. Cann (NMW, accession no. V.2007.1.225; IsoTyPI: BM, CGE). Diagnosis: S. devoniensi E. F. Warb. similis sed foliis in centro laminae 10—23(—25) partibus centesimis ad costam profundius lobatis (non, ut in S. devoniensi, 6—18(—23) partibus centesimis) differt. Similar to S. devoniensis E. F. Warb. but differing in having leaves more deeply lobed 10—23(—25)% of the way to the midrib at the centre of the lamina (not 6—18(—23)% as in S. devoniensis). Sorbus admonitor 1s a member of the S. latifolia group, and is characterised by the ovate to obovate, shallowly but acutely lobed leaves 1-2—1-:5 times as long as wide and cut 10-23% of way to the midrib with broadly rounded to broadly cuneate bases and greenish- white undersides, and by the large brownish berries. Compared to S. devoniensis the leaves are also glossier. Figure | shows typical leaves of S. admonitor and S. devoniensis. 208 T. C. G. RICH AND M. C. F. PROCTOR 000g FIGURE |. Typical leaves of $. admonitor (A, B) and S. devoniensis (C, D). Scale bar | cm. Sorbus admonitor has been known for many years as the large tree at Watersmeet car park, Devon with a ‘No Parking’ notice nailed to it, which had leaves more deeply lobed than the widespread form of S. devoniensis (Martin & Fraser 1939; Sell 1989). Studies of isozymes by Proctor et al. (1989) showed that S. admonitor differs consistently from S. devon- iensis. The specific epithet ‘admonitor’ (1.e. to admonish) was first coined by E. F. Warburg as a jocular reference to the ‘No Parking’ notice. Sorbus admonitor is endemic to the Watersmeet area where there are at least 108 trees in the East Lyn Valley and two trees nearby above Sillery Sands, Lynmouth (Rich & Cann 2009). Sorbus devoniensis sensu stricto does not grow in the East Lyn Valley area and the two taxa have not been confirmed growing together. Sorbus admonitor differs from S. subcuneata Wilm., with which it does grow, in having broader leaves with rounded bases, and a denser crown. The [.U.C.N. (2001) conservation assessment is ‘Endangered’ based on the Rich & Cann (2009) survey data. Sorbus cambrensis M. C. F. Proctor, sp. nov. HOLOTYPUS: face of railway cutting cliff, Cwm Clydachy SO213120 so uly e200 in CanG: Rich, R. Preece & G. Motley (NMW, accession no. V.2001.25.74a). DESCRIPTION Frutex vel arbor parva ad 8 m. Cortex truncorum majorum griseibrunneus. Gemmae lanceoloideae, acutae, pilis albis vestitae. Folia lata brachyblastorum 6:0—11-0(—12°5) x (4-0-) 4-5—8-0(—8:5) cm, obovata, 1-3—1-6plo longiora quam latiora, ad longitudinis suae 53-67 partes centesimas latissima, apice obtuso, basi cuneata (angulo inter marginem et costam 34— 54(—58)°), sine lobis vel lobis non profundis secta, marginibus biserratis, dentes plerumque exstantes ferentibus sed in 1-3 cm infimis prope petiolum fere integris, venis 18—24, in centro laminae ad angulum (28—)31-41(-48)° a costa tentis; pagina superior atro-viridis; pagina inferior griseialba tomentosa. Inflorescentiae ad 8 cm diametro, aggregatae, tholiformes, ramulis sparsim tomentosis instructae. Sepala triangularia vel deltata, viridia vel tomentosa. Petala 7-85 x 5-6 mm, elliptica, alba. Antherae cremeae vel cremeae roseitinctae. Styli” 2, plus minusve’ = pen aa dimarciinnn longitudinis conjuncti, ad basin pilosi. Fructus maximi 10-5—12:5 x 11-5-13 mm, 0-8—1-Oplo longiores quam latiores, saepe aspectu globoso, in maturitate saturate coccinei, lenticellarum mediocrium numero modico praediti. Shrub or small tree to 8 m or more tall. Bark of larger trunks greyish-brown. Buds lanceoloid, acute, clothed with white hairs. Broad leaves of short sterile shoots 6:0—11-0(-12:5) x (4:0-) 4-5-8-0(—8:5) cm, obovate, 1:3—1-6 times as long as wide and widest 53-67% of the way along the leaf length, with the apex obtuse and the base cuneate (with the angle between the margin and the midrib 34—-54(—58)°), unlobed or shallowly lobed, with the margins biserrate with teeth usually directed outwards but nearly entire in the lowest 1-3 cm near the petiole, NEW SORBUS TAXA 209 VOUS FIGURE 2. Typical leaves of S. cuneifolia (A, B) and S. anglica (C, D). Scale bar 1 cm. with 18—24 veins, held at an angle of (28—)31— 41(—48)° to the midrib at the centre of the lamina; upper surface dark green; lower surface greyish-white tomentose. Inflorescences to 8 cm in diameter, crowded, domed, with branchlets sparsely tomentose. Sepals triangular to deltate, green or tomentose. Petals 7-8:5 x 5—6 mm, elliptic, white. Anthers cream or cream with a hint of pink. Styles 2, joined for about half their length, pilose at the base. Largest fruits 10-5—12:5 x 11:-5-13 mm, 0-8— 1:0 times as long as wide, often looking globose, deep red at maturity (R.H.S. colour chart 33A, 42A, 44A, 45A, 46A), with a moderate number of medium lenticels. Sorbus cambrensis is a member of the S. porrigentiformis E. F. Warb. group within the S. aria (L.) Crantz group, and is characterised by the obovate, biserrate leaves which are greyish-white underneath and the large sub- globose fruits. Studies of S. porrigentiformis by Proctor & Groenhof (1992) demonstrated that trees from Coed Pantydarren (cited as Darren Disgwylfa), Craig-y-Cilau and other crags in the Llang- attock area had a different isozyme profile to the widespread lowland form of 5S. porrigentiformis, and also differed in leaf toothing and fruit size and shape. They also noted that plants at Tarren yr Esgob differed slightly from S. cambrensis, and these are described separately below as S. stenophylla. Sorbus cambrensis is endemic to v.c. 42 Breconshire, Wales. It has been recorded - scattered widely within Cwm Clydach, at - Blackrock, at Craig-y-Cilau, and at Coed Pantydarren. The total population is probably about 100 plants. Sorbus cambrensis is very similar to S. porrigentiformis and S. stenophylla. It tends to grow into a larger tree than S. porrigentiformis with usually more robust shoots, has leaves somewhat more oblong in outline and biserrate, and has larger, subglobose fruits (length: width ratio 0-85—0-95) which are brighter red. Sorbus stenophylla has narrower leaves with a narrower cuneate base and deeper biserrate toothing. Morphologically the leaves are similar to those of S$. leptophylla E. F. Warb. but that species has fruits longer than wide. For leaf silhouettes of S. cambrensis, see Figures 5j-l of Proctor & Groenhof (1992). The provisional I.U.C.N. (2001) ervation assessment is ‘Endangered’. cons- Sorbus cuneifolia T. C. G. Rich, sp. nov. HOLOTYPUS: Creigiau Eglwyseg (Cefn Fedw), Llangollen, v.c. 50 Denbighshire, Wales, 10 July 2002, T. C. G. Rich & S. O. Hand (NMW, accession no. V.2002.17.161). Diagnosis: S. anglicae Hedl. similis sed foltis basi angustius cuneatis, angulo inter marginem et costam (29—)32—44(46)° (non, ut in S. anglica, foliis basi late cuneatis, angulo inter marginem et costam (38—)40—59(—65)°) differt. Similar to S. anglica Hedl. but differing in having leaves more narrowly cuneate at the base with the angle between the margin and the midrib (29—)32—44(-46)° (not broadly cuneate with the angle between the margin and the midrib (38—)40—59(—65)° as in S. anglica). Sorbus cuneifolia is a member of the S. anglica/S. intermedia group, and is charact- erised by the elliptic to obovate leaves with distinct obtuse lobes cut up to about a quarter 210 T. C.G, RICH AND WIE C" EY PROCTOR to one third of the way to the midrib; they are greyish-white tomentose beneath with ‘feathery’ intermediate veins between the main veins, and have cuneate leaf bases. Sorbus cuneifolia was first recognised as distinct by A. J. Wilmott and E. F. Warburg who used a working epithet ‘castelli’ after its occurrence on the ruined walls of Castell Dinas but they never published it. The epithet cuneifolia refers to the narrower leaf bases, its distinguishing feature from the closely related S. anglica. Figure 2 shows typical leaves of S. cuneifolia and S. anglica. Sorbus cuneifolia is a rare tree confined to the west-facing Carboniferous Limestone crags of Eglwyseg Mountain north of Llangollen, Denbighshire (v.c. 50), Wales. About 240 plants are known (Morgan 1987, Woods 1998), and its J.U.C.N. (2001) conservation assess- ment is thus ‘Endangered’. Sorbus eminentiformis T. C. G. Rich, sp. nov. HOLOTYPUS: coppiced tree in woodland, Seven Sisters, Great Doward, v.c. 36 Herefordshire, England, SO546153, 23 September 2002, T. C. G. Rich (NMW,, accession no. V.2002.17.150). Diagnosis: S. eminenti E. F. Warb. similis sed foliis subrhombeis ad basin late cuneatis (non, ut in §. eminente, orbicularibus ad_ basin rotundatis) differt. Similar to S$. eminens E. F. Warb. but differing in having subrhombic leaves broadly cuneate at the base (not orbicular and rounded at the base as in S. eminens). Sorbus eminentiformis is a member of the S. aria group and is characterised by the obovate- orbicular to subrhombic leaves 1-0—1-3 times as long as wide with a broadly cuneate base (the lowest part of the leaf looking triangular) and regular, neat, uniserrate, acuminate teeth, and the broad, red fruits. When Warburg (1957) described S. eminens from Offa’s Dyke, Tidenham in the Wye Valley, he noted that plants from around Symonds Yat differed from the form he described from the Wye Valley and Avon Gorge in having subrhombic leaves with rather deeper teeth. Molecular studies by Proctor & Groenhof (1992), Chester et al. (2007) and Cowan ef al. (2008) have confirmed that they differ from each other. For leaf silhouettes of S. eminentiformis see Figures 4a—b of Proctor & Groenhof (1992), and Figures 4c-f for S. eminens. Sorbus eminentiformis 1s endemic to the Lower Wye Valley in England and Wales, where it occurs in v.c. 34 West Gloucester- shire, v.c. 35 Monmouthshire and v.c. 36 Herefordshire (Rich et al. 2009). The total population is probably under 100 trees, the bulk of which occur on the Great Doward. The I.U.C.N. (2001) conservation assessment is ‘Endangered’. Sorbus margaretae M. C. F. Proctor, sp. nov. HOLOTYPUS: Desolate, v.c. 4 North Devon, SS7849, 31 May 1997, ME {Grae rrocton (NMW, accession no. V.2003.19.388). Diagnosis: S. vexanti E. F. Warb. similis sed fructibus latis atrorubrioribus latioribus quam longioribus differt. Similar to S. vexans E. F. Warb. but differing in having broad, darker red fruits that are wider than long. Sorbus margaretae is a member of the S. aria group, and is characterised by the unlobed, obovate leaves with biserrate teeth which are densely white-tomentose underneath and the red, very broad fruits. Sorbus margaretae was first recognised as distinct by Margaret E. Bradshaw during her surveys of the rare whitebeams of South-west England in 1984, and it is named after her. It was provisionally named ‘Taxon D’, and isoenzyme studies showed that it is genetically distinct from the related species S$. rupicola (Syme) Hedl. and S$. vexans (Proctor ef al. 1989). It is probably the plant referred to as ‘S. rupicola “Devon form” * by Martin & Fraser (1939). Sorbus margaretae is endemic to the coast of South-west England from Combe Martin, North Devon (v.c. 4) to Culbone, South Somerset (v.c. 5). The total population is at least 120 plants. The I.U.C.N. (2001) conser- vation assessment is ‘Endangered’ based on the available data. Sorbus margaretae is distinguished from S. vexans by the darker red fruits which are much broader than long, and by minor leaf characters (Proctor et al. 1989). It is distinguished from S. rupicola — which has elliptic-obovate leaves with single toothing — by the broader obovate leaves with biserrate toothing. Sorbus porri- gentiformis also grows in some of the same localities and is similar but has broader leaves. Figure 3 shows typical leaves and fruits of S. margaretae, S. rupicola and S. vexans. NEW SORBUS TAXA 211 FIGURE 3. Typical leaves and fruits of S. margaretae (A—C), S. rupicola (D-F) and S. vexans (G—I). Scale bars 1 cm. Sorbus x motleyi T. C. G. Rich, hybr. nov. = Sorbus aucuparia L. x S. leyana Wilm. HOLOTYPUS: Coed Penmailard, Brecon (v.c. Ly acswmilewe. G. Rich, 11 June 2002 (NMW, accession no. V.2002.17.074). DESCRIPTION Arbor juvenis caulibus erectis. Petioli 10- 14 mm. Folia lata brachyblastorum 65-86 x 50-60 mm, 1:25-1-65plo longiora quam latiora, elliptica, ad longitudinis suae 40—55 partes centensimas latissima, venis 13-17, in centro laminae ad angulum 37-—55° a costa tentis, profunde lobata, ad basin pinnis lateralibus 4~—7 discretis ellipticis acutis et foliolo terminali multo ampliore late ovato _ obtuso lobato praedita, marginibus magis minusve uniserratis, dentes acutos prorsum projectos ferentibus, supra glabrescentia, infra viridulialba tomentosa, aestate exeunte glabrescentia. Inflorescentiae ad 8 cm diametro, ageregatae, tholiformes, ramulis sparsim tomentosis instructae. Flores 12-18 mm diametro. Sepala. triangularia, tomentosa, glandulis paucis in marginibus praedita. Petala 7-9 x 5-6 mm, elliptica, alba. Antherae cremeae roseitinctae. Styli 2, plus minusve per unum longitudinis trientem conjuncti, ad basin pilosi. Fructus maximi 9-12 x 10-13 mm, (0-75—)0-8—1-Oplo longiores quam _latiores, aspectu plerumque latiores quam longiores, in maturitate coccinei, lenticellis paucis parvis praediti. Young tree with erect stems. Petioles 10- 14 mm. Broad leaves of short shoots 65—86 x 50-60 mm, 1:25-1:65 times as long as wide, elliptic, widest 40-55% of the way along the leaf length, with 13-17 veins, held at an angle i) — ii) Te) G RICH AND MEG PROGHOR FIGURE 4. Leaves from Sorbus species at Coed Penmailard. A, B, Broad leaves of short shoots of S. leyana. C, D sucker leaves of S. leyana. E, F Broad leaves of short shoots of S. x motleyi. G, S. aucuparia. Scale bar = 1cm. FIGURE 5. Typical leaves of S. scannelliana (A, B), and S. arranensis (C, D). Scale bar 1 cm. ST eI — ——— NEW SORBUS TAXA 23 of 37—55° to the midrib at the centre of the lamina, deeply lobed with 4—7 free, elliptic, acute, lateral pinnae at the base and a much larger broadly ovate, obtuse, lobed terminal leaflet, with the margins + uniserrate with acute, forward-pointing teeth, glabrescent above, greenish-white tomentose — beneath, becoming glabrescent in late summer. Inflorescences to 8 cm in diameter, crowded, domed, with branchlets sparsely tomentose. Flowers 12-18 mm in diameter. Sepals triangular, tomentose, with a few glands on the margins. Petals 7-9 x 5—6 mm, elliptic, white. Anthers cream with a hint of pink. Styles 2, joined for about a third of their length, pilose at the base. Largest fruits 9-12 x 10-13 mm, (0-75—)0-8-1:0 times as long as wide and usually looking wider than long, widest at of below the middle, scarlet at maturity (R.H.S. colour chart 45B) with a few small lenticels. Sorbus x motleyi is a member of the S. anglica/S.intermedia group, and is_ charac- terised by the leaves with 4—7 free leaflets and a larger terminal lobe, and greenish white underneath. The wild saplings are too young to fruit, and the description of flowers and fruits here is based on grafted material which flowered at the National Botanic Garden of Wales in 2008. Sorbus Xx motleyi (Figure 4E—F) differs from S. leyana (Figure 4A, B) in having leaves with four or more free lateral pinnae, the lowest of which are typically constricted at their base where they join the petiole. Sorbus leyana leaves from sunlit situations are generally deeply lobed but lack free pinnae (Figure 4A, B), but in deep shade or on sucker regrowth from damaged stems, they may have |-3 free pinnae at the base of the leaf, usually with the lamina adnate to the midrib on both sides (Figure 4C, D). Sorbus x motleyi differs from S. aucuparia (Figure 4G) in having a broad, lobed, terminal leaflet much larger than the lateral leaflets. Sorbus x motleyi is more likely to be confused with two other hybrids than either of its parents. It differs from S$. x thuringiaca (ilse) Fritsch (=S. aucuparia x S. aria) in having broader leaves with fewer veins and obtuse apices (leaves 1-5—2-0 times as long as wide with 16—27 veins and acute apices in S. x thuringiaca). Sorbus x motleyi differs from S. Memijcorsii. Rich ~S. aueuparia x ‘S. intermedia) in having broader leaves with greenish-white-tomentose undersides to the leaf (1-4-2:0 times as long as wide and greenish-tomentose underneath in S$. x liljeforsit; Rich 2008). In 1999, G. S. Motley, after whom it is named, found a sapling with leaves inter- mediate between S. leyana and S. aucuparia under a S. aucuparia tree at Coed Penmailard, which he interpreted as a hybrid between them. A second, younger sapling was found nearby in 2004. AFLP data confirmed the hybrid origin of the first sapling (unpublished data; R. S. Cowan, pers. comm. 2002). Sorbus xX motleyi 1s analogous in origin to S. pseudofennica E. F. Warb., which arose as a S. arranensis Hedl. x S. aucuparia hybrid on Arran (Robertson et al. 2004a, b). The two S. x motleyi saplings appear to be different ages, and must have arisen independently; Robertson et al. (2004a, b) found that S$. pseudofennica had also arisen on more than one occasion. The older of the two S. x motleyi saplings probably originated in c. 1990 after the damage inflicted to the woodland by the second Great Storm of 1989 (cf. Rich et al. 2005). The chromosome number is not known. Sorbus x motleyi is endemic to Wales, and has only been found at Coed Penmailard, v.c. 42 Brecon. Currently only two saplings are known which could easily be accidentally trampled if care is not taken when visiting the Sie, It mS ILIWKCIN, (OOM) toeezie Carexony ‘Critically Endangered’. Sorbus scannelliana T. C. G. Rich, sp. nov. HOLOTYPUS: wooded limestone ridge east of Blue Pool, Ross Island, Killarney, V9469588161, v.c. H2 North Kerry, Ireland, 9 September 2008, R. Hodd, A. McVeigh & T. Rich (NMW, accession number’ V.2008.1.3; ISOTYPI: BEL, BM, DBN, TCD). Diagnosis: S. arranensi Hedl. similis sed foltis obtusis, in centro laminae 14-24 partibus centesimis ad costam non tantopere profunde lobatis (non, ut in S. arranensi, acutis, in centro laminae 30-60 partibus centesimis ad costam lobatis) differt. Similar to S. arranensis Hedl. but differing in having obtuse leaves not so deeply lobed 14-24% of the way to the midrib at the centre of the lamina (not acute and lobed 30-60% of the way to the midrib at the centre of the lamina as in S. arranensis). Sorbus scannelliana is a member of the S. anglica/S. intermedia group, and _ is characterised by the shallowly lobed, elliptic leaves which are greenish-white tomentose 214 T. C. G. RICH AND M. C. F. PROCTOR underneath. It has probably evolved as a hybrid between §. aucuparia and S. rupicola (anal- ogous in origin to §. arranensis) in the Killarney area where both parents occur. The epithet honours Maura J. P. Scannell, whose knowledge of Irish botany is unrivalled. Sorbus scannelliana is endemic to Ross Island, Killarney, Kerry (v.c. H2), Ireland, and was first found in June 1988 by T. Rich and N. Taylor. The population size in September 2008 was found to be only five plants, and the I.U.C.N. (2001) conservation assessment is ‘Critically Endangered’. In the Killarney area, S. scannelliana is only likely to be confused with S. anglica which is locally frequent on Ross Island and the Muckross Peninsula, and S. intermedia. Sorbus anglica has broader leaves which are whiter underneath with distinct cross venation between the main leaf veins (Figure 2c, d). Sorbus intermedia is also naturalised on Ross Island and differs in having more deeply lobed leaves with fewer veins. Sorbus scannelliana differs from S. leyana Wilm. in having shallower leaf lobes, and from S. minima (Ley) Hedl. in having larger, broader leaves. Figure 5 shows typical leaves of S. scannelliana and S. arranensis. Sorbus stenophylla M. C. F. Proctor, sp. nov. HOLORYRUS.) Ranent yi 2 SSObsmmnys Cama Breconshire, Wales, SO252305, 6 September 1970, M. Porter (NMW). DESCRIPTION Frutex vel arbor parva ad 8 m. Gemmae lanceoloideae, acutae, pilis albis vestitae. Folia lata brachyblastorum (5-5—)7-0—11-5(—12:5) x 4-5—7-0(—7:5) cm, obtrullata vel oblonga, 1-35— 1:75plo longiora quam latiora, ad longitudinis suae 52-67 partes centesimas latissima, apice obtuso vel rotundato interdum acuminato vel acuto, basi cuneata (angulo inter marginem et costam 30-42°), plerumque sine lobis sed interdum lobis non profundis secta, marginibus dentes acutos biserratos, prope apicem plerumque curviserratos, ferentibus sed in |-— 3 cm infimis prope petiolum fere integris, venis (13-)14—-20(-21), in centro laminae ad angulum (25—)27—33(—37)° a costa tentis; pagina superior atroviridis; pagina inferior griseialba tomentosa. Inflorescentiae ad 6 cm diametro, tholiformes, aggregatae, ramulis tomentosis instructae. Sepala triangularia, aut Sparsim tomentosa aut viridia et solum in marginibus tomentosa. Petala 6-9 x 6-7 mm, late elliptica vel late ovata, alba. Antherae roseae. Styli 2, plus minusve per dimidium longitudinis conjuncti. Fructus maximi 11-5— 13-5 x 12-14 mm, 0-85-1-Oplo longiores quam latiores, aspectu plerumque latiores quam longiores, in maturitate rubri, lenticellarum parvarum vel magnarum numero modico praediti. Shrub or small tree to 8 m tall. Buds lanceoloid, acute, clothed with white hairs. Broad leaves of short sterile shoots (5-5—)7-0— 11-5(-12°5) x 4:5-7-0(-7:5) cm, obtrullate to oblong, 1:35-1:75 times as long as wide and widest 52-67% of the way along the leaf length, with the apex obtuse to rounded, occasionally acuminate or acute, and the base cuneate (with the angle between the margin and the midrib 30-42°), usually unlobed or occasionally with shallow lobes, with the margins biserrate, with acute teeth usually curving towards the apex near the top, but with the lowest 1-3 cm near the petiole nearly entire, with (13—)14-20(-21) veins, held at an angle of (25—)27—33(—37)° to the midrib at the centre of the lamina; upper surface dark green; lower surface greyish-white tomentose. Inflorescences to 6 cm in diameter, domed, crowded, with branchlets tomentose. Sepals triangular, either sparsely tomentose or green and tomentose only on the margins. Petals 6—9 x 6-7 mm, broadly elliptic to broadly ovate, white. Anthers pink. Styles 2, joined for about half their length. Largest fruits 11:-5-—13-5 x 12- 14 mm, 0-85—1-0 times as long as wide and usually looking wider than long, red at maturity (R.H.S. colour chart 42A, 44A, 45A), with a moderate number of small to large lenticels. Sorbus stenophylla is a member of the S. porrigentiformis group within the S. aria group, and is characterised by the narrowly cuneate, obtrullate to oblong leaves with deeply incised teeth, and fruits which are wider than long. Specimens named by E. F. Warburg show that he included material from the Llanthony Valley in his S. porrigentiformis. Studies of the isozymes by Proctor & Groenhof (1992) demonstrated that the isozyme profiles differed from S. porrigentiformis and were similar to S. cambrensis, differing from the latter in clearer separation of two bands. As the leaf shape is consistently different, they are treated as a distinct species. For leaf silhouettes showing S. stenophylla see Figures 5m—o of Proctor & Groenhof (1992). For differences from S. cambrensis, see above. NEW SORBUS TAXA PN) Sorbus stenophylla is endemic to the Llanthony Valley, Wales at Darren, Cwmyoy (v.c. 35), Tarren yr Esgob (in both v.c. 35 and v.c. 42) and Darren Lwyd (v.c. 42). Some plants at Craig-y-Cilau (v.c. 42) and Pwll-du Quarry, Govilon (v.c. 35) match S. stenophylla in morphology but require verification from DNA analysis. The population size at Tarren yr Esgob is c. 50 plants, there are seven trees at Darren Lwyd, and 50-100 at Cwmyoy. The I.U.C.N. (2001) conservation assessment is ‘Endangered’. Sorbus stirtoniana T. Rich, sp. nov. HOLOTYPUS: Tree c. 5 m tall with many trunks above track just west of junction, West Crags, Craig Breidden, v.c. 47 Montgomeryshire, Wales, grid reference SJ288139, T. C. G. Rich, A. Law, L. Houston, C. Charles and A. C. Tillotson, 19 June 2001 (NMW,, accession no. V.2001.025.184), with photographs of fruits and measurements 2 October 2002. Diagnosis: A S. leptophylla E. F. Warb. foliis ovatis vel ellipticis (non obovatis), ad vel infra (non supra) medium latissimis, differt. Differs from S. leptophylla E. F. Warb. in having ovate or elliptic (not obovate) leaves, widest at or below (not above) the middle Full description: Tree or small shrub to 5 m, with trunk ascending or appressed to the rock surface. Broad leaves of short shoots (SO—)60— 110(-135) mm long x (32—)40-75(-77) mm wide, 1-3—1-8 times as long as wide, ovate to elliptic, widest at (37—)38-55(—56)% along the leaf length, cuneate at the base, acute at the apex, with margins simple or weakly lobed, serrate to biserrate, with a total of (16—)18—24 veins, the veins at the middle of the leaf held at an angle of (28—)29-47(-48)° to the midrib, glabrescent above, white-tomentose below. Fruits 11-15(-16) mm long x (10—)11—13 mm wide, 0-9-1-3 times as long as wide and usually appearing longer than wide, dark red (RHS 1966 colour chart 46b), with few medium-sized lenticels scattered mainly at the base. Sorbus stirtoniana 1s a member of the S. aria group, and is characterised by the elliptic to ovate, acute, cuneate leaves with 18—24 veins which are white-tomentose below, and the large fruits which are usually longer than wide. Leaf outlines are given in Figures 2a—e of Rich et al. (2005). It is tetraploid (Bailey et al. 2008). Sorbus stirtoniana differs from S. leptophylla E. F. Warb., to which it is most closely related, in having leaves wider below rather than above the middle (cf. Figure | of Rich et al. 2005). It shares the same Aria chloroplast type ‘D’ with S. leptophylla (Chester et al. 2007), which is also found in S. aria (L.) Crantz (frequent) and in S. porrigentiformis (rare). Sorbus stirtoniana differs from S. porrigentiformis and S. rupicola in having fruits longer than wide, and in having leaves widest below the middle. Sorbus stirtoniana differs from the diploid S. aria in having larger leaves and fruits, and in ploidy level. At least 35 plants are known on the cliffs of the West Crags of Craig Breidden, and two on the North Crags (Rich et al. 2005), to which it is endemic. It is LU.C.N. (2001) ‘Critically Endangered’. Named in honour of Prof. Charles Stirton in recognition of his inspirational work in estab- lishing the National Botanic Garden of Wales. ACKNOWLEDGMENTS Libby Houston, Andy Law, Graham Motley and Ashley Robertson for help with field work, Hanson ple for giving us access to Craig Breidden, P. D. Sell for sharing his knowledge and allowing us to publish some of his taxa, John Richards and Clive Stace for helpful comments, and Philip Oswald and David Price for Latin translations. REFERENCES AAS, G., MAIER, J., BALTISBERGER, M. & METZGER, S. (1994). Morphology, isozyme variation, cytology and reproduction of hybrids between Sorbus aria (L.) Crantz and S. torminalis (L.) Crantz. Botanica Helvetica 104: 195-214. BAILEY, J. P., KAY, Q. O. N., MCALLISTER, H. & RICH, T. C. G. (2008). Chromosome numbers in Sorbus L. (Rosaceae) in the British Isles. Watsonia 27: 69-72. CHESTER, M., COWAN, R. S., FAy, M. F. & RICH, T. C. G. (2007). Parentage of endemic Sorbus L. (Rosaceae) species in the British Isles — evidence from plastid DNA. Botanical Journal of the Linnean Society 154: 291-304. COWAN, R. S., SMITH, R. J., FAY, M. F. & RICH, T. C. G. (2008). Genetic variation in Irish Whitebeam, Sorbus hibernica E. F. Warb. (Rosaceae) and its relationship to a Sorbus from the Menai Strait, North Wales. Watsonia 27: 99-108. 216 T. C. G. RICH AND M. C. F. PROCTOR I.U.C.N. (2001). .U.C.N. Red List Categories. Version 3.1. International Union for Conservation of Nature, Gland. MARTIN, W. K. & FRASER, G. T. (1939). Flora of Devon. T. Buncle & Co., Arbroath. MorGAN, V. (1987-1989). Rare plant survey of Wales. Unpublished reports to the Nature Conservancy Council. PROCTOR, M. C. F. & GROENHOF, A. C. (1992). Peroxidase isoenzyme and morphological variation in Sorbus L. in South Wales and adjacent areas, with particular reference to S. porrigentiformis E. F. Warb. Watsonia 19: 21-37. PROCTOR, M. C. F., PROCTOR, M. E. & GROENHOF, A. C. (1989). Evidence from peroxidase polymorphism on the taxonomy and reproduction of some Sorbus populations in south-west England. New Phytologist 112: 569-575. RicH, T. C. G. (2008). Sorbus x liljeforsii, a name for the S. aucuparia x S. intermedia hybrid (Rosaceae). Nordic Journal of Botany 25: 339-341. RICH, T. C. G. & CANN, D. C. G. (2009). A survey of Sorbus species at Watersmeet, North Devon, September 2007. Transactions of the Devonshire Association for the Advancement of Science. 140: 185— 198. RICH, VT. €. G., HOUSTON, LU, CHARLES, ©: S TILLOTSON; Ay €: 2009). The diversitysotySojbusmle: (Rosaceae) in the Lower Wye Valley. Watsonia (in review). RICH, T. C. G., MOTLEY, G. S. & KAY, Q. O. N. (2005). Population sizes of three rare Welsh endemic Sorbus species (Rosaceae). Watsonia 25: 381-388. ROBERTSON, A., NEWTON, A. C. & ENNOS, R. A. (2004a). Multiple hybrid origins, genetic diversity and population genetic structure of two endemic Sorbus taxa on the Isle of Arran, Scotland. Molecular Ecology 13: 123-143. ROBERTSON, A., NEWTON, A. C. & ENNOS, R. A. (2004b). Breeding systems and continuing evolution in the endemic Sorbus taxa on Arran. Heredity 93: 487-495. ROYAL HORTICULTURAL SOCIETY (1966). R. H. S. colour chart. Royal Horticultural Society, London. SELL, P. D. (1989). The Sorbus latifolia (Lam.) Pers. aggregate in the British Isles. Watsonia 17: 385-399. WARBURG, E. F. (1957). Some new names in the British flora. Sorbus. Watsonia 4: 43-46. . Woops, R. G. (1998). The conservation of some Sorbus species in Wales. pp. 32-39 in A. JACKSON & M. FLANNAGAN, eds. The conservation status of Sorbus in the UK. Royal Botanic Gardens, Kew. (Accepted December 2008) Watsonia 27: 217-228 (2009) ANI Five new Sorbus (Rosaceae) taxa from the Avon Gorge, England WU, C, G, INUCi sk Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff, CF 10 3NP S. A. HARRIS University of Oxford, Department of Plant Sciences, South Parks Road, Oxford, OXI 3RB and S. J. HISCOCK School of Biological Sciences, University of Bristol, Woodland Road, Bristol, BSS 1UG ABSTRACT One new Sorbus species and four new hybrids are described from the Avon Gorge, Bristol. Sorbus x avonensis is a hybrid between S. aria and S. porrigentiformis. Sorbus X houstoniae is a hybrid between S. aria and S._ bristoliensis. Sorbus leighensis is a distinct clone of S. aria x S. porrigentiformis and is locally frequent in the Avon Gorge; it is treated as a species. Sorbus X robertsonii is a hybrid between S. aria and S. eminens. Sorbus X proctoris 1s a hybrid between S. aucuparia and the cultivated S$. scalaris. KEYWORDS: Apomixis, hybrid, species, whitebeam, rowan. INTRODUCTION The Avon Gorge in South-west England is a large Carboniferous Limestone gorge about 3 km long with wooded slopes and open rocks, cliffs and quarries on the western edge of the city of Bristol. It is designated as a Site of Special Scientific Interest (S.S.S.I.) and Special Area of Conservation (S.A.C.) for its out- standing flora and ravine woodlands (Ratcliffe 1977), and is the most diverse Sorbus site in Britain with endemic species such as S. bristoliensis Wilm. and S. wilmottiana E. F. Warb. (Rich & Houston 2004; Houston et al. 2008). During the course of a three-year project funded by the Leverhulme Foundation to elucidate the origin of the endemic Sorbus Species in the Avon Gorge, five new Sorbus taxa have been discovered and their origins *e-mail: tim.rich@museumwales.ac.uk ascertained by analysis of nuclear micro- satellites (Robertson et al., in preparation). In this paper the new taxa are described. All the taxa have originated by hybridisation. It is likely that most Sorbus taxa derived from independent crossings of an apomictic poly- ploid with a variable sexual diploid would result in genetically different, apomictic, polyploids. Such apomictic hybrids could be considered species in their own right as they are reproductively isolated and genetically distinct, but we have adopted the following approach in deciding whether to treat them as hybrids or species. Following normal practice in other groups, the taxa are treated as hybrids if there are one or a few individuals of the same parentage which have probably originated from different hybridisation events (i.e. are poly- phyletic), irrespective of whether they are apomictic or not, or polyploid or not. Where only a few such spontaneous hybrids occur in a number of different locations which cannot be differentiated morphologically, it is better to treat them as hybrids rather than describe a myriad of different species which are im- practical to tell apart without DNA markers. Additionally, as the biological success of many of these hybrids appears to be low since they do not increase with time (although we do not know when they originated), these may be only short-lived, ‘temporary’ taxa. However, where there are significant numbers of individuals of a morphologically distinct, monophyletic, apomictic clone (i.e. they are biologically successful), they are treated as species, even if 218 T€. G, RICH, S] A] HARRISVAND SEIS COCK of the same parentage as previously described taxa. For example, the locally frequent and distinct S$. leighensis and the two slightly different examples of the hybrid S. x avonensis described below are all derived from S. aria (L.) Crantz x S. porrigentiformis E. F. Warb., but are morphologically different from each other and from the readily recognised existing species S$. wilmottiana E. F. Warb. and S. whiteana T. Rich & L. Houston which also have the same parentage (Robertson ef al., in preparation). If some of the hybrids should subsequently prove to be successful apomictic clones which become more frequent in the future, their rank could be raised to species. METHODS Broad leaves from the short, vegetative shoots in sunlit situations, excluding the oldest and youngest leaf (Aas et al. 1994), were measured on herbarium material in NMW. Fruits were measured on fresh material, and the colours matched against the Royal Horticultural Society colour charts (Royal Horticultural Society 1966). Potential pollen viability was investigated using Alexander’s Stain (Alexander 1969) on the flowering collections available. Anthers were removed from herbarium specimens with tweezers under a low-power binocular micro- scope, and placed on a slide with a drop of Alexander’s Stain, warmed briefly on the hotplate, then broken up with the tweezers. The preparation was then covered with a cover-slip and replaced on the hotplate to improve the uptake of the stain. The slides were then examined under a_ high-power compound microscope for areas of dense pollen grains. Potentially viable grains were counted as those which were large and rounded-triangular with cell walls which stained green and with cytoplasm inside which stained uniformly bright red. Small deformed grains or those staining green only with very little or no red staining inside (i.e. no cytoplasm) were considered infertile. The numbers of pollen grains counted are cited for individual trees; averages and ranges of other species cited are taken from Rich (2009). Soil pH was measured with a calibrated pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water on soil samples collected from around the roots. For nuclear microsatellite methods and results, see Robertson ef al. (in preparation). DESCRIPTIONS Sorbus aria (L.) Crantz x S. porrigentiformis E. F. Warb. = S. X avonensis T. Rich, hybr. nov. HOLOTYPUS: St Vincent’s Rocks South, Avon Gorge, v.c. 34 West Gloucestershire, England, ST/565.730, flowers 21 May 2007, L. Houston, and fruits and photographs 19 September 2007, T. C. G. Rich & L. Houston (NMW, accession number V. 2007.1.132). ISOTYPI: BM, CGE. Vernacular name: Avon Gorge Whitebeam. Frutex vel arbor parva ad 7 m ut minimum. Gemmae ovoidoconicae, virides, ad apicem pilis exalbidis pilosae. Folia lata brachyblastorum 5-5—10-0 x 4:0-7:5 cm, 1-:2— 1-4plo longiora quam latiora, obovata vel late elliptica, ad longitudinis suae 51-62 partes centensimas latissima, apice obtuso, basi late cuneata, in centro laminae OQ-15 partibus centensimis ad costam leniter lobata, marginibus leniter biserratis dentes prorsum projectos ferentibus, sed 1—2 cm infimis prope petiolum fere integris; pagina superior atroviridis glabra; pagina inferior viridialba tomentosa, venis 15—20 ad angulum 25-—32° a costa tentis. Petioli 8-20 mm. Inflorescentia ad 8 cm lata, aggregata, ramulis tomentosis instructa. Flores circa 16-18 mm _/ati, fragrantes. Sepala anguste triangularia, ad apicem viridula sed ad basin tomentosa, in marginibus sine glandibus. Petala circa 6 xX 5 mm, late elliptica, cupulata, alba. Antherae roseae. Styli 2, usque ad basin discreti, virides, basi pilosa. Fructus maximi 10-11 x 10- 12 mm, 0-85—1-Oplo longiores quam latiores, plerumque aspectu magis minusve globoso, in medio latissimi, in maturitate rubri, lenticellis paucis parvis vel magnis praediti. Semina perfecte formata. Shrub or small tree to at least 7 m. Buds ovoid- conical, green, pilose at tip with whitish hairs. Broad leaves of short shoots 5:5—10-0 x 4-0- 7-5 cm, 1:2—1-4 times as long as wide, obovate to broadly elliptic, widest at 51-62% along leaf length, with apex obtuse and base broadly cuneate, weakly lobed 0-15% of way to midrib at centre of leaf; margins weakly biserrate with teeth directed forwards, but with lowest 1-2 cm near the petiole nearly entire; upper surface dark green, glabrous; lower surface greenish- white tomentose, with 15—20 veins held at an angle of 25—32° to the midrib. Petioles 8-20 mm. NEW SORBUS TAXA FROM THE AVON GORGE 219 FIGURE |. Leaves and fruits of Sorbus x avonensis. A—D, type. E-F, second tree. Scale bars 1 cm. Inflorescence to 8 cm across, crowded: branch- lets tomentose. Flowers c. 16-18 mm across, fragrant. Sepals narrowly triangular, greenish at tip but tomentose at base, without glands on margins. Petals c. 6 Xx 5 mm, broadly elliptic, cupped, white. Anthers pink. Styles 2, split to base, green, pilose at base. Largest fruits 10-11 x 10-12 mm, 0-85-1-0 times as long as wide but usually looking more or less globose, widest at the middle, red at maturity (RHS colour chart 44A, 45A), with a few small to large lenticels. Seeds fully formed. Sorbus X avonensis is a member of the Sorbus aria group. The description is based on both the type tree and the second tree known from the opposite side of the gorge, which is identical genetically but differs slightly morphologically, though this may be due to environmental conditions. Hybridisation seems to occur repeatedly between S. aria and S. porrigentiformis in the Avon Gorge. This has given rise to a series of morphological forms, some of which appear to have reproduced to form stable clones with a 220 T. C. G. RICH, S. A. HARRIS AND S. J. HISCOCK significant number of plants (e.g. S. eminens, S. whiteana, §S. wilmottiana and the new S. leighensis described below), and many of which have not; we intend this hybrid name to be applied to spontaneous S. aria xX S. porrigentiformis plants which do not form significant populations. Sorbus X avonensis (Fig. 1) resembles S. porrigentiformis more than S. aria, with broadly obovate, weakly biserrate leaves and fruits globose to wider than long (length/width ratio 0-85—1-0), and has been recorded as S. porrigentiformis in the past in the Avon Gorge. Sorbus porrigentiformis sensu stricto has obovate, unlobed, obtuse, weakly biserrate leaves, and fruits usually wider than long (length/width ratio 0-8—1-O(— 1-1)). Sorbus aria has white-tomentose undersides to the leaves, usually more veins and fruits usually longer than wide. Nuclear microsatellite data suggest that, in the Avon Gorge at least, S. porrigentiformis sensu stricto 1s consistently tetraploid and that S. X avonensis is triploid (Robertson ef al., in preparation). Counts of 2n = 51 previously relemed to) iS) \porrigentiformis (e225) by; Warburg 1962, who interpreted S. porrigentiformis broadly, including — both triploids and tetraploids) probably refer to hybrids. Pollen stainability in one S. xX avonensis tree was 66% (167 pollen grains counted), lower than in S. aria (average 85%, range 23—100%) and slightly lower than in S. porrigentiformis (average 73%, range 61- 84%). Fruits of S. Xx avonensis have apparently well-formed seeds, and some at least are viable. The plastid DNA of the type tree is the widespread Aria type ‘A’ (Chester et al. 2006). The only occurrences of S$. X avonensis confirmed by nuclear microsatellites are two places on opposite sides of the Avon Gorge (Fig. 2; v.cc. 6, 34). One isolated tree occurs on open Carboniferous Limestone rocks and the other in taller woodland with S. aria and S. bristoliensis between the railway line and the towpath. Neither tree occurs with S. porrigentiformis. Sorbus X avonensis 1s likely to occur wherever the parents overlap in SW England and Wales. In particular we have seen triploid material from Burrington Combe (v.c. 6) and Blaise Castle (v.c. 34), and material of unknown ploidy from the Wye Valley (v.cc. 34, 36) which could be this hybrid, but DNA analysis may be the only certain method of verifying their identity. The epithet ‘avonensis’ occurrence in the Avon Gorge. relates to its Sorbus aria (L.) Crantz x S. bristoliensis Wilm.= S. x houstoniae T. Rich, hybr. nov. HOLOTYPUS: rocks, Stokeleigh Camp, v.c. 6 North Somerset, England, ST/561.732, flowers 24 May 2005, T. C. G. Rich & L. Houston, and imuuits Sl October 2004.) ©. Gar Richee-e Robertson & L. Houston (NMW, accession numbers V.2005.1.148 and V.2004.26.276). Vernacular name: Houston’s Whitebeam. Arbor parva ad 5 m ut minimum. Gemmae ovoidoconicae, virides, ad apicem _pilis exalbidis pilosae. Folia lata brachyblastorum 7:0-9:'5 x 4:5-6:5 cm, 1:4-1-6plo longiora quam latiora, obovata vel obovatoelliptica, ad longitudinis suae 50-70 partes centensimas latissima, apice acuminato, basi late cuneata, in centro laminae 7—14 partibus centensimis ad costam valde lobata, marginibus _ leniter biserratis dentes prorsum projectos ferentibus, sed 1-2 cm infimis prope petiolum integris; pagina superior atroviridis glabra; pagina inferior viridialba tomentosa, venis 17—22 ad angulum 32-39° a costa tentis. Petioli 8— 25 mm. Inflorescentia ad 8 cm lata, aggregata, ramulis tomentosis instructa. Flores 16-18 mm lati, fragrantes. Sepala anguste triangularia, ad apicem viridula sed ad basin tomentosa, in marginibus sine glandibus vel glandes 1-2 ferentes. Petala 5-5—7-5 x 5:5-6 mm, late elliptica vel fere orbicularia, cupulata, alba. Antherae roseae. Styli 2, per unum longitudinis trientem conjuncti, virides, glabri. Fructus maximi (1 1—)12—14(—14-5) x (11-5—)12-14:5(— 15) mm, 0-9-1-Iplo longiores quam latiores, plerumque .aspectu magis minusve globoso, in medio latissimi, in maturitate croceorubri, lenticellarum parvarum mediocriumque numero modico praediti. Semina_perfecte formata. Small tree to at least 5 m. Buds ovoid-conical, green, pilose at tip with whitish hairs. Broad leaves of short shoots 7-0-9-5 x 4:5-6:5 cm, 1-4-1-6 times as long as wide, obovate to obovate-elliptic, widest at 50-70% along leaf length, with apex acuminate and base broadly cuneate, distinctly lobed 7—-14% of way to midrib at centre of leaf; margins weakly biserrate with teeth directed forwards, but with lowest 1-2 cm near the petiole entire; upper surface dark green, glabrous; lower surface greenish-white tomentose, with 17-22 veins held at an angle of 32-39° to the midrib. Petioles 8-25 mm. Inflorescence to 8 cm across, SS = NEW SORBUS TAXA FROM THE AVON GORGE i) 2) V.C. 34 West N Gloucestershire — = . oe tee v Leigh Woods -74 V.C. 6 North Somerset ee 56 $7 | | | FIGURE 2. Distribution of new Sorbus taxa within the Avon Gorge, South-west England. A, S. x avonensis. H, S. X houstoniae. L, S. leighensis. P, S. x proctoris. R, S. x robertsonii. Base map courtesy L. Houston. 22 T. C. G. RICH, S. A. HARRIS AND S. J. HISCOCK FIGURE 3. Leaves and fruits of Sorbus x houstoniae. Scale bars 1 cm. crowded; branchlets tomentose. Flowers 16— 18 mm across, fragrant. Sepals narrowly triangular, greenish at tip but tomentose at base, without or with 1—2 glands on margins. Petals 5-5—7:5 x 5-5-6 mm, broadly elliptic to nearly orbicular, cupped, white. Anthers pink. Styles 2, united for one-third of length, green, glabrous. Largest fruits (11—)12—14(-14:5) x (11-5—)12-14-5(-15) mm, 0-9-1-1 times as long as wide but usually looking more or less globose, widest at the middle, orangey-red at maturity (RHS colour chart 33A, 34A, 44A), with a moderate number of small and medium lenticels. Seeds fully formed. Sorbus X houstoniae is a clear member of the S. latifolia group, intermediate between the S. aria group and S. torminalis (L.) Crantz. Sorbus X houstoniae (Fig. 3) looks similar to S. bristoliensis, with which it is very easily confused. Both taxa have shallowly, acutely lobed, elliptic-obovate leaves that are greenish- white beneath, and orangey-red fruits. Sorbus x houstoniae differs from S. bristoliensis in having relatively broader leaves. Sorbus xX houstoniae is not likely to be confused with S. aria which has red fruits and leaves white- tomentose beneath and usually without acute lobes. Nuclear microsatellites and plastid DNA have shown that S. x houstoniae 1s a tetraploid derived from fertilisation of an unreduced triploid ovule of S. bristoliensis by normal haploid S. aria pollen (Robertson ef al. in preparation). Microsatellite data do not support Challice & Kovanda’s (1978) suggestion that an odd S. bristoliensis tree in the former Bristol University Botanic Garden at North Road, Leigh Woods is of this origin, though we are not certain whether we sampled the same tree. The one hybrid tree had an average of 98% (50 pollen grains counted) stainable pollen, probably equivalent in fertility to S. aria (average 85%, range 23-100%) but much higher than S. bristoliensis (average 27%, range 11—38%). Fruits with fully-formed seeds occur, but it is not known if they are viable. The sole tree confirmed by nuclear micro- satellites occurs on a north-facing, vertical Carboniferous Limestone cliff below Stoke- leigh Camp, Leigh Woods, North Somerset (v.c. 6), growing near both parents as well as other Sorbus species (Fis. 2) ihemtrcemis inaccessible without ropes. The main threat to its survival.1s growth of ivy. Named after Ms Libby Houston, the “guardian angel” of the Avon Gorge, who not only first discovered it, but whose natural habitat also seems to be the same — clinging to impossible cliffs in remote parts of the gorge. Sorbus leighensis T. Rich, sp. nov. HOLOTYPUS: Quarry 4, Leigh Woods, v.c. 6 North Somerset, England, ST/561.739, flowers 15 May 2007, T. C. G. Rich & A. Robertson, and fruits and photographs 19 September 2007, T. C. G. Rich & L. Houston (NMW accession number V. 2007.1.180). IsoTyPl: BM, CGE Vernacular name: Leigh Woods Whitebeam. Frutex vel arbor parva ad 10 m ut minimum. Gemmae ovoidoconicae, virides, ad apicem pilis albis pilosae. Folia lata brachyblastorum (6:5—)7-0-10°5 x (4-5—)5-0-7-0(-8:5) cm, 1-:2- 1-5(—1-6)plo longiora quam latiora, plerumque obovata, aliquando elliptica, ad longitudinis ———— re eC TCOC— a SS _ — NEW SORBUS TAXA FROM THE AVON GORGE 3} suae (50—)55—65 partes centensimas latissima, apice truncato vel late obtuso, basi anguste rotundata vel cuneata, in centro laminae 6-13 partibus centensimis ad costam plerumque lobis non profundis obtusis secta, marginibus leniter biserratis dentes parvos_ exstantes ferentibus, sed parte infima prope petiolum circa 1 cm fere integra; pagina superior pallide viridis glabra; pagina inferior alba _ vel viridialba tomentosa, venis 15—19 ad angulum (22—)24—31(—35)° a costa tentis. Petioli 8— 16 mm. Inflorescentia ad 9 cm lata, aggregata, ramulis tomentosis instructa. Flores 18-20 mm lati, fragrantes. Sepala anguste triangularia vel deltoidea, ad apicem viridula sed ad basin tomentosa, in marginibus sine glandibus. Petala 7-8 X 5-6 mm, late elliptica, cupulata, alba. Antherae cremeae, leniter roseae. Styli 2(—3), usque ad basin discreti, virides, basi pilosa. Fructus maximi 10—12(—13) x 10-14(-14) mm, 0-8—1-Iplo longiores quam latiores, plerumque aspectu globoso, in maturitate rubri, lenticellis paucis parvis vel mediocribus praediti. Semina perfecte formata. FIGURE 4. Leaves and fruits of Sorbus leighensis. Each leaf from a different tree. Scale bars 1 cm. 224 T. C. G. RICH, S. A. HARRIS AND S. J. HISCOCK Shrub or small tree to at least 10 m. Buds ovoid-conical, green, pilose at tip with white hairs. Broad leaves of short shoots (6:5—)7-0— 10°5 x (4:5—)5-0-7-0(-8:5) cm, 1:2—1-5(-1-6) times as long as wide, usually obovate, sometimes elliptic, widest at (50—)55-65% along leaf length, with apex truncate to broadly obtuse and base narrowly rounded to cuneate, usually with shallow, blunt lobes cut 6—-13% of way to midrib at centre of leaf; margins weakly biserrate with small teeth directed outwards, but with lowest c. 1 cm near the petiole nearly entire; upper surface light green, glabrous; lower surface white to greenish-white tomen- tose, with 15—19 veins held at an angle of (22-) 24—31(—35)° to the midrib. Petioles 8-16 mm. Inflorescence to 9 cm across, crowded; branchlets tomentose. Flowers 18-20 mm across, fragrant. Sepals narrowly triangular to deltoid, greenish at tip but tomentose at base, without glands on margins. Petals 7-8 x 5- 6 mm, broadly elliptic, cupped, white. Anthers cream with a hint of pink. Styles 2(—3), split to base, green, pilose at base. Largest fruits 10—12 (—13) x 10-14(—14) mm, 0-8—1-1 times as long as wide but usually looking globose, red at maturity (RHS colour chart 42A, 45A, 45B, 46A), with a few small to medium lenticels. Seeds fully formed. Sorbus leighensis 1s a clear member of the Sorbus aria group. It can be readily distinguished from the other species in the S. aria group by the obovate, truncate leaves with small but distinct, shallow, blunt lobes (Fig. 4). Usually only one leaf in each short shoot is truncate at the apex, the others are obtuse. This taxon has been known since at least the 1980s. It has been variously referred to as S. ‘Bristol porrigentiformis’ or “Bristol pseudo- porrigentiformis’ by Proctor et al. (1989) and Proctor & Groenhof (1992). The tree sampled as “S. wilmottiana 23’ by Lemche (1999) is also S. Jleighensis. Nuclear microsatellite markers show that S. leighensis is a triploid lOO Ckerahecl IKON S Gag X s, porrigentiformis (Robertson ef al. in preparation), and has the widespread S. aria plastid type ‘A’ (Chester et al. 2007). It has a distinct isozyme profile (Proctor & Groenhof 1992). Sorbus leighensis is endemic to the Somerset side of the Avon Gorge (Fig. 2), where it is locally frequent. At least two plants occur in Quarry 2, at least 12 plants (possibly 20) are known in Quarry 3, at least 63 plants in Quarry 4 and at least three along the towpath. One easily found tree occurs on the sea wall opposite the entrance to Quarry 2. Surprisingly, it has not been found on the Bristol side of the Avon Gorge (v.c. 34 West Gloucestershire). Some trees with virtually identical leaves occur in Burrington Combe, North Somerset (v.c. 6) but remain to be compared with plants from Avon Gorge using molecular markers. The pollen stainability of four trees was variable: 6% (54 pollen grains counted),13% (67 counted), 28% (72 counted) and 45% (88 counted; A. Patto, pers. comm. 2007), giving a mean of c. 23% which is fairly typical of triploids (Rich 2009). However it has fertile fruits and grows well from seed. Sorbus leighensis occurs on _ open Carboniferous Limestone rocks and scree, and in open scrub with S. aria, S. porrigentiformis and §. wilmottiana. Soil from under the type tree had pH 7°8. Sorbus leighensis has an I.U.C.N. (2001) conservation status of ‘Endangered’ due to the small population size (at least 80 plants, but not fully surveyed) and its restricted occurrence on one site. However, the whole population occurs within the Avon Gorge S.S.S.I. and S.A.C., and most trees occur on land owned by the National Trust. The main threats are scrub development, or clearance along the towpath. The epithet ‘leighensis’ is derived from its occurrence in Leigh Woods. Sorbus aria (L.) Crantz x S. eminens E. F. Warb.= S. x robertsonii T. Rich, hybr. nov. HOLOTYPUS: Great Fault, Avon Gorge, Bristol, v.c. 34 West Gloucestershire, England. ST/564.733, flowers 15 May 2007, T. C. G. Rich, and fruits and photographs 19 September 2007, T. C. G. Rich & L. Houston (NMW, accession number V.2007.1.175). ISOTYPI: BM, CGE, OXF. Vernacular name: Robertson’s Whitebeam. Arbor ad 12 m ut minimum, truncis multis praedita. Gemmae ovoidoconicae, virides, ad apicem pilis exalbidis pilosae. Folia lata brachyblastorum 10-0-14-0(-14:5) x 7-5—10-0 (-10°5) cm, (1-1—)1:2—1:6(-1-7)plo longiora quam latiora, obovata vel late elliptica, ad longitudinis suae 50—-63(-—69) partes centensimas latissima, apice obtuso, basi rotundata, marginibus non lobatis biserratis dentes parvos exstantes ferentibus, sed parte infima prope petiolum circa 1-3 cm fere integra; pagina superior viridis glabra; pagina inferior alba vel viridialba tomentosa, venis 20—26(—27) ad angulum 25—-40° a costa tentis. Petioli 12-26 mm. Inflorescentia ad 7 cm lata, NEW SORBUS TAXA FROM THE AVON GORGE 225 agegregata, ramulis tomentosis instructa. Flores 14-17 mm lati, fragrantes. Sepala anguste triangularia, ad apicem acuminata viridula sed ad basin tomentosa, in marginibus — sine glandibus. Petala 7-8 x 6 mm, late ovata vel late elliptica, cupulata, alba. Antherae cremeae. Styli 2, usque ad basin discreti, virides, glabri. Fructus maximi (12—)13—16(—17) x 10-15 mm, (0-95—)1-0-1:2plo longiores quam _latiores, plerumque aspectu longiore quam latiore, in medio latissimi vel obconici, in maturitate rubri, lenticellarum parvarum vel mediocrium numero modico praediti. Semina_perfecte formata. Tree to at least 12 m, with many trunks. Buds ovoid-conical, green, pilose at tip with whitish hairs. Broad leaves of short shoots 10-0—14-0(— 14:5) x 7-5-10-0(-10°5) cm, (1-1—)1-:2-1-6(- 1-7) times as long as wide, obovate to broadly elliptic, widest at 50-63(-69)% along leaf length, with apex obtuse and base rounded; margins unlobed, biserrate with small teeth directed outwards, but with lowest c. 1-3 cm near the petiole nearly entire; upper surface green, glabrous; lower surface white to greenish-white tomentose, with 20—26(—27) veins held at an angle of 25—40° to the midrib. Petioles 12—26 mm. Inflorescence to 7 cm across, crowded; branchlets tomentose. Flowers 14-17 mm across, fragrant. Sepals narrowly triangular, acuminate and greenish at tip but tomentose at base, without glands on margins. Petals 7-8 x 6 mm, broadly ovate to broadly elliptic, cupped, white. Anthers cream. Styles 2, split to base, green, glabrous. Largest fruits (12—)13—16(-17) x 10-15 mm, (0-95-) 1-0-1:2 times as long as wide but usually looking longer than wide, widest at the middle or obconical, red at maturity (RHS colour chart 44A, 45B, 46A), with a moderate number of small to medium lenticels. Seeds fully formed. Sorbus X robertsonii is a member of the Sorbus aria group. The above description applies to the one tree in the Avon Gorge, Bristol whose identity has been confirmed from nuclear microsatellites (Robertson ef al. in preparation). Apart from the obconical fruits, S$. x robertsonii (Fig. 5) is approximately morphologically intermediate between its parents. Sorbus aria has ovate to elliptic, often shallowly lobed leaves which are acute to obtuse at the apex and have weakly biserrate to strongly biserrate margins, with fruits usually longer than wide or globose (length/width ratio 0-9-1-2). Sorbus eminens in its strict sense (i.e. excluding the trees with (OO G FIGURE 5. Leaves and fruits of Sorbus 4000 50 Dry heath and road >2000 85 Dry heath & semi-improved 1200 60 pasture Dry heath & semi-improved 500 50 pasture Lowland blanket bog and road 800 50 Lowland blanket bog 450 <5 Lowland blanket bog 400 <5 Cutover lowland blanket bog 100 65 and road Lowland blanket bog 300 <<) Lowland blanket bog Ds) 30 Lowland blanket bog 20 <5) Lowland blanket bog 400 - Dry heath & lowland blanket 350 - bog Lowland blanket bog 190 - Lowland blanket bog 150 - Lowland blanket bog 50 - Lowland blanket bog DD - Lowland blanket bog 15 - Lowland blanket bog 13 - Lowland blanket bog 3 - Improved agricultural 250 - grassland Improved agricultural 170 - grassland Raised bog US = Whereas sites further inland are generally surrounded by unmodified Atlantic blanket bog. The flush habitat in which E. gracile grows, occurs in areas of intact Atlantic blanket bog where there is some convergence of water flow due to topography. There are eight of these sites, which also tend to be confined to the coastal fringe of the Connemara region (Fig. 2). At some of these sites, e.g. the flush north- west of Letterdife House, E. gracile grows in a quaking raft of vegetation which has developed along the margins of a narrow bog stream, while at others, e.g. the flush north-west of Errisbeg hill, the species grows in old drainage channels which have become filled with an unconsolidated, soupy peat. The dominant species in these flushed areas of bog are generally Carex lasiocarpa, Carex limosa, Juncus subnodulosus, Schoenus nigricans and Potamogeton poly- gonifolius. ERIOPHORUM GRACILE IN IRELAND 235 The population size of FE. gracile in Connemara varies with habitat (Table 2). Bog flush sites generally contain the smallest pop- ulations of the species, ranging from 3 to c. 400 flowering heads per site; the inland lake sites generally support larger populations, from 20 to c. 800 flowering heads. Populations found in coastal lakes are the largest, ranging from 500 to >5000 flowering heads per site. DISTRIBUTION IN OFFALY (V.C. H18) In August 1997 a small population of E. gracile was found growing along the margins of a base-rich pool in the cut-over edge of a raised bog at Sharavogue, Co. Offaly. This bog is notable in that it is one of the few remaining raised bogs in Ireland which retains lagg-type vegetation along its margins. This population of E. gracile is one of only two known east of the river Shannon. DISTRIBUTION IN WESTMEATH (V.C. H23) E. gracile occurs in Scragh Bog, one of the best remaining examples of a calcareous basin fen in Ireland (O’ Connell 1981). This popu- lation was first reported by Scannell ef al. (1968) and it remains the most easterly known population on the island. At this site E. gracile occurs in quaking scragh vegetation dominated by Schoenus nigricans, Menyanthes trifoliata, Scorpidium revolvens and the nationally rare bryophyte Cinclidium stygium. DISTRIBUTION IN WEST MAYO (V.C. H27) The most northerly population of E. gracile in Ireland occurs at Cloonooragh, north-west of the town of Crossmolina in County Mayo. At this location the species grows in a very wet flush, dominated by Carex limosa, which is surrounded by intact lowland blanket bog. This population was discovered in August 2002. DISCUSSION E. gracile is now known to be much more common than previously thought, especially in the Connemara region of county Galway. The recent discovery of additional sites for E. gracile outside Connemara suggests _ that further populations may exist in more remote areas of western Ireland. As E. gracile was re- found in many of its previously reported Irish sites, this suggests that the habitats of the species have not been subject to the same scale of degradation which has taken place in England over the past 150 years (see Winship 1994). However, the failure to re-find the species at five previously documented Irish sites gives cause for concern. It is also likely, given the comparatively recent discovery of E. gracile in Ireland, that many sites were destroyed prior to its discovery in 1966. Although documentary proof is lacking, many wetland areas which may have supported E. gracile have been lost in the past 150 years (Irish Peatland Conservation Council 1992). This loss is especially pronounced in the midlands and east of the country, which have experienced a greater level of agricultural intensification and peatland exploitation than in the west, particularly over the past 100 years. For Britain, it is presumed that E. gracile depends largely on vegetative spread as a means of population expansion at sites (Winship 1994). This is also apparent at many Irish sites. The spread of the species is most clearly seen in lakes where E. gracile grows in very sparsely vegetated areas e.g. Dohola Lough and the lake west of Cloonagat Lough. At these locations extensive networks of E. gracile rhizomes can be observed in shallow water close to the lake shores, spreading out from the scragh surface and colonising deep muddy peat. Although seeds of E. gracile germinate easily on moistened filter paper at 20 C (Conaghan 1995), seedlings of the species were not observed in the field during this survey. Indeed, little is known about the ability of seeds of E. gracile to germinate and establish in field conditions, and it seems likely that the establishment of seedlings in the wild is a comparatively rare event. What is apparent, however, is that once the species has become established at a site, it can spread well by means of vegetative growth, providing there is suitable habitat available to colonise. Hence it is likely that only a very low rate of seedling establishment may be necessary in order to ensure the survival of the species at a particular site. The occurrence of the species at certain sites suggests that it may be part of a relict community of early post-glacial times. If many of the populations encountered are relict populations, then conditions at the sites at time of establishment may have been quite different. Its occurrence on Scragh Bog, alongside several other nationally rare species of current more northerly distribution such as Pyrola rotundifolia, Dactylorhiza traunsteineri and the moss Tomentypnum nitens, indicates that this fen is a refuge for such species in the Irish midlands, a region where many fens have long 236 J. P. CONAGHAN AND M. SHEEHY SKEFFINGTON since given way to raised bog (O’Connell 1981). In the rare occasions where a lagg remains relatively intact around a raised bog, it supports a ‘relict’ fen community dominated by species such as Carex rostrata, C. lasiocarpa and Potentilla palustris. It is in such a community that E. gracile occurs in the N. Kerry and Offaly sites. It is also notable that a suite of rare bryophytes, with current boreal distributions, such as Tomentypnum _nitens, Paludella squarrosa and Leiocolea rutheana, have been located at a number of blanket bog flush sites in Co. Mayo (Lockhart 1999) which are within 5 km of the recently discovered E. gracile site in that county. The different types of habitat in which E. gracile was located contain populations which Vialiy sete auly aml |SIZ7e ane Semaditnerences sam population size are primarily related to the size of suitable habitat available to the species. For example, the larger scragh systems found in the Connemara coastal lake sites contain a large area of suitable habitat, 1.e. scragh edge, for E. gracile to colonise and therefore they contain the largest populations of the species. Since it is likely that many of the populations which occur in bog flush, calcareous fen and raised bog lagg habitats are relict populations, they may be remnants of much larger populations which occurred when there was more open water present in these locations. Although it may be argued that the habitats in which E. gracile grows often represent comparatively short-lived stages in seral succession, there is litthe doubt that drainage and/or eutrophication of water can accelerate the rate of succession from low-growing, species-rich plant communities to _ taller, species-poor ones (Tallis 1983; Wheeler 1983). The observed ecological preferences of E. gracile at its Irish sites, coupled with recent experiences in England (Winship 1994) and North America (Decker et al. 2006), suggest that the species is one of the first to disappear when open water conditions are eliminated. Therefore it is imperative that protective measures, such as the prevention of site drainage and surface water eutrophication, be imposed in order to slow down the rate of seral progression at sites. This safeguarding of the water quantity and quality at sites will ultimately result in a much longer “life span” of E. gracile populations. Although all the different E. gracile habitats are potentially threatened by drainage and eutrophication, the extent of these threats differs between sites. At present, lake sites occurring in the Connemara region are the least threatened because it is unlikely that they will be drained or otherwise interfered with in the near future. However, van Groenendael et al. (1993) noted that the rate of seral progression in coastal lakes in southwestern Connemara accelerated in the time period between the mid- 1970s and the early 1990s due to the increased use of artificial fertilisers on land within the lake catchments. This could accelerate the disappear- ance of E. gracile along with the open water. Populations of E. gracile which occur in blanket bog flush sites are more susceptible to drainage than those in lakes because their hydrology is intimately linked with that of the surrounding blanket bog and thus they are at a greater risk from the effects of damaging operations such as peat-cutting or afforestation in adjacent areas of land. This is further compounded by the fact that these populations are generally very small, both in flowering head number (Table 2) and extent and are therefore less likely to survive a damaging event such as drainage than a larger, more widely dispersed population. The sites where E. gracile occurs in Westmeath and North Kerry are both nature reserves and are thus legally protected from drainage and reclamation. However, the adjoining land is not protected at present and there is a danger that drainage or fertiliser application on this land may damage the ecology of the sites. From the. point of view of species conservation, one of the most crucial questions that needs to be addressed is at what point in the succession process does E. _ gracile disappear from a site. Winship (1994) has implicated the encroachment of dense reed swamp vegetation dominated by Phragmites australis in the elimination of E. gracile from English sites; however, in at least one of the Irish lake sites, the species is demonstrating the ability to persist for some time after open water conditions have disappeared. In the northern half of Cregduff Lough the species is common in an area of lake scragh, dominated by the tall reed swamp species Phragmites australis, Carex lasiocarpa and Schoenoplectus lacustris, in which there is no open water present. This suggests that the species may survive for a considerable time in areas of closed swamp vegetation which are still sufficiently wet. The future monitoring of E. gracile at this site will reveal much about the ability of the species to persist in habitats where open water 1s absent. ERIOPHORUM GRACILE IN IRELAND De CONCLUSIONS Although this survey suggests that there may have been a small decline in the number of Irish E. gracile sites in the past quarter of a century, it has also shown that the species is more common than was previously thought. Further intensive searches of the unconfirmed sites are recommended before it can be stated categorically that these populations have become extinct. Although this paper records 13 new sites for E. gracile in Ireland, it 1s felt that the species should remain on the Irish list of protected plant species. The main reasons for the continued protection of the species are (1) the decline of the species throughout the rest of Europe (Curtis & McGough 1988), (2) the relatively threatened nature of the species’ habitat in Ireland and (3) the fact that many populations are small and vulnerable to habitat interference. It is imperative that the remaining Irish sites be monitored on a regular basis to ensure that they are protected from damaging Operations. Experiences in other countries (Winship 1994; Decker et al. 2006) have shown that protection of the existing hydro- logical balance within sites is needed in order to ensure the long-term survival of E. gracile. ACKNOWLEDGMENTS We wish to thank Jim Ryan and Lady Rosemary Fitzgerald for permitting the use of their unpublished records. Thanks are also due to Dr Tom Curtis of the National Parks and Wildlife Service, Dublin who provided much help throughout the survey. Much of this research formed part of a Ph.D. thesis at the National University of Ireland, Galway and thanks for financial support are due to the National Parks and Wildlife Service, the National University of Ireland Galway and Donegal County Council. We are indebted to an anonymous referee for useful comments on an earlier draft of the paper. REFERENCES CLAPHAM, A. R., TUTIN, T. G. & MOORE, D. M. (1987). Flora of the British Isles (3rd edn). Cambridge University Press, Cambridge. CONAGHAN, J. P. (1995). The ecology of Eriophorum gracile Koch ex Roth and Eriophorum latifolium Hoppe in Ireland. Ph.D. Thesis, National University of Ireland. COUNCIL OF EUROPE (1983). List of rare, threatened and endemic plants in Europe. Nature and Environment Series, No. 27, Strasbourg. Cox, J. R. (1995). The conservation and management of Slender Cottongrass Eriophorum gracile in England. Hampshire and Isle of Wight Wildlife Trust, Eastleigh. CURTIS, T. G. F. & MCGOUGH, H. N. (1988). The Irish Red Data Book. 1. Vascular Plants. Stationery Office, Dublin. DECKER, K, CULVER, D. R. & ANDERSON, D. G. (2006). Eriophorum gracile W. D. J. Koch (slender cottongrass): a technical conservation assessment. [Online]. U.S.D.A. Forest Service, Rocky Mountain Region. DOUGLAS, C. & GROGAN, H. (1987). Survey to Locate Lowland Blanket Bogs of Scientific Interest in Connemara, Co. Galway. Internal Report, National Parks and Wildlife Service. FLORA (PROTECTION) ORDER 1999 S.I. No. 94/1999. Government Publications, Dublin. HULTEN, E. (1962). The Circumpolar Plants. 1. Vascular Cryptograms, Conifers, Monocotyledons. Almquist and Wiksell, Stockholm. IRISH PEATLAND CONSERVATION COUNCIL (1992). Irish Peatland Conservation Council Policy Statement and Action Plan 1992-1997. Irish Peatland Conservation Council, Dublin. KERTLAND, M. P. H. (1971). Eriophorum gracile Roth. in Ireland. Jrish Naturalists’ Journal 17: 62. LOCKHART, N. D. (1999). Paludella squarrosa (Hedw.) Brid., a Boreal relic moss new to Ireland. Journal of Bryology 21: 305-308. O’CONNELL, M. (1981). The phytosociology and ecology of Scragh Bog, Co. Westmeath. New Phytologist 87: 139-187. PATON, J. A. (1999). The liverwort flora of the British Isles. Harley Books, Colchester. ROSE, F. (1967). Eriophorum gracile Roth. new to Ireland. Jrish Naturalists’ Journal 15: 361-362. SCANNELL, M. P. J. & SYNNOTT D. M. (1987). A Census Catalogue of the Flora of Ireland (2nd edn). The Stationery Office, Dublin. SCANNELL, M. P. J., SYNNOTT D. M. & WEBB, D. A. (1968). Eriophorum gracile in Ireland. Irish Naturalists’ Journal 16: 12. SMITH, A. J. E. (2004). The Moss Flora of Britain and Ireland. Cambridge University Press, Cambridge. 238 J.P. CONAGHAN AND M. SHEEHY SKEFFINGTON STACE, C. (1997). New Flora of the British Isles. 2nd ed. Cambridge University Press, Cambridge. STELFOX, A. W. (1970). A possible early record for Eriophorum gracile from West Galway by C. C. Babington in 1836. Irish Naturalists’ Journal 16: 282. TALLIS, J. H. (1983). Changes in wetland communities, in A. P. J. GORE, ed. Ecosystems of the World; 4A, Mires: Swamp, Bog, Fen and Moor pp. 311-347. Elsevier, Amsterdam. TUTIN, T. G., HEYwoop, V. H., BURGES, N. A., MoorE, D. M., VALENTINE, D. H., WALTERS, S. M. & WEBB, D. A. (1980). Flora Europaea, Volume 5, Cambridge University Press, Cambridge. VAN GROENENDAEL, J. M., HOCHSTENBACH, S. M. H., VAN MANSFELD, M. J. M. & ROOZEN, A. J. M. (1979). The influence of the sea and of parent material on wetlands and blanket bog in west Connemara, Treland. M.Sc. Thesis, Catholic University Nijmegen. VAN GROENENDAEL, J. M., VAN MANSFELD, M. J. M., ROOZEN, A. J. M. & WESTHOFF, V. (1993). Vegetation succession in lakes in the coastal fringe of West Connemara, Ireland. Aquatic Conservation: Marine and Freshwater Ecosystems 3: 25-41. WHEELER, B. D. (1983). Vegetation, nutrients and agricultural land use in a north Buckinghamshire valley fen. Journal of Ecology 71: 249-256. WIGGINTON, M. J., ed. (1999). British Red Data Books 1: Vascular Plants. 3rd ed. Joint Nature Conservation Committee, Peterborough. WINSHIP, H. (1994). The conservation of Slender Cottongrass Eriophorum gracile Koch ex Roth in England. Hampshire and Isle of Wight Wildlife Trust, Eastleigh. (Accepted December 2008) Watsonia 27: 239-242 (2009) 239 Neotypification of Aira setacea Hudson (Poaceae) J.O. CHIAPELLA IMBIV, Universidad Nacional de Cordoba, Vélez Sarsfield 1601 X5016GCA Cordoba, Argentina ABSTRACT Deschampsia setacea (Hudson) Hackel is a slender grass described originally as Aira setacea Hudson on a plant collected in Stratton Strawless (Norfolk) in the first edition of the Flora Anglica. It has never been typified, and since the herbarium of Hudson has been lost to fire, a neotype is chosen in order to preserve the current application of the name. KEYWORDS Deschampsia_ setacea, Hudson, Neotypification, Flora Anglica INTRODUCTION The preparation of a revision of Deschampsia (Poaceae: Aveneae) in South America has made evident the lack of an extant type specimen for Deschampsia setacea (Hudson) Hackel. The “Bog Hair-grass’ is a slender grass occasionally found in bogs, pools and ditches in Great Britain (Perring & Walters 1962; Stace 1997). Outside Britain D. setacea is found in Ireland, Central Europe (Conert 1987; Jansen 1951) and Scandinavia (Hylander 1953). Some scattered populations also exist in central Chile (Parodi 1949), thus giving this taxon a disjunct distribution between Europe and _ South America. Deschampsia setacea (Hudson) Hackel is based on Aira setacea Hudson, which was described in the Flora Anglica (Hudson 1762) on the basis of a plant collected in East Anglia. CHARACTERS USEFUL IN DISTINGUISHING DESCHAMPSIA SETACEA FROM OTHER DESCHAMPSIA SPECIES Deschampsia setacea is a perennial cespitose grass with densely packed erect bristle-like blades 2:5-10 cm long, 1-1:'5 mm _ wide, inrolled, sharply pointed, scabrous on mid and lateral nerves of both abaxial and adaxial sides. The ligules are narrowly lanceolate, 4:5—11 mm long, hyaline, acuminate, and the panicles E-mail: jchiapella@imbiv.unc.edu.ar loose, lanceolate, 8-18 cm long, 1:5-5 cm wide, with spikelets 2-flowered, purplish, often clustered at the end of branches, the lower glume is narrowly lanceolate, (3)-4—-5 mm long, I-nerved, the upper glume lanceolate (3:5)—4:5—6 mm long, the lower lemma is 3—4 mm long, 4—-(5) toothed, teeth irregular, the lateral longer; the awn bent and twisted, 4:5—6 mm long, inserted in the inferior third or at the base, brownish in the lower half, purple in the upper half, scabrous. It can be readily distinguished from the widespread D. cespitosa by the leaves, which are almost filiform in D. setacea and up to 5 mm wide in D. cespitosa, and the narrower and shorter inflorescences than that of D. cespitosa. The awn differs too in the two species, being generally more bent and exserted in D. setacea and nearly straight and barely exceeding the glumes in almost all specimens of D. cespitosa. Another related taxon, Avenella flexuosa (L.) Drejer (formerly treated as Deschampsia flexuosa (L.) Trin.) differs by the obtuse ligules, the larger panicles and by forming rhizomes. All the mentioned taxa are remarkably variable and show often clinal variation along environmental gradients. PROTOLOGUE OF HUDSON’S AIRA SETACEA The protologue of Aira setacea Hudson (Fig. 1) does not mention a particular specimen or illustration to be selected as type material (ICBN Art. 9.2, McNeill et al. 2006), only indicating the locality of the type collection: ‘Habitat in ericeto Strattoniensi in comitatu Norfolk, D. Stillingfleet reperit’, roughly translated as [Habitat in peat heath in Stratton, County Norfolk, collected by Stillingfleet]. SELECTION OF A NEOTYPE While the specimens adjoined to Hudson located at other herbaria (CGE, E, K, LINN). are not those on which he made his original 240 J. O. CHIAPELLA fetacea, 6. AIRA folits fetaceis, panicula feeunda coarctata, a- riftis flofeulis duplo longtoribus recurvis: Anglis, {nvall leaved Hair-grals. Fiabitat in ericeto Strattonienfi i comitatu Norfolk. D. otllingfleet reperit, 2. VE. VII. = FIGURE 1. Protologue of Aira setacea Hudson (FI. Anglica 1: 30. 1762), photograph taken from an original copy held at the Library of the Institute of Botany, University of Vienna, Austria. @ 2000+ @ 1987-1999 e 1970-1986 e 1930-1969 ° -1929 FIGURE 2. Distribution of D. setacea (Huds.) Hackel in Britain & Ireland 10 km squares using size scaled dots to show different date classes, with progressively larger dots for more recent records. Where a plant has been recorded more than once in a grid square, only the most recent record is mapped. a NEOTYPIFICATION OF AJRA SETACEA 241 "WE NATURAL POETRY MUSEU LOnOOe: nH WEL ll HERB, MUS. BRIT. copyright reserved FIGURE 3. Type specimen of Aira setacea Huds. [=Deschampsia setacea (Huds.) Hack]. 242 J. O. CHIAPELLA descriptions (Stafleu & Cowan 1979), most of Hudson’s herbarium is considered lost to the fire that destroyed his house in 1783 (Dixon 1959), and the remaining specimens (excluding algae) are located at the Buddle Herbarium in BM (M. Spencer, pers. comm.). Inquiries with curators of the herbaria above mentioned resulted in no single sheet of Aira setacea, thus allowing for the selection of a specimen not belonging to Hudson’s collection. The recom- mendations ICBN 9A and 9B on designation of lecto and neotypes state that all aspects of the protologue should be considered for the selection of the new type material; in the present case the only clue is the place of the original collection, a locality in Norfolk close to Norwich, Stratton Strawless. A search in BM yielded a specimen collected in Norfolk, Litcham Common, ‘Ex Herb. F. J. Hanbury, 18th July 1883’ (Fig. 3), which is a good example of the species and therefore design- ated here as Neotype, in agreement with ICBN Art. 9.14. Deschampsia setacea was _ last recorded in Norfolk in 1974 (Fig. 2), and probably represents a case of local extinction due to either urban or agrarian development common in nearby regions of Cambridgeshire (Preston 2000). ACKNOWLEDGMENTS I am very grateful to M. Spencer (BM) for helping to find a specimen, and for providing the image, A. Lockton for preparing the distribution map and S. Renvoize and T. Cope (K) for checking the specimen. Curators at CGE, E, K, LINN helped with the search of Hudson’s specimens, M. Sanford critically read the manuscript, helping to improve it. REFERENCES DIXON P. S. (1959). Two important algal herbaria. Brit. Phycol. Bull. 1: 35-42. HUDSON, W. (1762). Flora Anglica 1. C. Moran, London. HYLANDER, N. (1953). Nordisk Kdrlvdxtflora. Almqvist & Wiskell, Stockholm. JANSEN, P. (1951). Gramineae, in WEEVERS, T., HEIMANS, J., DANSER, B. H., KLoos, A. W., VAN OOSTROOM, S. J. & WACHTER, W. H., eds. Flora Neerlandica, pp. 1-274. Royal Botanic Gardens, Amsterdam. MCNEILL, J., BARRIE, F. R., BURDET, H. M., DEMOULIN, V., HAWKSWORTH, D. L., MARHOLD, K., NICHOLSON, D. H., PRADO, J., SILVA, P. C., SKOG, J. E. & WIERESEMA, J. H. (2006). International Code of Botanical Nomenclature (Vienna Code). Regnum Vegetabile 146. A.R.G. Gantner Verlag KG. PARODI, L. R. (1949). Las gramineas Sudamericanas del género Deschampsia. Darwiniana 8: 415-475. PERRING, F. H. & WALTERS, S. M. (1962). Atlas of the British Flora. Thomas Nelson Ltd., London. PRESTON, C. D. (2000). Engulfed by suburbia or destroyed by the plough: the ecology of extinction in Middlesex and Cambridgeshire. Watsonia 23: 59-81. i STACE, C. A. (1997). New Flora of the British Isles, 2nd Ed., Cambridge Univ. Press, Cambridge. STAFLEU, F. A. & COWAN, R. S. (1979). Taxonomic Literature IT. W. Junk, The Hague. (Accepted November 2008) Watsonia 27: 243-257 (2009) 243 Notes BRITISH SPECIES OF RUBUS L. (ROSACEAE) IN MAINE AND NORTH BRITTANY, FRANCE For the latest in a long-running series of cross- Channel forays to investigate how far, if at all, British representatives of the Rubus fruticosus aggregate have ranges that extend to the nearer parts of France, two sharply contrasting areas were chosen in 2006. The last week in June proved ideal timing that season to catch the group in full flower in the largely agricultural belt between Normandy and the Loire, the northern part of which holds the two départements, Mayenne and Sarthe, that comprise the administrative region (and former province) of Maine. Extensive study of the Rubus flora of the département immediately to the south, Maine-et-Loire - roughly coinciding with the former province of Anjou — was undertaken during the first quarter of the twentieth century by one of France’s then leading specialists in the group, the Angers- based G. Bouvet; that area above it, however, seems to have been left virtually unexplored. The particular part of it selected on this occasion, approximately midway between the cities of Laval and Le Mans, has large tracts of basic soils, with poor Rubus diversity as a consequence; but the primarily Maine-et-Loire speciality that Bouvet named R. andegavensis (after Angers) is abundant and conspicuous in the hedges on those, in compensation. Con- siderable broad-leaved woodland, some clearly of very considerable age, nevertheless provides oases of interest in the more acid parts. Though at least half the Rubus species encountered were ones not known to occur in Britain, three of the commonest — R. questieri Lef. & P. J. Mueller, R. adscitus Genev. and R. rubritinctus W. C. R. Watson — are shared with England, though their range on the north side of the Channel is markedly south-western. To a British batologist they thus give the Rubus florula of Maine (and of the regions of France to the east as well) an unexpectedly western tilt. Adding emphasis to that, two species with a more south-eastern range in England, R. asperidens (Sudre ex Bouvet) Bouvet and R. flexuosus P. J. Mueller & Lef., are less in evidence, and on the whole less plentiful than in comparable habitats across the Channel. Well behind all those in frequency, however, and mostly represented by a single bush at best, are such very widespread English species as R. bertramii G. Braun, R. plicatus Weihe & Nees, R. pyramidalis Kaltenb., R. cardiophyllus Lef. & P. J. Mueller and R. echinatus Lindley, the last two, seen only in field hedges on the sum- mit of the principal hill, Mont Richard (363 m), probably brought from afar by birds. That greater affinity with England’s south- western Rubus florula than its south-eastern one that the brambles of Maine display is all the more obvious, as is only to be expected, on the northern seaboard of Brittany, the other region chosen for investigation — a week later, given that degree of retardation imposed on the Rubus flowering period by its more northern position. On a previous visit, in 1999, it had nevertheless come as a surprise how strongly south-western, “Cornubian’ even, that English affinity is along the comparatively sheltered, west-facing stretch of that coastline that is popularly known as the Céte de Granit Rose. Chiefly responsible for that is the prevalence and even local abundance there of R. hastiformis W. C. R. Watson. Two further species, up till then similarly believed endemic to the south-west of the British Isles, R. aequalidens Newton and R._ peninsulae Rilstone, were also met with there on that earlier occasion but very much more sparsely (Allen 2002). A wider search, to check the accuracy of that initial impression and in the hope of finding at least one further compatriot of those, was many years overdue. In the combined list below of the more noteworthy finds in one or other region 7 indicates a specimen from the locality in question has been seen also by A. Newton and the determination confirmed by him. All voucher material has been lodged in BM. MAINE R. elegantispinosus (A. Schumach.) H. E. Weber Mayenne: Bois de la Forge, Chammes, colony in a clearing near the south margin. This is now the fourth wood (and fourth département) in the general region of the middle Loire in which finds of this species 244 NOTES Watsonia 27 (2009) have been made since 1994 (contrary to the statement in Allen 2007), earlier herbarium specimens having been detected from Sarthe (Forét de Pincé, Sablé) and Loire-et-Cher (Forét de Boulogne, nr. Chambord), a solitary clump in each case. This virtually exclusive woodland distribution is sharply at variance with the situation in Britain, where the species is patently a garden escape and more or less confined to waste places in urban areas. The possibility that it is indigenous in that part of France seems to deserve consideration. R. neomalacus Sudre Mayenne: grassy roadside by south margin of Bois de la Forge; Bois de Bergault. Sarthe: along roadside margin of Forét de Grande Charnie W. of Etival-en-Charnie. Single clump or patch in each case. These finds back up the two recorded from Maine-et-Loire by Bouvet (1912), suggesting that this region constitutes a ‘stepping-stone’ between the two strikingly disparate centres otherwise known for this species: the mouth of the Loire and the west half of Surrey, v.c. 17. R. trichodes W.C.R. Watson — Sarthe: margin of broad-leaved woodland by Les Faucheries cross-lanes, Forét de Grande Charnie +. Third record for France, extending the known Conti- nental range some 300 km south-westwards. R. botryeros (Focke ex Rogers) Rogers Mayenne: clump in field hedge, summit of Mont Rochard. A strongly western species near its eastern limit here. R. tamarensis Newton Mayenne: one patch, Bois de la Forge. A semi-western species with isolated outliers southwards from its centre in Normandy’s Orne Valley. R. scaber Weihe Mayenne: one bush, Bois de la Forge. Apparently very rare in N. W. France, in contrast to its frequency in southern England and very incongruous local abundance in the west of Connaught in Ireland. NORTH BRITTANY Coéte de Granit Rose (dép. Cdtes-du-Nord): R. leuchostachys Schleicher ex Sm. Patch among bracken on cliff-top, public footpath between Truzugal and Kerbalanec, N. W. of Louannec — the same unusual habitat for this species in which it occurs, apparently exclu- sively, in Guernsey and Sark (Allen 2001). Technically the first record for the European mainland, but subsequently a specimen coll- ected by Genevier (Vendée: Mortagne, 1857, sub nom. R. vestitus) has come to light in herb. Déséglise in NMW (xerox copy in BM). R. aequalidens Newton Locally common 2 km inland in plantations and hedges around Varac’h Chateau, S. of Louannec; a bush in a Thuja hedge | km E. of Beg Léguer. (Also common in woodland divided by D786 S. W. of Lanmeur, 8 km into dép. Finisteére.) R. peninsulae Rilstone Two populations c. 200 m apart in scrub and full sun, cliff-top public footpath between Truzugal and Kerbalanec; in and around Marais du Quellen (wooded lagoon, now a nature reserve, on N. outskirts of Trébeurden)+; one bush in hedge of house by Beg Léguer viewpoint. R. venetorum D. E. Allen Though common over most of Brittany, anomalously rare on this stretch of coast: patch among bracken, roadside S. of Marais du Quellen, Trébeurden. Continual re-scrutiny of herbarium material collected on previous cross-Channel forays has also resulted in one further addition for mainland Europe: R. lanaticaulis Edees & Newton Manche: hedge of D122 W. of La Glacerie (village), 3-5 km S. of Cherbourg, 1987 (BM)7y. A starved example of this common species of western Britain and Ireland long expected to be found in Normandy and/or Brittany. REFERENCES ALLEN, D. E. (2001). Brambles (Rubus L. sect. Rubus and sect. Corylifolii Lindley (Rosaceae) of the Channel Islands. Watsonia 23: 421-435. ALLEN, D. E. (2002). A third list of British species of Rubus L. (Rosaceae) in north-west France. Watsonia 24: 220-222. ALLEN, D. E. (2007). A fourth list of British species of Rubus L. (Rosaceae) in north-west France. Watsonia 26: 485-486. BOUVET, G. (1912). Florule des Rubus de I’ Anjou, pt. 2. Bulletin de la Société d’Etudes scientifiques d’Angers n.s. 41: 9-36. D. E. ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ NOTES Watsonia 27 (2009) 245 THE STATUS OF RUBUS BABINGTONIANUS W. C. R. WATSON (ROSACEAE) Like many Rubus species, R. babingtonianus has had a chequered history since it was first determined as R. althaeifolius of Host. by Babington in Babingtons’ British Rubi in 1869. Babington assigned a number of sites for it round Cambridge but also gave it a much wider range in England. W. Moyle Rogers does not mention the plant at all in his Handbook of British Rubi published in 1900. W.C. R. Watson gave the plant a new name R. babingtonianus in 1946 with the note “Rubus althaeifolius Bab 1869 non Host. 1823” However, in Rubi of Great Britain and Ireland (1956), Watson gives a distribution covering 14 British vice counties from Kent to South Uist and one Irish v.c. Edees & Newton (Brambles of the British Isles 1988), having sorted the wheat from the chaff, gave the plant local endemic status almost confined to Cambs v.c. 29 with one record just in West Suffolk v.c. 26, and this status remained in place in the Atlas of British and Irish Brambles (Newton & Randall 2004) with just one extra hectad dot since the 1988 publication. The present author had not knowingly seen Rubus babingtonianus so was delighted when A. Newton gave him a grid reference for a site between Chippenham and Snailwell in east Cambridgeshire, where he himself had coll- ected specimens. A visit was made on 6 August 1997, speci- mens collected and these were duly confirmed by A. Newton. On 16 July 2001 whilst paying a visit to some “Breckland rarities’, a stop for lunch was made beside a Forest Enterprise plantation at Barton Mills v.c. 26 and, after eating, a stroll down the road revealed a low thicket of brambles that had the appearance of Rubus babingtonianus. Once seen, it was found again less than three weeks later whilst plant recording for the Flora of Suffolk on 4 August 2001 in an area at Thetford known as the Horse Meadows (administrative Norfolk but v.c. 26). A third site was found on 20 August 2001, this time at Thompson Common, a_ Norfolk Wildlife Trust reserve in v.c. 28. All were confirmed by A. Newton. Since that date, new sites have been found almost annually in v.cc. 25, 26, 27 and 28 as well as a colony found with Dr A. C .Leslie deep in the Fens at Littleport v.c. 29. The statement in Edees & Newton (1988) that it is a plant of clay soils appears to be not strictly true as the subsequent finds have covered a wide range of soil types; from sandy Breck to chalky clay to peaty fen, though it does seem to like stream banks and the margins of wet ditches and pools. Once acquainted with Rubus babingtonianus it is quite an easy plant to pick out, especially as it grows in areas where other species are limited in diversity. In addition to the regularly lobate leaf margin, the panicle becomes increasingly compound as the season advances and the neat pink flowers can be found from late May to early August when most other Corylifolii have long since gone over. It is also remarkably constant throughout its range. No attempt has yet been made to map all the sites for Rubus babingtonianus but in order to confirm its status as a regional rather than a local endemic, there follows a list of exsiccatae to be found in Herb. A. L. Bull most of which have been approved by A. Newton. Vice-county Date Location Grid Ref. 25 East Suffolk 17 06 2006 Wortham Ling TM0879 07 08 2004 Hedgerow, Wilby TM 2370 26 West Suffolk 28 07 2004 Castlings Heath, Groton TL9743 26 07 2003 Coney Weston Crossroads TL9438 16 07 2001 Roadside, Barton Mills WEI Z| 04 08 2001 Horse Meadows, Thetford TL8483 27 East Norfolk 24 06 2006 Saxlingham Road, Bayfield TG0340 31 07 2004 Bylaugh, near Swanton Morley bridges TG0218 09 09 2004 Stokesby road Billockby near the Muck Fleet TG4111 _ bridge 28 West Norfolk 2008 2001 Thompson Common TL9396 11 08 2008 By bridge at Madhouse Plantation Tottington TL9094 246 NOTES Watsonia 27 (2009) REFERENCES BABINGTON, C. C. (1869). Babington’s British Rubi. J. Van Voorst, London. EDEES, E. S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. MOYLE ROGERS, W. (1900). Handbook of British Rubi Duckworth, London. NEWTON, A. & RANDALL, R. D. (2004). Atlas of British & Irish Brambles. Botanical Society of the British Isles, London. WATSON, W. C. R. (1956). Rubi of Great Britain and Ireland. Cambridge University Press, Cambridge A. L. BULL Hillcrest, East Tuddenham, Dereham, Norfolk, NR20 3JJ ELEVEN NEW COMBINATIONS IN THE BRITISH FLORA Some new combinations are required for forthcoming floristic works. 1. Ficaria Schaeff. Molecular work has confirmed that Ficaria Schaeff. should be treated as a separate genus from Ranunculus L., so that R. ficaria L. should now be known as F. verna Huds. More- over Laegaard (2001) demonstrated that the type of Ranunculus ficaria is the tetraploid taxon hither to known as subsp. bulbilifer Lambinon, not the fertile diploid as concluded by Sell (1994). The correct subspecific epithets for the tetraploid and diploid respectively are therefore verna and fertilis. There appears to be no valid combination for the latter: Ficaria verna Huds. subsp. fertilis (Lawralrée ex Laegaard) Stace, comb. nov. Basionym: Ranunculus ficaria L. subsp. fertilis Lawralrée ex Laegaard, Nordic J. Bot. 20: 526 (2001) Laegaard (2001) actually gave the authority for his combination as “A. R. Clapham ex Laegaard”, but in fact Clapham published his invalid fertilis as a variety, not as a subspecies; Lawralrée later published it illegitimately as a subspecies (Sell 1994). Of the other two subspecies recognised in Britain, the combination F. verna_ subsp. ficariiformis (F. W. Schultz) B. Walln. already exists, but subsp. chrysocephala requires validation: Ficaria verna Huds. subsp. chrysocephala (P. D. Sell) Stace, comb. nov. Basionym: Ranunculus ficaria L. subsp. chrysocephalus P. D. Sell, Bot. J. Linn. Soc. 106: 117 (1991). 2. Dysphania R. Br. The glandular species of Chenopodium are now generally segregated into the genus Dysphania, under which two casual hybrids require combinations: Dysphania carinata (R. Br.) Mosyakin & Clemants (C. carinatum R. Br.) x D. cristata (F. Muell.) Mosyakin & Clemants (C. cristatum F. Muell.) = Dysphania x bontei (Aellen) Stace, comb. nov. Basionym: Chenopodium x_ bontei Aellen, Verh. Nat. Ges. Basel 44: 317 (1933) Dysphania pumilio (R. Br.) Mosyakin & Clemants (C. pumilio R. Br.) x OD. carinata (R. Br.) Mosyakin & Clemants (C. carinatum R. Br.) = Dysphania x christii (Aellen) Stace, comb. nov. Basionym: Chenopodium x christii Aellen, Verh. Nat. Ges. Basel 44: 317 (1933) 3. Callitriche L. With justification Lansdown (2006) relegated Callitriche hamulata to an infraspecific variant of C. brutia, on the grounds that the two taxa are not distinguishable unless fruiting in dry or damp (as opposed to aquatic) conditions, when C. brutia alone develops fruit-stalks 2-10 mm long. The presence or absence of fruit-stalks is, however, correlated with a striking difference in chromosome number: 2n = 28 in C. brutia; 2n = 38 in C. hamulata. This correlation has been demonstrated in 29 accessions (6 of C. brutia, 23 of C. hamulata) according to data in the B.S.B.I. Cytological Database. Accordingly, I consider that subspecific status is more appropriate than the varietal status chosen by Lansdown. Callitriche brutia Petagna subsp. hamulata (Kiitz. ex W. D. J. Koch) Stace, comb. nov. Basionym: C. hamulata Kitz. ex W. D. J. Koch, Syn. Fl. Germ. 246 (1837) Synonyms: C. intermedia WHoffm. subsp. hamulata (Kiitz. ex W. D. J. Koch) A. R. Clapham, in Clapham, Tutin & E. Warb., Fl. Brit. Isl. 622 (1952) C. brutia var. hamulata (Kitz. ex W. D. J. Koch) Lansdown, Watsonia 26: 113 (2006) NOTES Watsonia 27 (2009) 247 4. Mimulus L. The plant described by Clos as a variety of M. luteus is better recognised at species level according to Silverside (1994); the combination is made here: Mimulus nummularius (Clos) Stace, comb. nov. Basionym: Mimulus luteus L. var. nummularius Clos, in Gay, Fl. Chil. 5: 140 (1849) 5. x Schedolium Holub The three broad-leaved species of Festuca L. are now segregated into the genus Schedonorus P. Beauv., and their hybrids with species of Lolium L. have been allocated to nothogenus x Schedolium. Combinations under the latter have already been provided by Holub (1998) and Scholz (2007) for four of the six hybrids between the three species of Schedonorus (S. pratensis, S. arundinaceus and S. giganteus) and Lolium perenne and L. multiflorum; the other two are made here: Schedonorus pratensis (Huds.) P. Beauv. (Festuca pratensis Huds.) x Lolium multiflorum Lam. = x Schedolium braunii (K. Richt.) Stace, - comb. nov. Basionym: Festuca x braunii K. Richt., Pl. Eur. 1: 103 (1890) Synonym: x Festulolium braunii (K. Richt.) A. Camus, Bull. Mus. Hist. Nat. (Paris), Sér I, 33: 538 (1927) Schedonorus giganteus (L.) Holub (Festuca gigantea (L.) Vill.) x Lolium multiflorum Lam. = x Schedolium nilssonii (Cugnac & A. Camus) Stace, comb. nov. Basionym: x Festulolium nilssonii Cugnac & A. Camus, Bull. Soc. Bot. France 91: 19 (1944) 6. x Dactylodenia Garay & H. R. Sweet When the three British Gymnadenia taxa are treated as separate species rather than as sub- speeicssson. G. conopsea (L.) R. Br. the allocation and synonymy of the binomials for hybrids between them and species of Dactyl- orhiza Necker ex Nevski (= x Dactylodenia) have to be reconsidered. Of the ten combi- nations of x Dactylodenia known from the British Isles, five already have a legitimate binomial, four have no epithets that can be applied, and one can be provided with a legitimate binomial by the following new combination: x Dactylodenia evansii (Druce) Stace, comb. nov. Basionym: Orchis x evansii Druce, Rep. Bot. Exch. Club Brit. Isles 2: 199 (1907) Synonyms: Habenaria x evansii (Druce) Druce, Rep. Bot. Exch. Club Brit. Isles 8: 875 (1929) x Orchigymnadenia evansii (Druce) T. & T. A. Stephenson, J. Bot. 60: 35 (1922) = Gymnadenia borealis (Druce) R. M. Bateman, Pridgeon & M. W. Chase x Dactylorchis maculata (L.) So6 The above new combination was previously published invalidly and inaccurately by Avery- anov in Bot. Zhurn. (Moscow & Leningrad) 71: 93 (1986). The type of Orchis x evansii came from Berwickshire, where G. borealis rather than G. conopsea occurs. The publication of legitimate binomials for the remaining four combinations demands the collection of new material in order to provide type specimens. 7. Dactylorhiza Necker ex Nevski Wilmott’s Orchis kerryensis remains a distinctive taxon, best recognised as a variety of Dactylorhiza occidentalis, for which a new combination is required: Dactylorhiza_ occidentalis (Pugsley) P. Delforge var. kerryensis (Wilmott) R. M. Bateman & Denholm, comb. nov. Basionym: Orchis kerryensis Wilmott, Proc. Linn. Soc. Lond. 148: 126 (1936) 8. x Anacamptorchis E.G. Camus The transfer of Orchis morio to Anacamptis means that the hybrid O. mascula x O. morio becomes an _ imntergeneric hybrid under x Anacamptorchis, for which the new combination is made: x Anacamptorchis morioides (Brand) Stace, comb. nov. Basionym: Orchis x morioides Brand, Syn. Deut. Schweiz. Fl., ed. 3, 3: 2427 (1905) = Anacamptis morio (L.) R. M. Bateman, Pridgeon & M. W. Chase (O. morio L.) x Orchis mascula (L.) L. ACKNOWLEDGMENT I am grateful to Katherine Challis for inform- ation concerning the publication of Orchis x morioides Brand, and to Richard Bateman and Ian Denholm for publishing their new combination in this note. 248 NOTES Watsonia 27 (2009) REFERENCES Hows, J. (1998). Reclassifications and new names in vascular plants, 1. Preslia, Praha 70: 97-122. LAEGAARD, S. (2001). Validation of Ranunculus ficaria L. subsp. fertilis Clapham ex Laegaard. Nordic J. Bot. 20: 525-526. LANSDOWN, R. V. (2006). Notes on the water-starworts (Callitriche) recorded in Europe. Watsonia 26: 105— 120. SCHOLZ, H. (2007). xSchedolium krasanii — ein neuer Nothospeziesname. Kochia 2: 51-52. SELL, P. D. (1994). Ranunculus ficaria L. sensu lato. Watsonia 20: 41-50. SILVERSIDE, A. J. (1994). Mimulus: 180 years of confusion, in PERRY, A. R. & ELLIS, R. G., eds. The Common Ground of Wild and Cultivated Plants, pp. 59-64. National Museum of Wales, Cardiff. C. A. STACE Cringlee, Claybrooke Road, Ullesthorpe, Leics, LEI7 5AB THE RE-DISCOVERY OF ARENARIA NORVEGICA SUBSP. NORVEGICA IN IRELAND The only Irish record for Arenaria norvegica was made by the late John [Jack] Heslop Harrison in June 1961 during a visit to the Burren, County Clare (H9), with a party of botany students from the University of Birmingham (Heslop Harrison, Wilkins & Green 1961). During their stay they discovered a small colony growing in “shallow crevices and solutions hollows on an area of limestone pavement” at c. 800 ft (244 m) on “the south slope of Gleninagh Mountain overlooking Caher Lower” (Heslop Harrison et al. 1961). The precise locality is unclear as Gleninagh Mountain does not have a southern slope, but extends southwards in a long ridge to the east of Caher Lower. A more likely locality occurs on the adjacent mountain of Carnsefin at c. M164092 “or not far away” on a series of south-facing limestone pavements (Webb & Scannell 1983) but despite repeated searches no plants have been found in either locality. This led some botanists to conclude that the plant is either extinct or that the original record was erroneous (Webb & Scannell 1983; Curtis & McGough 1988). There is no doubt, however, as to the identity of Heslop Harrison’s specimen: this was shown to Geoffrey Halliday who counted the chromosomes of a_ plant grown from seed extracted from it (2n = 80) (G. Halliday, pers. comm.). This confirmed the presence of Arenaria norvegica subsp. norvegica for Ireland, a taxon then only known from Scotland, Norway, Sweden and Iceland. On 25 May 2008 two small colonies of A. norvegica subsp. norvegica were discovered a few kilometres away from Gleninagh Mountain on the southwest slopes of Carnsefin Mountain overlooking the coast at Fanore. The first colony, of around 10-20 plants, was confined to Carboniferous limestone outcrops in the middle of the “green road” between Morroogh and Black Head. Plants were growing in shallow gravel filled solution hollows extending over 20 m of the track. Only a few other species were growing in the same habitat (Bellis perennis, Minuartia verna, Plantago lanceolata, Thymus polytrichus). All other species listed in Table 1 (quadrat 1) were associated with pockets of Festuca ovina- Sesleria caerulea grassland. The community most closely resembled Festuca ovina- Minuartia verna_ grassland (67% fit to Cladonia species sub-community, OV37c,) and was almost identical to sites in Yorkshire where Arenaria norvegica subsp. anglica grows on a green lane on _ Ingleborough (Walker 2000). A second larger population was_ located nearby on limestone exposures at a slightly higher altitude (c. 100 m). Here, around 30 flowering plants extended over 50 m growing in soil and gravel-filled crevices and solution hollows adjacent to Dryas heath with abundant Gentiana verna and Helianthemum oelandicum subsp. canum. Again the vegetation was very open with scattered plants of Arenaria norvegica, Agrostis capillaris, Asperula cynanchica, Carex flacca, Festuca ovina, Saxifraga tridactylites and Thymus polytrichus (Table 1; quadrats 2 & 3). This community most closely resembled Festuca ovina-Agrostis capillaris-Thymus polytrichus — grassland (CG10c; 26 and 34% fit to Saxifraga aizoides- Ditrichum flexicaule sub-community). NOTES Watsonia 27 (2009) 249 % COVER OF SPECIES GROWING WITH ARENARIA NORVEGICA SUBSP. NORVEGICA IN THE BURREN, COUNTY CLARE (1 M* QUADRATS) Species 2 3) S)NSSIES | 2) 3 Achillea millefolium + + Festuca ovina 20 5) 5) Agrostis capillaris 2 5 Gentiana verna + + Antennaria dioica Helianthemum oelandicum + Anthyllis vulneraria 3 + Lotus corniculatus + + + Arenaria norvegica + + + Minuartia verna + Armeria maritima fe Plantago lanceolata + Asperula cynanchica + Plantago maritima + Bellis perennis + Prunella vulgaris + Briza media 1 Saxifraga tridactylites + Calluna vulgaris 2 Sesleria caerulea yy Campanula rotundifolia + Succisa pratensis 2 + Carex caryophyllea + Thymus polytrichus 2 3} 1 Carex flacca 7) i Viola rivinianna + + Carex pulicaris + Vegetation cover Ms) 20 IS) Dryas octopetala + Bare rock, soil and gravel nS 80 Species confined to solution hollows are highlighted in bold. Site 1, “ruts” of green road between Morroogh and Black Head, M15091021, alt. 80 m; sites 2 & 3, solution hollows on limestone pavement nearby, M15111059 & M15081060, alt. 100 m. Irish plants of Arenaria norvegica were similar to Scottish subsp. norvegica in having small flowers (<10 mm) with 4 styles and obovate leaves (subsp. anglica has larger flowers, greater than 11 mm, with 3 styles and narrowly ovate leaves (Stace 1997)). The leaves were characteristically succulent, glossy-green and had an indistinct midrib and a few basal cilia (greater than a third in A. ciliata subsp. hibernica). However, there were notable differences. Irish plants appear to flower about a month earlier than in Scotland as in May 2008 many capsules already contained seed. The flowers were also noticeably smaller. Although [Irish plants displayed the characteristic trailing habit of Scottish subsp. norvegica the leaves were noticeably smaller (2-5-4 mm) and more obovate. The reasons for these differences are unclear and would warrant further investigation. How Arenaria norvegica has _ eluded botanists for so long is a mystery. Heslop Harrison only realised the significance of his find on examining specimens at the end of a long day’s botanising and neither he, nor any of his students ever returned to confirm the locality (G. Halliday, pers. comm.). If their recollections of the precise locality were incorrect then subsequent botanists have been looking in the wrong place which is admittedly a very large area that “would take a dozen botanists...a week’s hard work to demonstrate conclusively that the plant was absent” (Webb & Scannell 1983). The 2008 discovery therefore confirms the presence of this rare arctic-montane species for Ireland at its most southerly station in the world (53°N). A more thorough examination of the adjacent slopes is planned and _ will hopefully reveal a more extensive population. ACKNOWLEDGMENTS We would like to thank Geoffrey Halliday for his comments on the original discovery and the botanical group who-helped us to survey both populations: Isabel Alonso, Margaret Bradshaw, Carl Borges, Jonathan Cox, David Heaver, Dawn Isaac, Chris Gibson, Phil Horton, Richard Jefferson, Carol Laverick, Mina Patel, Clare Pinches, Jill Sutcliffe, Mike Sutcliffe, Ann Skinner and Chris Walker. 250 NOTES Watsonia 27 (2009) REFERENCES CURTIS, T. F. G. & MCCouah, H. N. (1988). The Irish Red Data Book 1. Vascular Plants. Stationary Office, Dublin. HESLOP HARRISON, J., WILKINS, D. A. & GREENE, S. W. (1961). Arenaria norvegica Gunn, a species new to the Irish flora, in Co. Clare. Irish Naturalists’ Journal 13: 267-268. STACE, C. A. (1997). New flora of the British Isles. Second Edition. Cambridge University Press, Cambridge. WALKER, K. J. (2000). The distribution, ecology and conservation of Arenaria norvegica subsp. anglica Halliday (Caryophyllaceae). Watsonia 23: 197-208. WEBB, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and the Burren. Royal Dublin Society and Cambridge University Press, Cambridge. K. J. WALKER 97 Dragon Parade, Harrogate, North Yorkshire HGI 5DG S. WARD & S. PARR 14 Ballyvaughan Cottages, Green Road, Ballyvaughan, County Clare, Ireland WATER-BORNE DISPERSAL OF CARDAMINE BULBIFERA (L.) CRANTZ (BRASSICACEAE) IN MID-WEST YORKSHIRE (V.C. 64) In Britain Cardamine bulbifera (Coralroot) is a localised perennial herb of ancient woodlands, shady stream-sides and roadside banks in South East England (Showler & Rich 1993). It is classified as Nationally Scarce (Cheffings 2004) and has been recorded as a native in 143 tetrads in 25 10-km squares (Nicholson ef al. 2004). These populations are concentrated in two main areas: on dry, basic soils in beech woods in the Chilterns (Buckinghamshire, Hertfordshire and Middlesex) and_ wet, generally acidic woodlands of the High Weald (Kent and Sussex). Elsewhere in the British Isles, it has been recorded as a naturalised alien in a variety of habitats. These include large populations in the ancient Needwood Forest, Staffordshire (v.c. 39) where it was formerly considered to be native (Nicholson et al. 2004). The majority of established populations are referable to the alpine forma ptarmicifolia (DC) O. E. Schulz which has been confirmed from a number of vice-counties (South & North Devon, North Somerset, South Essex, Salop, Mid-west Yorks, Westmorland, Dumfries, Kincardinshire, County Dublin). This form, which originates from _ the mountains of Central and Eastern Europe, 1s often grown in gardens, where it can become invasive, and is easily distinguished from native plants (forma bulbifera) by the shape of the leaves, which are more ovate and asymmetrical in outline and with broadly serrate margins (see figures in Showler & Rich 1993; Rich & Jermy 1998). It is also taller with larger, browner bulbils produced in the leaf axils. This purpose of this note is to document the water-borne spread of this uncommon alien along one river valley near to Harrogate, Mid- west Yorks (v.c. 64) where it’s range is increasing rapidly via the dispersal of bulbils during floods. ORIGIN Cardamine bulbifera has been known from the banks of the Oak Beck (SE25), a small tributary of the River Nidd, to the west of Harrogate, since at least 1998 (Manson, 2003; Wallace, 2005). This population was thought to have originated from the Royal Horticultural Society's gardens at Harlow Carr through which a small tributary of the Oak Beck stream flows. This was confirmed in April 2007 when 1000s of plants were found along the rear perimeter fence of the gardens and in an adjacent area of woodland next to this stream. Here C. bulbifera was the dominant species over approximately 50 x 30 m area growing with a number of other garden plants that have also. become established along the lower reaches of the stream (e.g. Cardamine raph- anifolia, Doronicum pardalianches, Lysichiton americanus, Petastites japponicus, Ranunculus aconitifolius, Tellima grandiflora, Symphytum grandiflorum). The origin of the Coralroot was thought to be in peat brought with lily bulbs from Essex sometime in the 1980s (Manson 2003). It soon became invasive in the garden borders, where it still survives despite excessive weeding. NOTES Watsonia 27 (2009) USS FIGURE 1. The distribution of Cardamine bulbifera along Oak Beck and the River Nidd in v.c. 64. Arrows denote the direction of river flow and the size of circles the size of populations. DISTRIBUTION AND SPREAD A survey of the entire course of the Oak Beck and River Nidd, as far east as Knaresborough, was carried out by the author in April 2007 (Fig. 1). This showed C. bulbifera to be present along c. 5 km of the Oak Beck from its confluence with the Harlow Carr stream up to where it meets the River Nidd. It also occurs sporadically on both banks of the Nidd up to c. 6 km downstream from the Oak Beck as far as Knaresborough. Plants are more or less cont- inuously encountered along the upper stretches of the Oak Beck where extensive patches, in a few cases with 1000s of stems, occur on the shaded, sandy banks, islands or in damp woodland in the adjacent flood zone. In a few places the plant also grows above the flood zone aS well as in grassland bordering a residential estate. The species is much less common along the River Nidd where small populations occur in similar habitats, but also on sandy river banks with no tree cover. The reason for this may be the greater ‘scouring’ of the riverbanks during winter rains (Wallace 2005) but also because these colonies have had less time to become established than on the Oak Beck; it will be interesting to observe the behaviour of these small populations over the coming years. HABITATS AND ECOLOGY These established populations are almost exclusively confined to sandy soils within the narrow flood zone of the two main water- courses. These areas are subject to frequent ‘scouring’ and re-deposition of sand/silt during short floods which occur throughout the year. Consequently the habitats often show signs of disturbance with large areas often covered in deposited sand and tree litter. The ground flora is relatively uniform along both rivers, and is typical of Allium ursinum sub-community of Fraxinus excelsior-Acer campestre-Mercurialis perennis woodland (W8f; Rodwell 1991). Stands with C. bulbifera are usually shaded and dom- inated by a dense carpet of Allium ursinum, with Galium aparine, Geum urbanum, Ranun- culus ficaria and Urtica dioica. The canopy is usually a mixture of Acer pseudoplatanus, 252 NOTES Watsonia 27 (2009) TABLE 1. VASCULAR PLANT SPECIES RECORDED WITH CARDAMINE BULBIFERA IN 2 x 2 M QUADRATS IN MID-WEST YORKSHIRE (N = 11) Species % sites Canopy Acer pseudoplatanus 18 Alnus glutinosa 36 Corylus avellana 9 Crataegus monogyna 18 Fagus sylvatica 9 Ground flora Aegopodium podagraria 45 Allium ursinum 100 Anemone nemorosa 45 Angelica sylvatica 18 Anthriscus sylvestris 45 Brachypodium sylvaticum DY Cardamine flexuosa 27 Carex pendula D7 Species % sites Circaea lutetiana 18 Deschampsia cespitosa 27 Galium aparine 64 Geranium robertianum 36 Geum urbanum 3)5) Hedera helix 18 Heracleum sphondylium Di Hyacinthoides non-scripta A5 Impatiens glandulifera 64 Luzula sylvatica 45 Mercurialis perennis Dy Oxalis acetosella Diy, Ranunculus ficaria 100 Rubus spp. 45 Silene dioica 36 Urtica dioica a5) Only species recorded in 15% or more of sites are included; additional species recorded are listed below: Agrostis capillaris, Alliaria petiolata, Blechnum spicant, Cardamine raphanifolia, Cardamine sp., Conopodium majus, Doronicum pardalianches, Epilobium sp., Filipendula ulmaria, Geranium sp., Juncus effusus, Lysichiton americanus, Narcissus sp., Petasites hybridus, Phyllitis scolopendrium, Poa trivialis, Pteridium aquilinum, Ranunculus acris, R. repens, Ribes nigrum, Rosa canina, Sanicula europaea, Stellaria holostea, Taraxacum officinale, Ulmus sp., Valeriana dioica, Veronica beccabunga, V. chamaedrys, V. montana, Vicia sepium. Fraxinus excelsior with Alnus glutinosa confined to the edge of the watercourses (Table 1). The riverbanks also support large populations of J/mpatiens glandulifera which are present as seedlings during the flowering period of C. bulbifera. Other frequent asso- ciates are Aegopodium podagraria, Anemone nemorosa, Anthriscus sylvestris, Hyacinthoides non-scripta, Luzula_ sylvatica and Rubus fruticosus agg. C. bulbifera also occurs in Carex remota-Cirsium palustre sub- community of Alnus — glutinosa-Fraxinus excelsior-Lysimachia nemorum woodland (W7b) but in these stands many of the NVC constants are absent. Many other neophytes originating from Harlow Carr grow with C. bulbifera including Cardamine_ raphanifolia, Doronicum pardalianches, Lathraea_ cland- estina, Lysichiton americanus, Ranunculus aconitifolius, Tellima grandiflora and Symph- ytum grandiflorum. C. bulbifera is a vernal woodland-gap species which, like many shade-loving plants, comp- letes its reproductive cycle before ground flora vegetation has reached its full extent in July. In Harrogate leaves of C. bulbifera are clearly visible in February and by early March these have extended well above the surrounding leaves of Allium ursinum. Flowering is usually in full force by early April and completed by mid-June after when the leaves rapidly senesce to leave bare, withered stems in situ (these persist through the winter). In Harrogate a variety of insects have been observed visiting flowers, mostly notably Orange-tip (Antho- charis cardamines L.) and Green-veined White (Pieris napi L.) butterflies (the former mainly visiting C. amara). However, most flowers fail to set seed. Almost uniquely for a British woodland plant dispersal is therefore primarily by bulbils. These are produced very early in the leaf-axils and are dislodged by physical movement of the stems from late April onwards. Plants in Harrogate produce around twice as many bulbils as flowers: the average for 20 plants measured in 2008 was 12°8 (+ 0-7) bulbils (range 7-18) versus 6:4 (+ 0:5) flowers (range 2-10). Patches of say 500 stems, which are not uncommon on riverbanks, therefore produce anywhere between 3500 and 9000 bulbils annually. Individual bulbils vary in size but are relatively heavy weighing around 0-15 g. The majority therefore probably fall close to the parent plant. Secondary NOTES Watsonia 27 (2009) 253 dispersal probably occurs in a number of ways. Hegi (1958) reports dispersal by ants in continental Europe but Showler & Rich (1993), in noting that localities in Kent and Sussex are on, or close to river-banks and ditches, favour water-borne dispersal, presumably via ground- wash into streams and then physical movement by the current. Rose (1966) also suggested this as the main mechanism accounting for the micro-distribution of C. bulbifera in the Weald. Physical transport in the soil (as in this case of bulbils being introduced to Harlow Carr in peat) is likely to be less common (e.g. on boots, paws, hooves) but may account for the occurrence of isolated patches tens of metres above the flood zone. CONCLUSION The garden form of the Biodiversity Action Plan species C. bulbifera is now well naturalised along the River Nidd in v.c. 64 where it is undergoing a dramatic expansion in range, via water-borne dispersal of bulbils, into disturbed, sandy niches within the flood zone. Over the last decade tens of thousands of plants have become established and it has spread over 10 km downstream from its point of origin. It will be interesting to observe the behaviour of these populations, especially the rate and extent to which they expand along the River Nidd into the Vale of York over the next ten years. REFERENCES CHEFFINGS, C. (2004). New plant status lists for Great Britain. BSBI News 95: 36-43. HEGI, G. (1958). Ilustrierte Flora von Mittel-Europa 4(1). Miinchen. MANSON, D. C. (2004). Cardamine bulbifera in Harrogate. BSBI News 97: 22. NICHOLSON, R. A., SHOWLER, A. & RICH, T. C. G. (2004). Further observations on the ecology and distri- bution of Cardamine bulbifera (L.) Crantz (Brassicaceae) in the British Isles. Watsonia 25: 99-106. RICH, T. C. G. & JERMY, A. C. (1998). Plant Crib 1998. B.S.B.1., London. RODWELL, J. (Ed.) (1991). British Plant Communities. Volume 1. Woodlands and Scrub. Cambridge University Press, Cambridge. ROSE, F. (1966). Distribution maps of Kent plants. Proceedings of the Botanical Society of the British Isles 6: 279-281. SHOWLER, A. & RICH, T. C. G. (1993). Cardamine bulbifera (L.) Crantz (Cruciferae) in the British Isles. Watsonia 19: 231-245. WALLACE, I. (2005). Cardamine bulbifera (Coralroot) in Harrogate. BSBI News 98: 35. K. J. WALKER 97 Dragon Parade, Harrogate, North Yorkshire, HGI 5DG Email: Kevinwalker @bsbi.org.uk CIRSIUM x SEMIDECURRENS H. RICHTER IN GLAMORGAN The distinctive hybrid between Cirsium tuberosum (L.) All. and C. palustre (L.) Scop. (C. x semidecurrens H. Richter) has been re- corded from three sites in Britain growing amongst large stands of C. tuberosum in mosaics of scrub and calcareous grassland on moderate N- to NW-facing slopes with C. palustre growing nearby. This hybrid was first reported in Britain in 1932 by J. L. Bruce and J. S. L. Gilmour “...in the well-known locality for C. tuberosum near Nash Point” (BM) Glamorgan, although there is an earlier 1928 specimen in NMW collected by S. L. Bruce (Gilmour 1933). In 1947 Donald Grose discovered a single plant in South Wiltshire (v.c. 8) whilst searching for C. tuberosum close to where it was originally discovered by A. B. Lambert in 1812 although -the site was subsequently ploughed-up (Grose 1942, 1949, 1953, 1957). A second specimen from this site (Great Ridge), collected in 1939 and mis-labelled as Cirsium tuberosum, was re-determined as C. x semidecurrens (RNG; conf. W. A. Sledge). In 1986 a further plant was discovered nearby (Stockton Down, K, Everett 1993). The record for Montgomeryshire (v.c. 47) given in Stace et al. (2003) is an error and it has not been recorded in North Wiltshire (v.c. 7) despite both parents growing together in rank Bromopsis erectus grassland on Salisbury Plain. On the continent it has been recorded in France, Germany, Switzerland and Italy. 254 NOTES Watsonia 27 (2009) i 4 a ik ~\ if i ind fll aon, FIGURE |. Cirsium x semidecurrens growing at Cwm Nash, Glamorgan, August 2001. Cirsium xX semidecurrens was refound at Cwm Nash in Glamorgan (v.c. 41) by the author in August 2001 (Fig. 1) presumably growing very close to where it was originally discovered by Bruce over 70 years ago. A specimen was collected and verified by Clive Stace (NMW). This is the fifth British record (Appendix) and the first for Glamorgan since 1932 although botanists have noted putative hybrids in the Nash Point area in recent years (J. Woodman and Q. Kay, pers. comm.). Indeed, Kay & John (1995) cultivated a specimen of “C. tuberosum” from Cwm Nash which was morphologically unlike its seed parent with a tall branching flowering stem suggesting that it may have been an inter- specific hybrid, most likely with C. palustre, although other characters were more typical of C. tuberosum. In addition, achenes collected from a plant of C. tuberosum at Cwm Marcross, Glamorgan in 1994 also turned out to be this hybrid (Kay & John 1995). At Cwm Nash Cirsium x_ semidecurrens grows amongst an apparently pure colony of C. tuberosum on the moderately steep (30°) NW- facing slope of a small valley running down to the sea. The site overlies Jurassic limestone and is c. 2 km to the NW of the main Glamorgan colony of C. tuberosum at Nash Point Light- house. Cattle grazing has been re-introduced to the site recently (J. Woodman, pers. comm.) TABLE 1. FREQUENCY AND ABUNDANCE OF SPECIES GROWING WITH CI/RSIUM TUBEROSUM AND C. x SEMIDECURRENS AT CWM NASH, GLAMORGAN Species Frequency % Species Frequency % Agrostis capillaris 3) 3) Koeleria macrantha l + Anthoxanthum odoratum | + Leontodon hispidus 1 + Brachypodium sylvaticum 1 D Linum catharticum 1 + Briza media 1 4p Lolium perenne Z + Campanula glomerata 1 + Lotus corniculatus 3) 1 Carex flacca 3 1 Phleum pratense 1 + Centaurea nigra 3 7 Plantago lanceolata 3) 3) Cirsium arvense 2 + Potentilla erecta 1 + Cirsium tuberosum 3 VW Potentilla reptans 1 + Cirsium X semidecurrens 1 pe Primula veris 1 + Cynosurus cristatus 1 + Prunella vulgaris 3 ot Dactylis glomerata 3 D) Ranunculus bulbosus 1 + Daucus carota y) ct Sanguisorba minor 3 4 Festuca rubra 3 AQ Serratula tinctoria 1 1 Galium verum 3 + Stachys officinalis 3) Holcus lanatus 3 3 Succisa pratensis 2 4 Hypochaeris radicata 2 + Trifolium pratense 1 4 (Recorded in three 1 x Im quadrats). Species with average abundance < 1% are denoted by a “+”. NOTES Watsonia 27 (2009) and this has created a relatively short (c. 17 cm), species-rich sward (22 species m~-) dominated by Festuca rubra, with abundant Agrostis capillaris, Centaurea nigra, Cirsium tuberosum, Sanguisorba minor and Succisa pratensis and frequent Carex flacca, Galium verum, Lotus corniculatus and Stachys officinalis (Table 1). The grassland most closely resembles Galium verum sub-community of Cynosurus cristatus-Centaurea nigra mesotrohic grassland (55% fit to MG5b) but with affinities to Festuca rubra-Holcus lanatus maritime grassland (MC9). Over 60 patches of Cirsium tuberosum are scattered across the slope as well as a single patch of the hybrid between C. tuberosum and C. acaule (C. x NW Nn N medium) which has been known from this locality since 1935 (Bruce & Vachell 1935). Cirsium palustre does not occur on the same slope but is present by the stream at the bottom of the valley. The stems of the hybrid were tall (to 60 cm) and branched, with a mixture of arachnoid and jointed hairs especially towards the apex, and shortly decurrent, spinous wings extending for a few centimetres below the leaf bases (not present on all stems). The stems were topped with very lax racemose clusters of up to four capitula (as in Wiltshire) on relatively short (1— 10 cm) densely felted peduncles. The capitula were intermediate in size and shape, being narrower and shorter (mean 11 x 12 mm) than Op ee, Cirsium palustre O Cirsium tuberosum England ° @ Cirsium tuberosum Wales ® 4 © Cirsium x medium England s o8 @ Cirsium x medium Wales o @ Cirsium x semidecurrens, Ze A Cirsium acaule O lex N TX 3 A (ae A D Ik IX A A Ame A (| -8 -8 -6 -4 -2 0 2 4 PCA 1 FIGURE 2. A Principal Components Analysis (PCA) showing the similarity between Cirsium tuberosum and British hybrids (C. x semidecurrens and C x medium) and parental species (C. acaule, C. tuberosum, C. palustre) based on morphological characters. English and Welsh plants of C. tuberosum and C. x medium are shown with different symbols. 256 NOTES Watsonia 27 (2009) those of C. tuberosum (mean 13 x 14 mm), and with more-or-less adpressed, sparsely arachnoid outer phyllaries with the purple callosity indicative of C. palustre. The plant therefore differed from C. tuberosum in having almost twice as many capitula on each branch, shortly-decurrent, spinious wings and a mixture of arachnoid and jointed hairs on the stem, smaller, narrower involucres (Fig. 1), and + adpressed, sparsely arachnoid outer phyllaries. A Principal Components Analysis of a range of morphological characters measured on British hybrids and the parental species (C. acaule, C. palustre and C. tuberosum) 1s shown in Figure 2. These data show a clear separation between C. acaule, C. x medium and C. tuberosum on the first axis and between these species and C. palustre and C. x semidecurrens on the second (Rice): Although originally thought to be sterile (Gilmour 1933, Grose 1949), a small sample of hybrid fruits collected from the Glamorgan plant showed 50% germination (2 of 4 seeds). These were presumably backcrosses with C. tuberosum and in cultivation these hybrids continued to flower and produce a small number of viable achenes for over four years. C. palustre appears to hybridise readily with other thistles and is a parent to half of the natural thistle hybrids including the two most frequently recorded in the British Isles, C. x forsteri and C. x celakovskianum. The reason for this is unknown but may be due to its wider ecological range and the nectar being more readily accessible, and therefore more regularly visited by insects, than in other species of Cirsium. C. X semidecurrens is one of the rarest, presumably because of the rarity of C. tuberosum, but also because C. palustre is confined to moister mesotrophic and/or shaded habitats (e.g. colluvium, scrub edge) on calcareous soils where C. tuberosum 1s abundant. Cross-pollination is therefore likely to be limited in such habitats. Successive backcrossing and introgression with one or more parents has been reported for a few Cirsium hybrids, most notably between C. x medium and C. tuberosum on heavily grazed downland in Wiltshire where pure C. tuber- osum iS now very uncommon. This is also likely to occur at Cwm Nash where presumably all the seed produced by the hybrid will be back-crossed with C. tuberosum as the male parent. The Cwm Nash plant appears to be long-lived and, like C. tuberosum, may be spreading indefinitely via its large underground tubers. It will be interesting to see how long the plant survives and the extent to which backcrossing occurs in the future. | ACKNOWLEDGMENTS I am grateful to Clive Stace for determining the specimen, Tim Rich and Stephen Jury for the loan of specimens and Julian Woodman and Gill Barter (Countryside Council for Wales) for providing information on the Glamorgan populations. REFERENCES BRUCE, G. L. & VACHELL, E. (1936). 396/4 x 6. Cirsium acaule (L.) Weber x tuberosum = fraserianum Dr. Report of the Botanical Society and Exchange Club for 1935 11(1): 32. EVERETT, S. (1988). Rare Vascular Plants Survey for S. England. Nature Conservancy Council, Peterborough. EVERETT, S. (1993). Cirsium tuberosum (L.) All. Tuberous Thistle in The Wiltshire Flora, B. GILLAM (ed.). Pp.83-90. Pisces Publications, Newbury. GILMOUR, J. S. L. (1933). Cirsium tuberosum All. x C. palustre Scop. Journal of Botany 71: 17. GROSE, J. D. (1949). A hybrid thistle from Wiltshire. Watsonia 1: 91. GROSE, J. D. (1942). The tuberous thistle in Wiltshire. Wiltshire Archaeological and Natural History Magazine 49: 557-561. KAY, Q. O. N. & JOHN, R. (1995). The conservation of scarce and declining species in lowland Wales: population genetics, demographic ecology and recommendations for future conservation of 32 species of lowland grassland and related habitats. Countryside Council for Wales, Contract Science Report 110, Bangor. STACE, C. A. (1975). Hybridisation and the flora of the British Isles. Academic Press, London. APPENDIX Confirmed records of Cirsium x semidecurrens in the British Isles: SS96, GLAMORGAN, V.C. 41: With both parents at St. Donat’s Cliff [Cwm Nash], near Nash Point, J. L. Bruce & J. S. L. Gilmour, August 1932 (Gilmour 1933), BM. NOTES Watsonia 27 (2009) Di, Cwm Nash, near Nash Point, single plant growing amongst colony of Cirsium tuberosum, SS905699, K. J. Walker, 7 August 2001, NMW. ST93, SOUTH WILTSHIRE, V.C. 8: Great Ridge, J. D. Grose, 7 August, 1939, RNG [specimen labelled C. tuberosum but later re- determined as C. x semidecurrens by W. A. Sledge, det. 10 August, 1951]. Great Ridge, growing with C. tuberosum on a N-facing dowland in A. B. Lambert’s original 1812 station, J. D. Grose, July 1947 (Grose 1942), det. W. A. Sledge. Stockton Down near Great Ridge, single plant growing with C. tuberosum on edge of scrub, ST975358, S. Everett, 30 July 1986 (Everett 1993), K [the specimen could not be traced but there is a photocopy of the specimen in Everett (1988)]. K. J. WALKER 97 Dragon Parade, Harrogate, North Yorkshire HGI 5DG Email: kevinwalker@bsbi.org.uk VASCULAR PLANTS FROM SWANSEA UNIVERSITY HERBARIUM (UCSA) NOW INCORPORATED INTO THE WELSH NATIONAL HERBARIUM (NMW) In 2006, the vascular plants component of the Swansea University herbarium (UCSA) were donated to the National Museum of Wales (NMW). The bryophyte collection has been retained at Swansea. The collection consisted of about 5000 vascular plant specimens, but after curation consisted of 4256 specimens as material which had been seriously damaged by insects or was inadequately labelled was discarded. The material has all been remounted on conservation grade mounting card, and documented on the museum’s collection management system. Swansea University was founded in 1920, and the herbarium probably dates from about that time with 586 specimens collected from 1920 to 1949. Most of the material was collected after 1950 (3290 specimens) coincident with expansion of the school of Biological Sciences. Some historic material was also acquired, an interesting undated and anonymous collection from about the 1830s centred around Monmouthshire. As might be expected from a university herbarium compiled by staff and students, about half of the material was collected in South Wales. The following vice-counties were represented: 1-31, 33-46, 48-50, 52-59, 61- 66, 69-70, 72-73, 78, 83-85, 88-94, 96-97, 100, 103-105, 107-108, 110-112, HI—H3, H9, H15-H16, H24, H27-H28, H38 and the Channel Islands. Material had also been collect- ed on university field trips or whilst attending conferences in Greece (423 specimens), Spain (395), Iceland (212), Norway (114), India (88), France (46), Sweden (41), Austria (35) and Kenya (23), with small amounts of material from Australia, Belgium, Bulgaria, Denmark, Germany, Greenland, Italy, Malawi, Nether- lands, New Zealand, Portugal, Switzerland, Tanzania, West Indies and Zaire. There was a broad taxonomic coverage, but notable was the relatively large Poaceae collection (1100 specimens compared with the three top other families Asteraceae (329), Cyperaceae (306) and Fabaceae (274)) which included a collection which had formerly belonged to C. Bucknall. The only significant research collections were of Q. O. N. Kay (mainly Asteraceae, unfortunately badly damaged) and some Melampyrum material from A. J. E. Smith. The main collectors are Q. O. N. Kay (1224 specimens collected between 1959 and 1990), A. J. E. Smith (982 specimens 1956-1963), J. Hayward (226 specimens, 1950-1962), C. R. Hipkin (176 specimens, 1962-1984, mostly collected with H. Hipkin), R. Webb (128 specimens, 1954-1970) and M. L. Page (108 specimens, 1975). The data from the collection will be made available on the National Museum of Wales website, or can be requested from T. Rich. ACKNOWLEDGMENTS We would like to thank Rosalind Codd, Chris Heron, John Owen and David Price for helping to curate the collection. T. HUMPHREY, C. TAYLOR & T. C. G. RICH Welsh National Herbarium, Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP. C. R. HIPKIN Biological Sciences, Swansea University, Singleton Park, Swansea SA2 SPP. Watsonia 27: 259-272 (2009) 259 Plant Records Records for publication must be submitted to the appropriate Vice-county Recorder (see BSB/ Year Book 2008), and not to the Editors. Following publication of the New Atlas of the British & Trish Flora and the Vice-county Census Catalogue, new criteria have been drawn up for the inclusion of records in Plant Records. (See BSBI News no. 95, January 2004 pp10 & 11). These are outlined below: e First records of all taxa (species, subspecies and hybrids) included in the VCCC, designated as native, archaeophyte, neophyte or casual. e First record since 1970 of the taxa above, except in the case of Rubus, Hieracium and Taraxacum. e Records demonstrating the rediscovery of all taxa published as extinct in the VCCC or subsequently. e Newly reported definite extinctions. e Deletions from the VCCC (e.g. through the discovery of errors, the redetermination of specimens etc.) NB — only those errors affecting VCCC entry. e New 10km square records for Rare and Scarce plants, defined as those species in the New Atlas mapped in the British Isles in 100 10km squares or fewer. (See BSBI News no. 95, January 2004 pp 36-43). Records for the subdivisions of vice-counties will not be treated separately; they must therefore be records for the vice-county as a whole. However, records will be accepted for the major islands in v.cc. 100, 102-104, 110 and 113. In the following list, records are arranged in the order given in the List of Vascular Plants of the British Isles and its supplements by D.H. Kent (1992), from which the species’ numbers, taxonomy and nomenclature are taken. The Ordnance Survey national grid reference follows the habitat and locality. With the exception of collectors’ initials, herbarium abbreviations are those used in British and Irish Herbaria by D. H. Kent & D. E. Allen (1984). Records are field records if no other source is stated. For all records ‘det.’ or ‘conf.’ appear after the herbarium if the determination was based on material already in an institutional herbarium, otherwise before the herbarium. The following signs are used: * before the vice-county number: to indicate a new vice-county record. + before the species number: to indicate that the plant is an archaeophyte. + before the species number: to indicate that the plant is a neophyte. © before the species number: to indicate that the plant is a casual. The above 3 signs may also used before the vice-county number to indicate the status of the plant in that vice-county. ®) before the vice-county number: to indicate that this is an additional hectad for a Rare or Scarce plant. ® at end of entry: established taxon not in Vice-County Census Catalogue. Name of authority provided. [ ] enclosing a previously published record: to indicate that that record should be deleted or changed. From now on records will be published in two separate sections — 1) NATIVES (including archaeophytes) and 2) ALIENS (neophytes and casuals). The following list contains the second set of records up to and including the year 2007. Records up to and including the year 2008 will be published in the next issue of Watsonia. 260 PLANT RECORDS ARCHAEOPHYTES AND NATIVES Osmunda regalis 6/1.1. | *31, Hunts.: wet ditch, Holme Fen NNR, TL1989, D. A. Broughton, 2007, conf. T. C. E. Wells. Origin unknown but not planted; colonisation from native source no less likely than non-native source. Gymnocarpium dryopteris 16/4.1. *17, Surrey: TQ05, R. Skipper, 2007. Polystichum setiferum 17/1.1. | 31, Hunts.: ancient woodland, Waresley Wood, TL263546, T. Balbi, 2000. Ist record since 1846. Polystichum xbicknellii (P. setiferum x aculeatum) 17/1.1x2. *64, Mid-W. Yorks:.: limestone pavement, Malham Cove, SD897641, B. Brown, 2007. Dryopteris xcomplexa (D. filix-mas x affinis) 17/3.2x3. | *64, Mid-W. Yorks.: woodland on hillside, Otley Chevin, SE213444, B. Brown, 2007, det. K. Trewren. Helleborus foetidus 28/3.1. *© 45, Pembs.: frequent, in full flower, on roadside bank, Uzmaston Road, Haverfordwest, SM956155, S. B. & A. E. Evans, 2007. A location where garden waste 1s dumped. *+Fumaria parviflora 31/5.9. | *44, Carms.: waste ground opposite end of New Dock Road, Llanelli, SS510988, G. Hutchinson & I. K. Morgan, 1992, det. T. C. G. Rich, NMW. +Chenopodium ficifolium 43/1.13. *50, Denbs.: field edge, Sutton Green, SJ415488, M. Stead, 2007, conf. G. Kay. Atriplex portulacoides 43/3.10. *t64, Mid-W. Yorks.: road verge, Stonehouse, SE1558, Wharfedale Nats. Soc. 2007, det. N. Vernon. Spergularia marina 46/17.3. *t31, Hunts.: roadside, Great Stukeley, TL219746, K. J. Walker, 2004. +Silene gallica 46/20.11. 70, Cumberland: 30 stems in disused sandpit, New Cowper, NY117458, D. Muscat, 2007, LANC. Ist record since early 19th C. | Rumex xabortivus (R. conglomeratus x obtusifolius) 47/8.14x19. *44, Carms.: by cycle track (northern verge), Penclacwydd Wildfowl & Wetlands Centre, E. of Llanelli, SS533980, M. O. Stead & BSBI meeting, 2006, det. G. Kitchener, NUW. Hypericum linariifolium 51/1.11. 43, Rads.: Dolyhir near Stanner, SO246.583, A. Shaw, 2006. Ist record since 1945. Lavatera arborea 53/2.1. | *£62, N. E. Yorks.: several self-sown plants in rough grassland, Skelton, NZ658193, V. Jones & M. J. Yates, 2007. [Helianthemum canum subsp. levigatum 56/2.3c. 64, Mid-W. Yorks. & 69, Westmorland. Delete. The VCCC records are errors for subsp. incanum.] Frankenia laevis 59/1.1. — *£46, Cards.: mat 50 x 25 cm in shingly waste ground, presumably self-sown from flowerbed 7 m away where it grows, S end of Borth, SN607889, A. O. Chater & J. P. Poland, 2007. +Salix xrubens (S. fragilis x alba) 61/2.2x3. *78, Peebless.: one large tree onethe tyre boundary, Well Burn, Eastfield, near Biggar, NT035356, D. J. McCosh, 1999. Salix xreichardtii (S. caprea x S. cinerea) 61/2.10x11. 44, Carms.: willow-dominated scrub, Huntsman Chemical Works, Bynea, Llanelli, SS556986, Llanelli Naturalists meeting, 2007, det. G. Hutchinson. Ist recent record. y*Erysimum cheiranthoides 62/7.1. 51, Flints.: many plants in weedy field, Tyn-y-ffordd, Bodfari, SJ087701, J. A. Green, 2007. Ist recent record. Rorippa islandica 62/12.3. *70, Cumberland: grazed, muddy hollow, Gilwilly Industrial Estate Penrith, NY507307, R. W. M. Corner, 2007, conf. T. C. G. Rich, LANC. *+Camelina sativa 62/24.1. *69, Westmorland: abundant in field by A684, Killington, SD599918, A. M. Boucher, 2007, LANC. Relict of ‘conservation mixture’. Pyrola rotundifolia subsp. maritima 66/1.3b. | *49, Caerns.: disused quarry floor, Trefriw, SH7861, I. Fraser, 2007. Potentilla tabernaemontani 75/9.12. ®64, Mid-W. Yorks.: grassland, Whitewell, SD653465, P. C. G. Green, 2007. Rosa xsabinii (R. spinosissima x mollis) 75/21.5x17. *78, Peebless.: Innerleithen, NT33, A. Craig Christie, 1873, E, det A.L. Primavesi. Amongst large collection at E determined by A. L. Primavesi. PLANT RECORDS 261 Rosa xdumalis (R. canina x caesia subsp. vosagiaca), 75/21.12x13b. *105, W. Ross: roadside scrub, Camusterrach, NG7141, R. Maskew, 2005. Hybrid with R. caesia as female parent. [Rosa xmolletorum (R. canina x mollis) 75/21.12x17. 37, Worcs.: delete from VCCC. Specimen corresponding to single 1999 record now considered a complex indeterminate hybrid (R. Maskew 2006). | [Rosa caesia subsp. caesia, 75/21.13a. 37, Worcs.: delete from VCCC. Specimen corresponding to single 1991 record re-determined as R. canina c.v. (R. Maskew 2004). | Rosa caesia subsp. caesia, 75/21.13a. *46, Cards.: about ten bushes in roadside hedges, 1.5 km E of Oakford, SN467578, R. Maskew & A. O. Chater, 2007, det. R. Maskew, NMW. Rosa caesia subsp. vosagiaca 75/21.13b. 105, W. Ross: roadside scrub, Camusterrach, NG7141, R. Maskew, 2005. Last confirmed v.c. record 1938. Rosa xrogersii (R. caesia subsp. vosagiaca x tomentosa) 75/21.13bx15. *37, Worcs.: scrub on edge of bridleway, Abbots Morton, SP036555, R. Maskew, 2001, herb. R. Maskew, det. R. Maskew & A. L. Primavesi as hybrid with R. tomentosa as female parent. O Rosa caesia subsp. vosagiaca x sherardii 75/21.13bx16. 105, W. Ross: roadside scrub, Camusterrach, NG7141, R. Maskew, 2005, herb. R. Maskew, hybrid with R. caesia as female parent. Last confirmed v.c. record 1938. @ Rosa xobovata (R. caesia subsp. vosagiaca x rubiginosa) 75/21.13bx18. *108, W. Sutherland: single bush on roadside, Kyle of Tongue, NC5554, R. Maskew, 2007, herb. R. Maskew, hybrid with R. rubiginosa as female parent. @ Rosa xavrayensis (R. tomentosa x rubiginosa) 75/21.15x18. | *37, Worcs.: two bushes with both parents, in scrub, Penny Hill Quarry, Martley, SO752614, R. Maskew, 2006, herb. R. Maskew, hybrid with R. tomentosa as female parent. Rosa mollis 75/21.17. 36, Herefs.: several bushes on rocky hillside, Black Darren, SO0295295, M. Jannink, 2007, conf. R. Maskew in situ, herb. R. Maskew. Only previous confirmed v.c. record 1907. Rosa xbigeneris (R. rubiginosa x micrantha) 75/21.18x19. | *37, Worcs.: few bushes with both parents, in scrub, Penny Hill Quarry, Martley, SO752615, R. Maskew, 2006, herb. R. Maskew, hybrid with R. micrantha as female parent. +Pyrus pyraster 75/26.2. 44, Carms.: one tree near old cottage, bank of Nantrhydw, Pontantwn, SN434139, J. A. Green, 2007, det. A. O. Chater & G. M. Kay, NMW. Vicia lutea 77/14.13. *t44, Carms.: six robust plants in compacted development plateau, presumably introduced with seed mix, Parc Hendre development site, Capel Hendre, SN599109, B. Stewart, 2007, NMW. Vicia bithynica 77/14.14. +17, Surrey: near to remains of gravel pit lake, Molesey Heath, West Molesey, TQ132673, R. Alder, R. D. Baker, M. Parslow, B. M. Spooner & Whitfield, 2007. A good colony with Rumex palustris and Euphorbia xpseudovirgata. |st record since 1970. Epilobium xsubhirsutum (E. hirsutum x parviflorum) 84/1.1x72. *38, Warks.: with both parents in field near disused railway, Burton Green, near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. With 7 other Epilobium hybrids! Epilobium xbrevipilum (E. hirsutum x tetragonum) 84/1.1x<5.’ *38, Warks.: with both parents in set-aside field, Burton Green, near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. Epilobium xwaterfallii (EK. hirsutum x palustre) 84/1.1x9. *44, Carms.: rank grassland, with both parents, West Tip, Cross Hands, SN561134, R. D. & K. A. Pryce, 2007, det. G. Kitchener, NMW. Ist confirmed record. Epilobium xpalatinum (EF. parviflorum x tetragonum) 84/1.2x5. — *38, Warks.: with both parents in set-aside field, Burton Green near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. Epilobium xdacicum (E. parviflorum x obscurum) 84/1.2x6. 38, Warks.: with both parents in field near railway, Burton Green, near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. Ist record since 1874. Epilobium xinterjectum (E. montanum x ciliatum) 84/1.3x8. *38, Warks.: about 10 plants, Stables farmyard, Snitterfields, SP252597, J. W. Partridge, 2004, det. G. D. Kitchener, WARMS. Epilobium xlamotteanum (E. lanceolatum x obscurum) 84/1.4x6. *44, Carms.: lane bank off bend in driveway up to Forge Farm, Forge Quarry, Cwmdwyfran, SN408257, G. Hutchinson, 1997, det. G. Kitchener, NMUW. 262 PLANT RECORDS Epilobium xsemiobscurum (E. tetragonum x obscurum) 84/1.5x6. | *38, Warks.: with both parents in set-aside field, Burton Green, near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. Epilobium xmentiens (E. tetragonum x ciliatum) 84/1.5x8. 38, Warks.: with both parents in set-aside field, Burton Green, near Coventry, SP2775, G. D. Kitchener, 2007, WARMS. Ist record since 1965. Euphorbia serrulata 91/2.7. *©44, Carms.: garden weed, abundantly self-sowing, Garregwen Farm, Penrhiwgoch, SN561185, G. Hutchinson, R. D. & K. A. Pryce, 2005, det. T. Walker, NMW. Also seen, in 2007, by M. Stead & BSBI group in river shingle at Llanwrda (SN716307). Geranium rotundifolium 103/1.4. *31, Hunts.: improved grassy verge by cyclepath, Woodston, TL1797, D. A. Broughton, 2007. Pimpinella major 107/13.1. 70, Cumberland: roadside verge, Storms, east of Keswick, NY290243, R. E. Groom, 2007, LANC. Ist record since 1939. Ligusticum scoticum = 107/36.1. *£70, Cumberland: one non-flowering plant in dunes, Blitterlees, Silloth, NY099519, R. E. Groom, 2007, LANC. *+Vinca minor 109/1.1. 48, Merioneth: in quantity near houses on a roadside, near Mallwyd, SH81, P. M. Benoit, 2007. Ist record since 1970. +Hyoscyamus niger 110/4.1. ®52, Anglesey: disturbed ground, Penial Dowyn, SH293832, J. Rees, 2007. New hectad and Ist record in the v.c. since 1979. Calystegia sepium subsp. roseata, 111/3.2b. *31, Hunts.: locally abundant on edge of birch woodland in fen, Holme Fen NNR, TL1989, D. A. Broughton, 2007. +Lamium hybridum 118/5.4. *43, Rads.: near Howey, SO067586, B. Laney, 2007. Nepeta cataria 118/13.1. +49, Caerns.: many plants on disturbed soil mound, Llandygai industrial estate, SHS5971, W. McCarthy, 2007. Ist record for at least 40 years. Clinopodium acinos 18/18.5. 50, Denbs.: rocky path, Eglwysleg, Llangollen, SJ223468, J. L. Roberts, 2007, conf. J. A. Green, NMW. Ist record since 1970. Mentha xgracilis (M. arvensis x spicata) 118/23.1x3. **31, Hunts.: woodland ride, Folksworth, North Wood, TL133897, K. J. Walker, 2004. Mentha suaveolens | 18/23.4. +50, Denbs.: roadside, Coedpoeth, $J302512, J. L. Roberts, 2007. Ist record since 1970. Callitriche brutia 120/1.6. | *51, Flints.: mesotrophic pond, Maes-y-grug between Alltami and Northophall, $J262665, S. Whild, 1994. Ist record (although in VCCC). Kuphrasia nemorosa x confusa 124/20.7x9. *17, Surrey: two colonies in firebreak, Chobham Common, SU968652, A. Wragg, 2007. Voucher specimen retained. *48, Merioneth: in long grass by road junction, Trystion valley above Cynwyd, SJ04, S. E. Stille, 2007, det. A. J. Silverside. Pedicularis palustris 124/25.1. 31, Hunts.: mixed fen community, Woodwalton Fen, TL229848, N. Millar, 2005. First record since 1913. *+Sambucus ebulus 131/1.4. 78, Peebless.: North Slipperfield, NT1251, J. H. Penson, 1970. Source - J. H. Penson’s notebooks. Ist record since 1970. Site searched in 2007; not found. 7Valerianella carinata 133/1.2. *31, Hunts.: ditch bank by road, Conington, TL194869, W.R. Meek, 2005. Centaurea xmoncktonii (C. nigra x jacea) 135/11.7xjac. *35, Mons.: 2 plants near the bottom of a barish grassy bank on the edge of A466, near Wyndcliff, ST523970, T. G. Evans, 2007. Ist record since 1970 & Ist Welsh record as a spontaneous hybrid. This rayed knapweed has colourless combs of different shapes on its outer phyllaries ranging from those on C. jacea to some approaching those of C. nigra. Picris hieracioides 135/17.2._ 69, Westmorland: one plant on verge of M6, Dillicar, Lowgill, SD622982, M. S. Porter, 2007. Ist record since 1970. Taraxacum naevosiforme 135/25.5.41. *46, Cards.: road verge, SW of Maesnant, Pumlumon, SN773878, S. P. Chambers, 2007, det. A. J. Richards, NMW. Hieracium consociatum Jord. ex Boreau 135/28.7.con. *44, Carms.: on the cliff E of the spillway and the parking area, Llyn Brianne, SN793484, J. Sawtschuk, 2006, det. D. J. McCosh, NMW. = *47, Monts.: a few plants on old bridge, Abermule Dingle, SO167935, A. K. Thorne, 2007, det. D. J. McCosh. @ Hieracium cinderella 135/28.8.135. 47, Monts.: a few plants on rock-face in woodland, Dolforwyn Wood, SO156958, A. K. Thorne, 2007, det. D. J. McCosh. Ist record since 1970 and new Site. ; PLANT RECORDS 263 Senecio xostenfeldii (S. jacobaea x aquaticus) 135/62.10x11. *78, Peebless.: Ada Hill, Eddleston, NT234488, L. Gaskell, 2006. Senecio erucifolius 135/62.12. *69, Westmorland: north side of canal, opposite garage, Crooklands, $D539828, J. Clarke, 2007, LANC. Ist definite record. Alisma lanceolatum 137/4.2. 51, Flints.: mesotrophic pond, Maes-y-grug between Alltami and Northophall, SJ262665, S. Whild, 1994. Ist record since 1930. Potamogeton xsparganiifolius (P. natans x gramineus) 142/1.1x6. *46, Cards.: abundant on silty shoals among Myriophyllum alterniflorum, in Afon Teifi just above Pont Einon, Cors Caron, SN671614, R. V. Lansdown, 2005, det. C. D. Preston, NUW, CGE. The P. gramineus parent is not known in the river or in the v.c. Juncus xdiffusus (J. inflexus x effusus) 151/1.25x26. 49, Caerns.: disused quarry floor, Trefriw, SH7862, W. McCarthy, 2007, conf. P. Benoit. 1st record since 1970. Juncus xkern-reichgeltii (J. effusus x conglomeratus) 151/1.26x27. *44, Carms.: frequent in unimproved acid grassland, Hafod Wenol SSSI, Mynydd Betws, SN676095, BSBI meeting, 2007, conf. A. O. Chater & J. P. Poland. Ist published record. Trichophorum cespitosum nothossp. foersteri (T. cespitosum subsp. cespitosum x cespitosum subsp. germanicum) 152/2.2ax2b. *46, Cards.: hollows in raised bog, Gors Goch, 800 m NW of Llyn Fanod, SN597649, A. O. Chater & S. D. S. Bosanquet, 2007. Schoenoplectus lacustris 152/7.1. 35, Mons.: many patches on south margin of Tesco attenuation pond, between Magor and Llanwern, ST404864, S. J. Tyler, 2007, conf. T. G. Evans. lst record since 1970 and a new hectad for this very uncommon v.c. plant. Carex paniculata 152/16.1. 48, Merioneth: four plants on wet wooded slope, Dolgoch ravine, SH60, P. M. Benoit, 2007. 1st recent record: a rediscovery of the locality and updating for the hectad from K. M. Stevens, 1969. Carex diandra 152/16.3. 43, Rads.: Rhosgoch Common, SO194481, S. D. S. Bosanquet, 2007. Also seen in 2007, by A. Ferguson, at SO200486 on Rhosgoch Common. Ist recent records. Previously reported from Llandrindod Wells (1919) and Rhosgoch Common (1948). Carex xsubgracilis (C. acutiformis x acuta) 152/16.25x65. —_*17, Surrey: large colony with both parents present, Chertsey Meads, Area 3b, TQ063662, G. Hounsome & B. W. Phillips, 2007. Known for some years but only just determined. Carex viridula subsp. brachyrrhyncha 152/16.46a. 47, Monts.: upland runnel, Bryncoch Farm, SJO13081, D. Reed, 2007, conf. A. Jones. 4th site but 1st confirmed record. Festuca ovina subsp. hirtula 153/12.8b. *62, N. E. Yorks.: several plants in heathy grassland, Biller Howe Dale, NZ912017, V. Jones & M. J. Yates, 2007. Lolium xboucheanum (L. perenne x multiflorum) 153/13.1x2. *31, Hunts.: margin of arable adjacent to improved grassland, Sawtry, TL168826, D. A. Broughton, 2008. *62, N. E. Yorks.: set-aside, by River Rye near Helmsley, SE626827, V. Jones & W. A. Thompson, 2007, conf. A. Copping, herb. V. Jones. Poa infirma 153/18.1. | *£31, Hunts.: road verge, Conington, TL189863, W. R. Meek, 2005, det. A. C. Leslie. Agrostis xmurbeckii (A. capillaris x stolonifera) 153/39.1x4. | *44, Carms.: sunny places on poached ground by gate at N side of enclosure, Caeau Blaenau Mawr SSSI, Capel Hendre, SN594122, A. O. Chater, 2006, NMW. xAgropogon littoralis (Agrostis stolonifera x Polypogon monspeliensis) 153/39x46/4x1. *©63, S. W. Yorks.: waste ground, old mill site, Bradford, SE133333, B. A. Tregale & M. Wilcox, 2007. y+Apera spica-venti 153/44.1. | *©48, Merioneth: two plants in dirt at base of fence separating the railway track from the footway, Barmouth Viaduct, SH61, P. M. Benoit, 2007. Alopecurus xbrachystylus (A. pratensis x geniculatus) 153/47.1x2. *63, S. W. Yorks.: damp grassland edge near Maythorn, SE1805, P. Middleton, 2007. Phleum arenarium 153/49.5. *©63, S. W. Yorks.: sandy acidic terrain, Lower Whitley Edge, SE1804, M. J. Lucas, 2003. Brachypodium pinnatum 153/54.1. *42, Brecs.: patch about 2 sq.m. in disused limestone quarry, Penwyllt, SN857156, M. & C. Porter, 2007, det. M. Porter. herb. M.P. 264 PLANT RECORDS Elytrigia xacuta subsp. obtusiuscula (E. atherica x juncea) 153/56.2x3a. *44, Carms.: Dune slope, W of Ginst Point, Laugharne, SN322077, A. O. Chater & S. D. S. Bosanquet, 2006, NMW. Pollen empty, somewhat shrunken. Hordelymus europaeus 153/58.1. ®69, Westmorland: several tufts on semi-shaded limestone slope in woodland, Witherslack, SD428870, I. Taylor, 1994, LANC (2007). Allium schoenoprasum 158/24.1. *t64, Mid-W. Yorks.: road verge, Greenhow Hill Road, SE1558, Wharfedale Nats. Soc. 2007, det. N. Vernon. xDactylodenia st-quintinii (Gymnadenia conopsea x Dactylorhiza fuchsii) 162/16x18.1x1. *78, Peebless.: Glenbield Farm, Peebles, NT24, W. Bell, 1998, det. R. Bateman. Dactylorhiza fuchsii x D. incarnata subsp. coccinea 162/18.1x3b. #63, S. Wo orkse Woodhall Quarry near Calverley, SE197353, M. Wilcox, 1997. NEOPHYTES AND CASUALS tSelaginella kraussiana 2/1.2. *35, Mons.: 10 x 2 m on barish patches in grass, below N wall of church, Llangybi, ST373967, S. D. S. Bosanquet, 2006. tPinus nigra subsp. nigra 20/7.2a. 44, Carms.: disturbed, quite dry dune slack, Tywyn Burrows, SN363055, BSBI meeting, 2007, det. A. O. Chater. Also seen in 2007, by M. & J. Iliff, in Ystrad Ffin Churchyard (SN787471). Ist records as neophyte. +Helleborus orientalis 28/3.3. *42, Brecs.: hedge near Llangynidr, SO165198, M. & C. Porter, 2007. First seen here in 2000. tNigella damascena 28/5.1. 46, Cards.: pathside bank disturbed in c.2005, Cefn-llan, Llanbadarn Fawr, SN599814, S. P. Chambers, 2007. Ist record as neophyte. Anemone blanda 28/9.bla. *46, Cards.: well naturalised on shaded stream bank, Llanrhystud, SN541698, A. O. Chater, 2007, NMW. ; +Ranunculus ficaria subsp. ficariiformis 28/13.17c. | *31, Hunts.: shady footpath in village, Alwalton, TL134962, R. Woodall, 2002, conf. T. C. E. Wells, Specimen in ABRN. tBerberis thunbergii 29/1.2. 46, Cards.: self-sown bush in scrub on pasture slope, north bank of Cwm Rheidol Reservoir, SN698793, A. O. Chater, 2007. Ist record as neophyte. *62, N. E. Yorks.: bird-sown in rough wasteland, N of A66, South Bank, NZ532209, V. Jones & W. A. Thompson, 2007. tBerberis wilsoniae 29/1.4. 44, Carms.: two self-sown bushes plus several seedlings and small plants in vicinity on limestone-aggregate surfaced yard of derelict Ennis Caravans site, Cross Hands, SN559128, R. D. & K. A. Pryce & G. Hutchinson, 2007. tBerberis wilsoniae x aggregata 29/1.4x5. | *70, Cumberland: bank of River Ehen, exit from Ennerdale Lake, NY088152, R. E. Groom, 2007, LANC. Ist noted 2003. @ tBerberis julianae 29/1.7._ *44, Carms.: waste ground at front of disused caravan-park, Cross Hands, SN559128, G. Hutchinson, 2007, NMW. Ist Welsh record. tBerberis darwinii 29/1.9. | *52, Anglesey: well established around rocky outcrop, Trearddur, SH265796, I. Bonner & C. Aron, 2007. *62, N. E. Yorks.: several bird-sown bushes on bare ground by railway N of Teesside Park, NZ476194, V. Jones & W. A. Thompson, 2007. *+Morus nigra 35/1.1. *44, Carms.: hedgerow near to exit of field, near house, Glan Rhydw, Pontantwn, SN433134, G. M. Kay ef al. 2007. Ist v.c. record outside garden. +Nothofagus alpina 39/2.2._ *69, Westmorland: two trees by path east of Birkbeck Cottage, Coniston, SD317965, G. Lines, 2007, LANC. *Betula ermanii Chamisso 40/1.erm. *51, Flints.: Nature Reserve on old industrial site, Rhydymwyn, SJ205668, J. Phillips, 2006, det. G. Hutchinson, NMW. @ Ist Welsh record. *Alnus rubra 40/2.rub. *62, N. E. Yorks.: several, naturally regenerating, by side of forestry track, Battersby Plantation near Ingleby Greeenhow, NZ598066, V. Jones, 2007, herb V. Jones. ©Chenopodium urbicum 43/.1.11. *77, Lanarks.: rough ground at King George V Dock, Glasgow, NS5366, P. Macpherson, 2007, det. E. J. Clement. +Atriplex halimus 43/3.9. *7(), Cumberland: one bush in shelter-belt by railway sidings, Maryport, NY030361, R. E. Groom, 2007, LANC. ©Corispermum leptopterum (Asch.) 43/COR.lep. *51, Flints.: scattered plants in light sandy soil on industrial site, Paper Mill, Deeside, SJ307719, L. Charter, 2007, det. G. M. Kay, conf. E. J. Clement. © PLANT RECORDS 265 ©Amaranthus hybridus 44/1.2. *31, Hunts.: disturbed ground, Upwood Airfield, TL265846, A. Wallis, 2006, conf. T. C. E. Wells. tLychnis chalcedonica 46/18.cha. *7(0), Cumberland: roadside bank, Borrowdale Road, Keswick, NY267229, Moffat, 2006, LANC 2007. tPersicaria capitata 47/1.cap. *70, Cumberland: cracks in garden path, Cockermouth, NY 117306, S. R. Cowen, 2006, det. R. E. Groom, LANC. tFallopia xbohemica (F. japonica x sachalinensis) 47/5.1x2. *62, N. E. Yorks.: large patches in woodland, Peasholm Glen, Scarborough, TA033892, V. Jones & W. A. Thompson, 2007, herb. V. Jones. +Rheum xhybridum (R. palmatum x rhaponticum 47/7.palxrha. — *49, Caerns.: persisting on soil dumped by coast path, Criccieth, SH4837, W. McCarthy & M. Stead, 2007. +Rumex scutatus 47/8.2. *77, Lanarks.: trackside, Harthill, NS9064, P. Macpherson & J. R. Hawell, 2007, det. E. J. Clement, herb. P.M. Probably the result of fly tipping. +Paeonia officinalis 49/1 .off. 51, Flints.: dense birch woodland, no doubt an established garden throw-out but some distance from houses, Caerestyn, E of Caergwrle, SJ316575, G. Wynne, 1999. 2nd published record but predates that published in Welsh Plant Records 2006. tHypericum xinodorum (H. androsaemum x hircinum) 51/1.3x¢4. *31, Hunts.: near buildings, Monks Wood Experimental Station, TL201798, K. J. Walker, 2006. Colonised from nearby garden. *52, Anglesey: naturalised by former habitation, Holyhead, SH224823, O. Mountford, 2006. 69, Westmorland: plant 2—3 m tall in old quarry, Levens, SD499870, G. Halliday, 2007, LANC. Garden outcast. Ist record since early 19th C. tHypericum ‘Hidcote’ 51/1.Hid. *44, Carms.: established on roadside bank near Amroth, SN182076, A. O. Chater et al. 2007, det. G. Hutchinson. ©Hypericum olympicum 51/l.oly. L. *62, N. E. Yorks.: a few plants self-sown at base of wall, Kirklevington, NZ429099, V. Jones, 2007, conf. E. J. Clement, herb. V. Jones. @ +Lavatera xclementii (L. olbia x thuringiaca), 53/2.olbxthu. *69, Westmorland: waste ground by infilled canal, Kendal, SD518930, A. M. Boucher, 2007, LANC. +Cucurbita pepo 60/CUT.pep. *17, Surrey: probably a single large plant on top of manure heap on arable field edge to W of Raikes Farm, Sutton Abinger, TQ103463, S. D. Mellor, 2007. Large, glaucous, globose fruits. tSisymbrium altissimum 62/1.5. 46, Cards.: reseeded road verge, Waun crossroads, Aberystwyth, SN599821, S. P. Chambers, 2007, conf. T. C. G. Rich, NMW. 2nd post-1930 record & Ist as neophyte. tErophila glabrescens 62/22.3. *48, Merioneth: weed of flower beds, Barmouth, SH61, P. M. Benoit, 2007. tLepidium sativum 62/30.1. 46, Cards.: path-side, Cwm Padarn, Llanbadarn Fawr, SN600814, S. P. Chambers, 2007. 2nd post-1950 record and Ist as a neophyte. tBrassica napus subsp. oleifera 62/34.2a. 44, Carms.: several plants in corner of construction site, Whitland, SN203167, R. D. Pryce, 2007. Ist record of subspecies. tLysimachia ciliata 69/4.4. *42, Brecs.: probable bird-seed introduction into garden near Bwlch, SO142229, J. Hieatt, 2007, det. M. Porter, herb. M.P. tLysimachia punctata 69/45. *31, Hunts.: legume-rich verge, Conington Airfield, TL194868, N. Crossman, 2006, conf. T. C. E. Wells. *Deutzia scabra 71/DEU.sca. *17, Surrey: small bush in hedgerow, Hoyle Hill, Capel, TQ176421, P. Follett & S. A. Follett, 2007. tHydrangea macrophylla 71/HYD.mac. *49, Caerns.: sandy bank at top of beach, Porth Dinllaen, SH2741, W. McCarthy, 2007. Also seen by W. McCarthy on slate quarry waste at Llanllechid (SH6369). *52, Anglesey: naturalised by former habitation, Holyhead, SH224823, O. Mountford, 2006. tRodgersia podophylla 74/2.1. | *52, Anglesey: several sizeable clumps well established in woodland, Pwll-fanogl, Llanfair-Pwllgwyngyll, SH5271, N. Brown, 2007. tBergenia xschmidtii (B. crassifolia x ciliata) 74/3.1xcil. *46, Cards.: one clump established on grassy slope by scrub, SSW of Lifeboat Station, New Quay, SN389597, A. O. Chater & J. P. Poland, 2007. @ +Tolmiea menziesii 74/7.1. *44, Carms.: originally introduced but well established on shady riverside bank, Maes-yr-Hafod, Trelech, SN303331, R. D. & K. A. Pryce, 2007. 266 PLANT RECORDS tTellima grandiflora 74/8.1. 43, Rads.: near Ednol Hill, Radnor Forest, SO235642, R. G. Woods, 2007. Ist record since 1961. tSpiraea xrosalba (S. salicifolia x alba) 75/3.1x2. *52, Anglesey: naturalised by former habitation, Holyhead, SH224823, O. Mountford, 2006. +Aruncus dioicus 75/4.1. *62, N. E. Yorks.: persisting and naturally regenerating by pool in village, Thornton-le-Dale, SE836831, V. Jones & W. A. Thompson, 2007. +Holodiscus discolor 75/5.1. *62, N. E. Yorks.: three large and four small plants in mere-side scrub, the latter presumably natural regeneration, The Mere, Scarborough, TA035859, V. Jones & W. A. Thompson, 2007, herb. V. Jones. tFilipendula xpurpurea, 75/6.pur. *62, N. E. Yorks.: several long-established plants at edge of damp woodland, Grey Towers, Nunthorpe, NZ533135, V. Jones & W. A. Thompson, 2007, conf. E. J. Clement, herb. V. Jones. Previously known as Filipendula camtschatica +Kerria japonica 75/7.1. *42, Brecs.: hedge, near Bwlch, SO162223, M. & C. Porter, 2006. *44, Carms.: self-sown in mortared stone wall, Llandovery Castle, SN767342, R. D. Pryce, 2007, NMW.1st published record. tRubus armeniacus 75/8.134. *31, Hunts.: woodland edge on edge of village, Alwalton, TL132961, D. A. Broughton, 2007. *Potentilla recta 75/9.7. *42, Brecs.: seeding on gravel paths in garden (not planted), Llangynidr, SO151198, M. Tillotson, 2006, det. M. Porter, herb. M.P. *Duchesnea indica 75/12.1. *69, Westmorland: beneath trees in St Thomas’s churchyard, Kendal, SD513931, A. M. Boucher, 2007, LANC. tAcaena ovalifolia 75/18.3. *62, N. E. Yorks.: appeared as a garden weed, Ingleby Greenhow, NZ580063, V. Jones, 2007, conf. E. J. Clement, herb. V. Jones. May have been brought in by visiting birds. +Acaena inermis 75/18.4. *69, Westmorland: weed of garden path, Kendal, SD5192, M. James, 2007, LANC. tAlchemilla conjuncta 75/19.2. *44, Carms.: garden escape at seaward end of valley, Morfa Bychan, near Pendine, SN2207, J. Rees, 1996, det. G. Hutchinson, NMW. Also seen, by G. Hutchinson, in 2007, spreading on Carboniferous Limestone wall top, Crwbin (SN473131). tRosa rugosa 75/21.6. *31, Hunts.: road verge on urban fringe, Peterborough, Orton Southgate, TL140942, D. A. Broughton, 2007. *108, W. Sutherland: a few scattered bushes on roadside, Scourie, NC1543, R. Maskew, 2007. +Rosa rubiginosa 75/21.18. 48, Merioneth: in quantity and flourishing in a newly planted hedge near Coleg Meirion, Dolgellau, SH71, P. M. Benoit, 2007. Ist recent record. tRosa majalis Herrm. 75/21.maj. *37, Worcs.: Wannerton Downs near Kidderminster, SO8778, W. Mathews, 1849, WOS, det. R. Maskew. @ Only confirmed v.c. record. *+Rosa sempervirens L. 75/21.sem. *37, Worcs.: bushy place around pit in field, Madresfield, SO8047, R. F. Towndrow, 1881, MANCH, NMW WOS, det. R. Maskew. @ Many specimens, the latest dated 1920. Only confirmed British records. *tPrunus persica 75/22.1. *69, Westmorland: small sapling in layby outside rugby ground, Kendal, SD521940, A. M. Boucher, 2007, LANC. +Sorbus ‘Joseph Rock’, 75/28.Jos. _*44, Carms.: tree planted in waste ground with seedlings becoming established in vicinity, Bynea, SS551987, G. Hutchinson, 2006, NMUW. @ +Cotoneaster dammeri 75/32.11. *38, Warks.: ten in fruit on grass road verge, Manby Road, Castle Vale, Birmingham, SP143917, J. W. Partridge, 2007, det. J. Fryer, WARMS. Var. ‘major’. *44, Carms.: naturalized & spreading in garden of Dylan Thomas’ Boathouse, West Tip, Cross Hands, SN306110, R. D. Pryce, 2001, det. J. Fryer, NMW. Also seen, in 2007, by R. D. & K. A. Pryce on trackside at Laugharne (SN561131). +Cotoneaster xsuecicus (C. dammeri x conspicuus) 75/32.11x18. *38, Warks.: three or four on trackside, Warmington Road, Birmingham, SP153841, J. W. Partridge, 2007, det. J. Fryer, WARMS. Birmingham & Black Country Flora Project. +Cotoneaster lacteus 75/32.13. *62, N. E. Yorks.: bird-sown bush on waste ground by railway, E of Riverside Stadium, Middlesbrough, NZ510205, V. Jones & W. A. Thompson, 2007. tCotoneaster integrifolius 75/32.16. 38, Warks.: roadside cement kerb, Sydenham, Leamington Spa, SP328644, J. W. Partridge, 2007, det. J. Fryer, WARMS. Ist record since 1932. PLANT RECORDS 267 tCotoneaster hjelmqvistii 75/32.22. *69, Westmorland: cemetery wall by infilled canal, Parkside Road, Kendal, SD518919, A. M. Boucher, 2007, conf. J. Fryer, LANC. +Cotoneaster divaricatus 75/32.26. | *38, Warks.: one on wall by pavement, The Boulevard, Chester Road, Birmingham, SP116937, J. W. Partridge, 2007, det. J. Fryer, WARMS. Birmingham & Black Country Flora Project. +Cotoneaster villosulus 75/32.29. *17, Surrey: one small shrub in pit, Howell Hill, Ewell, TQ239619, E. J. Taylor & P. Wakeham, 2007. +Cotoneaster dielsianus 75/32.37. | *70, Cumberland: west end of roadside crags, north side of A66, Dubwath, NY 194310, M. S. Porter, 2002, conf. J. Fryer, LANC photo. +tCotoneaster franchetii 75/32.39. *38, Warks.: two or three by fence on derelict land, Gravelly Industrial Park, Birminghmam, SP100894, J. W. Partridge, 2006, det. J. Fryer, WARMS. Birmingham & Black Country Flora Project. +Cotoneaster sternianus 75/32.40. | *38, Warks.: one fruiting seedling in alley, Red House, Leamington Spa, SP328654, J. W. Partridge, 2005, det. J. Fryer, WARMS. tPyracantha rogersiana 75/33.2. *69, Westmorland: thicket near end of canal, Stainton, SD520853, A.M. Boucher, 2007, LANC. +Crataegus persimilis 75/35.5. *62, N. E. Yorks.: single large bush in old hedgerow, by A168 near Thornton-le-Moor, SE393899, V. Jones & M. J. Yates, 2007, conf. W. A. Thompson. tCercis siliquastrum 76A/CER.sil. *17, Surrey: a self-sown seedling in a flower bed in the War Museum Park, Geraldine Mary Harmsworth Park, Lambeth, TQ315790, N. H. Bertrand, 2007. @ tTrifolium incarnatum subsp. incarnatum 77/19.19a. 70, Cumberland: single large plant on graded soil on site of flood defence works, Strand Road, Carlisle, NY407502, H. & B. Spencer, 2007, LANC. Ist record since 1970. tLaburnum xwatereri (L. anagyroides x alpinum) 77/22.1x2. _*44, Carms.: hedge, Llech- Sion, NE of Cwmduad, SN394325, A. O. Chater, 1997, NMW. Ist record as a neophyte and pre- dating records already published. +Elaeagnus umbellata 78/2.1. *44, Carms.: Bronwydd Arms, SN4123, A. S. Lewis, 2002, det. G. Hutchinson, NMW. tMyriophyllum aquaticum 79/2.2._*31, Hunts.: in lake, Crown Lakes Country Park, Farcet, TL196942, N. Millar, 2002, conf. T. C. E. Wells. Fuchsia magellanica 84/5.1. | *70, Cumberland: one bush on wall by River Caldew, Carlisle, NY397556, R. E. Groom, 2007. +Cornus sericea 85/1.2. *31, Hunts.: scrub between river and railway, Woodston, TL1898, D. A. Broughton, 2006. Well naturalised in public open spaces around Peterborough. +Euonymus fortunei (Turcz.) Hand.-Mazz. 88/1.for. *64, Mid-W. Yorks.: Blubberhouses, SE1658, Wharfedale Nats. Soc., 2007. @ tIlex xaltaclerensis (I. aquifolium x perado) 89/1.1xper. *62, N. E. Yorks.: several self- sown bushes in woodland by railway bank, Valley Road, Weaponess, Scarborough, TA035876, V. Jones & W. A. Thompson, 2007. @ tEuphorbia dulcis 91/2.5. *44, Carms.: Dylan Thomas’ Boathouse garden, Laugharne, SN306110, J. Rees, 1992, det. T. Walker, NMW. tEuphorbia amygdaloides subsp. robbiae 91/2.16b. *38, Warks.: on quarry spoil heap, Griff, Nuneaton, SP364893, B. Proctor, 2007, det. J. W. Partridge, WARMS. ©Euphorbia characias subsp. characias 91/2.17a. *62, N. E. Yorks.: a few self-sown plants at base of wall, Briggswath, NZ874089, V. Jones, 2007, conf. Perry’s garden centre. ©Euphorbia characias subsp. veneta 91/2.17b. *62, N. E. Yorks.: several self-sown plants at top of wall, back alley by Askham Bryan College, Guisborough, NZ617163, V. Jones, 2007, conf. E. J. Clement, herb. V. Jones. tKuphorbia oblongata 91/2.obl. *38, Warks.: building site, New Bond Street, Birmingham, SP085863, M. W. Poulton, 2004, det. E. J. Clement, WARMS. +Acer cappadocicum 99/1.2. _*52, Anglesey: woodland, Plas Newydd, SH524704, P. Denne, — 2007. *69, Westmorland: 2 m sapling in hedgerow near lake shore, south of Pooley Bridge, NY468240, C. S. Crook, 1991, LANC. _ Acer saccharinum 99/1.5. *62, many self-sown amongst planted trees and in adjacent wasteland, N of A66, South Bank, NZ533209, V. Jones & W. A. Thompson, 2007, herb. V. Jones. 268 PLANT RECORDS ©Ruta graveolens L. 1O1A/RUT.gra. *17, Surrey: one plant on a waste site behind railings, the ruins of St. Agnes Terrace, south of Waterloo, Lambeth, TQ314775, N. H. Bertrand, 2007. @ ©Impatiens balfourti Hook.f. 105/1.bal. *17, Surrey: N side, at entrance to yard between nos. 20b and 24, on small patch of waste ground, Chestnut Grove, Balham, TQ2873, R. Vickery, 2007, BM. © ©Lycopersicon esculentum | 10/7.1. *48, Merioneth: one plant at base of wall, Barmouth railway station, SH61, P. M. Benoit, 2007. ©Solanum sisymbriifolium 110/8.sis. | *35, Mons.: plants of unknown origin in greenhouse and garden, Steppes Cottage, Jingle Street, Wonastow, SO476107, H. V. Colls, 2007. +Petunia xhybrida (P. axillaris x integrifolia) 110/PET.axixint. *64, Mid-W. Yorks.: town centre, Otley, SE202455, B. Brown, 2007. *69, Westmorland: pavement weed, Library Road, Kendal, NY513928, A. M. Boucher, 2007, LANC. +tCalystegia silvatica subsp. disjuncta Brummitt 111/3.4b. *44, Carms.: roadside hedge- bank, NW of Pont-ar-llechau, SN719254, A. O. Chater & R. D. Pryce, 1997, NMW. Ist v.c. record pre-dating previously reported records. © ©Gilia achilleifolia Benth. 114/GIL.ach. *62, N. E. Yorks.: four plants self-sown at edge of woodland by road verge, N of Seamer, NZ492112, E. Gendle, 2007, det. E. J. Clement, herb. V. Jones. © +Symphytum orientale 1|16/4.6. *46, Cards.: road verge, N side of Ffordd Sulien, Llanbadarn Fawr, SN598809, S. P. Chambers, 2007, conf. C. O’ Reilly, NMW. +Anchusa azurea 116/6.3. *52, Anglesey: roadside hedgerow, 20yds. from old school, Llangoed, SH619799, E. Phenna, 2006. *Amsinckia micrantha 116/12.2. *46, Cards.: re-seeded road verge, Waun crossroads, Aberystwyth, SN599821, S. P. Chambers, 2007, NMW. 4 Sieber si £U ZUUY Botanical Society of the British Isles Registered as a charity in England and Wales Number 212560 and in Scotland Number SC038675 Applications for membership should be addressed to the Hon. General Secretary, c/o Department of Botany, The Natural History Museum, Cromwell Road, London, SW7 5BD, from whom copies of the Society’s Prospectus may be obtained. Officers for 2008-2009 President, Mr M. E. Braithwaite Vice-Presidents, Mr R. D. Pryce, Mr R. G. Woods, Dr F. J. Rumsey, Mrs J. M. Croft Honorary General Secretary (joint), Miss L. Farrell and Mr D. A. Pearman Honorary Treasurer, Mr A. M. Nixon Editors of Watsonia Papers and Notes, M. J. Y. Foley, M. N. Sanford* Book Reviews, J. R. Edmondson Plant Records, M. S. Porter Obituaries, M. Briggs *Receiving editor, to whom all MSS should be sent (see inside back cover). © 2009 Botanical Society of the British Isles The Society takes no responsibility for the views expressed by authors of Papers, Notes, Book Reviews or Obituaries. The cover illustration of Euphorbia hyberna L. (Irish Spurge) was drawn by Rosemary Wise. eee ee r—— -utsonia 27: 283-300 (2009) | } ! i | | ABSTRACT An account of the Whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, Somerset, England is given. Sorbus aria is abundant, and variable. Sorbus anglica and S. eminens are quite widespread, but S. porrigentiformis is rarer. Three new polyploid species endemic to the gorge are described supported by DNA analysis: S. cheddarensis, S. eminentoides ind S§. rupicoloides. The gorge is one of the most mportant sites for Sorbus in Britain. Browsing is fecting significant numbers of Sorbus trees, and azing is affecting regeneration. KEYWORDS: Somerset. Endemic, goats, microsatellites, INTRODUCTION Cheddar Gorge in the Mendips, Somerset, is Britain’s largest gorge and one of its best known limestone features. The Carboniferous Limestone cliffs tower up to c. 120 m above the gorge bottom on the south side, with extensive screes, Slopes and smaller cliffs on the north side. The gorge is part of the Cheddar Complex Site of Special Scientific Interest (SSSD) (which includes the former Cheddar Gorge SSSI), designated for a suite of rare plants, animals and communities and the geological interest. The north side is owned by the National Trust, the south side by the Longleat Estate. Cheddar Gorge has been known for many years as a rich botanical site with nationally rare plants such as Dianthus gratianopolitanus Vill, Galium fleurotii Jord. and the Cheddar endemic Hieracium stenolepiforme (Pugsley) P. D. Sell & C. West, and other uncommon species such as Geranium purpureum Vill., Cerastium pumilum Curtis and Sedum forsterianum Sm. The gorge is also home to * . iS . . e-mail: tim.rich@museumwales.ac.uk n account of the Whitebeams (Sorbus L., Rosaceae) of Cheddar Gorge, England, with description of three new species | _L. HOUSTON, A. ROBERTSON, K. JONES, S. C. C. SMITH, S. J. HISCOCK School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG and T. C. G. RICH Department of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP* rare Sorbus species, but there is relatively little data on them compared to other Cheddar rare plants. In their review of the rare plants in Cheddar, FitzGerald & McDonnell (1997) pointed out that the gorge was of great importance for Sorbus and recommended a full audit of the species present. The common and widespread Sorbus aria (L.) Crantz has been known in Cheddar Gorge for many years and is abundant. Sorbus anglica Hedl. was first recorded in 1901 by A. Ley (Roe 1981), and has always been regarded as rare. Sorbus porrigentiformis E. F. Warb. was first confirmed in 1966 from material collected by A. W. Gravestock, and is also rare (Green et al. 1997). Surprisingly, S. eminens E. F. Warb., now known to be the second commonest species in the gorge, was only first reported by Proctor & Groenhof (1992), though it had been previously collected without a name by E. C. Wallace in 1935 and by F. Rose in 1948. Sorbus aucuparia L. and S. torminalis (L.) Crantz are known in the Cheddar area, but neither has been recorded from the gorge itself (Roe 1981; Green et al. 1997). Cheddar Gorge and the surrounding area were grazed for many centuries by cattle, sheep and goats, when trees and scrub were largely confined to the cliffs. After the Second World War, stock grazing decreased and scrub species such as Cornus sanguinea L., Crataegus monogyna Jacq., Rhamnus cathartica L., Taxus baccata L., Viburnum lantana L. and Sorbus aria began to invade the grasslands, the scrub eventually developing locally into secondary woodland. The whitebeams, especially S. aria, benefited during this time and spread into the grasslands. Scrub clearance has been carried out to maintain open grasslands for the rare herbaceous species, but as rare whitebeams 284 L.HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S. J. HISCOCK & T. C. G. RICH were known to be present, few Sorbus trees were cleared, and soon small groves of whitebeams were quite frequent in the gorge. To enable some S. aria trees to be cleared without affecting the rare Sorbus species, L. Houston was commissioned to survey the rare whitebeams by the National Trust in 2005 and by the Longleat Estate in 2006. Full details are given in Houston (2006, 2009) and the results are summarized here with descriptions of three new species found during the surveys. METHODS DNA MICROSATELLITES Bud samples were collected for DNA analysis from ten S. aria, six S. anglica, six S. eminens and ten S. cheddarensis during the autumn of 2005 and from a single S$. eminentoides and a single S$. rupicoloides in 2006. They were placed in separate bags that were sealed and stored at 4°C for four days after which they were transferred to a freezer and stored at -20°C until needed. DNA _ was extracted using Qiagen® DNeasy® Plant Mini Kit 2004 protocol. Four nuclear microsatellite loci were amplified using two sets of primers developed for Malus x domestica (Gianfranceschi et al. 1998) and two pairs developed for Sorbus torminalis (Oddou-Muratorio ef al. 2001). Each of these primer-pairs were redesigned, from sequences deposited in Genbank, to facilitate amplification of a wide range of Sorbus taxa (Table 1). Forward primers were labelled with fluorescent dyes. Reverse primers were tailed with a non-target-specific sequence to overcome allele-sizing problems associated with the primer sequence dependant adenylation of PCR products by most Tag polymerases (Brownstein et al. 1996). PCR reaction mixtures (20ul total volume) contained the following components/concen- trations: one unit of Hotstart Taq DNA polymerase (Qiagen), 2ul of 10x reaction buffer (Qiagen), 2 mM of each dNTP (Bioline), with 0-2uM of each forward and reverse primer and 20-50 ng of genomic DNA. PCR reactions were performed in a MJ-Research PTC-200 Peltier Thermal Cycler. The following programme was used: | step at 95°C for 15 minutes followed by 35 cycles of | minute at 94°C, 1 minute at 55°C, 1 minute and 30 seconds at 72°C, and a 10 minute extension step of 72°C. PCR amplification products from the labelled primers were separated on a laser based capillary electrophoresis instrument, the ABI 3730xl. Alleles were sized relative to an internal size standard and resulting electro- pherograms were analysed using GeneMarker genotyping software. Each individual electro- pherogram was scored manually. DATA ANALYSIS The most appropriate measure to study genetic diversity for apomictic taxa is to compile and compare multilocus genetic phenotypes (Gornall 1999). Therefore, for each individual analysed in this study, microsatellite allele phenotypes were determined for each of the four amplified loci. These data were combined to construct the multilocus phenotypes of each sample. The numbers of individuals belonging to each multilocus phenotype were determined for each species. Ploidy level estimates were based on the maximum number of displayed alleles at a single locus. For example for S. cheddarensis, three alleles were displayed at TABLE 1. NUCLEOTIDE SEQUENCES OF NUCLEAR AND CP DNA PRIMERS USED IN THIS STUDY Locus Dye Primers Repeat Na CHOIFO2! 6-Fam Forward CCACATTAGAGCAGTTGAGGATGA (GA)>> 9 Reverse ATAGGGTAGCAGCAGATGGTTGT CH02D11' Vic Forward AAATAAGCGTCCAGAGCAACAG (AG)», 10 Reverse GGGACAAAATCTCACAAACAGA MSS5" 6-Fam Forward CCCCAACAACATTTTTCTCC (GA)j9 12 Reverse CCTCTCGCTCTTTGCCTCT MSS16 | Vic Forward ATGTCACATCTCTCCCCTTGTGT (GA)og 5 Reverse TTTTGCCCTCAAAGAATGCCTTA ‘Microsatellite Primers derived from Malus x domestica DNA (Gianfranceschi et al. 1998) ' Microsatellite Primers derived from S. torminalis (Oddou-Muratorio et al. 2001). Na = the number of alleles amplified across all taxa. SORBUS IN CHEDDAR GORGE 285 locus FO2 suggesting it is at least triploid. Each multilocus phenotype was transformed into bi- nary code and metric distances were computed between them using Dice similarity co- efficients (Dice 1945) with NTSYSPC version 2.1la software (Rohlf 2002). The relationships among the multilocus phenotypes were visualised using Principal Coordinate Analysis (PCO) with NTS YSPC software (Rohlf 2002). MORPHOLOGY AND DESCRIPTIONS OF NEW SPECIES Broad leaves from the short, vegetative shoots in sunlit situations, excluding the oldest and youngest leaf (Aas ef al. 1994), were measured on herbarium material of the three new taxa in NMW. Fruits were measured on fresh material, and the colours matched against the Royal Horticultural Society colour charts (Royal Horticultural Society 1966). POLLEN STAINABILITY Pollen stainability, an estimate of potential pollen viability (Rich 2009), was investigated using Alexander’s Stain (Alexander 1969) on the flowering collections available. Anthers were removed from herbarium specimens with tweezers under a low-power binocular micro- scope, and placed on a slide with a drop of Alexander’s Stain, warmed briefly on the hotplate, then broken up with the tweezers. The preparation was then covered with a cover-slip N | S500 m and replaced on the hotplate to improve the uptake of the stain. The slides were then examined under a _ high-power compound microscope for areas of dense pollen grains. Potentially viable grains were counted as those which were large and rounded-triangular with cell walls which stained green and _ with cytoplasm inside which stained uniformly bright red. Small deformed grains or those staining green only with very little or no red staining inside (i.e. no cytoplasm) were considered infertile. The numbers of pollen grains counted are cited for individual trees. SOIL PH Soil samples were collected from around the roots of a few trees in 2007, and air-dried. The pH was subsequently measured on a 50 soil:50 distilled water mixture with a calibrated pHep2 Hanna pocket-sized pH meter. FIELD SURVEY Field surveys were mainly carried out by L. Houston during the period 14 July to 10 August 2005, and 24 August to 27 October 2006, with extra visits with T. Rich during 2005, 2006 and 2007. Existing records were also used to direct surveys (e.g. M. C. F. Proctor unpublished 1989 data held at NMW;; FitzGerald & McDonnell 1997; T. Rich unpublished data 2001-2004). Figure | shows the areas of the gorge searched; _=_—"—“~“~_—_~__. -2 FIGURE |. Areas of Cheddar Gorge surveyed for Sorbus. W~-=- =~ - > eee tL 286 L.HOUSTON, A. ROBERTSON, K. JONES, S. C. C. SMITH, S. J. HISCOCK & T. C. G. RICH the sheer size and scale of the gorge meant that it was not possible to survey it compre- hensively, and thus the coverage shown in Figure 1 must be borne in mind when interpreting the distribution maps. All locations and details of the rarer taxa were recorded, but for S. aria which is frequent throughout the site usually only brief details were noted. For the rarer trees, GPS grid references were recorded using a Garmin Etrex hand-held unit where possible. Tree height was estimated in metres, and the girth of accessible trees measured with a tape measure at 1:3 m. Notes were made on the growth form (maiden or coppice), and damage by goats, from conservation management work or from landscape work was noted. Numerous photographs were taken. Most trees were named in the field. Trees on cliffs or with leaves high in the canopy were either sampled using a long-handled pruner or identified as far as possible with binoculars. No abseiling or climbing with or without ropes was undertaken. It was not possible to name with certainty many trees which were inaccess- ible on cliffs or trees which had lost leaves. Voucher specimens for some rarer trees or those requiring a second opinion were collected and named in consultation with T. Rich: the vouchers have been deposited in NUW. RESULTS NUCLEAR MICROSATELLITE ANALYSIS In total, 36 alleles from four microsatellite loci were amplified from 34 adult Sorbus individuals. The number of alleles amplified at each locus across all taxa ranged from five at locus MSS16 to 12 at locus MSS5 (Table 1). The total number of different alleles displayed by each taxon ranged from nine for S. cheddarensis to 15 for S. rupicoloides (Tables 2 and 3). When data from all loci were combined, the total number of detectable multilocus phenotypes scored across all taxa was 15. Sorbus anglica and S. eminens each consisted of just a single multilocus phenotype, as would be expected of a clonally reproducing species (Table 3). Similarly each of the S. cheddarensis individuals shared the same multilocus phenotype, supporting the hypothesis that S. cheddarensis is a novel Sorbus species that, like S. anglica and S. eminens, reproduces apomictically (Table 3). In contrast, each S. aria individual had a unique TABLE 2. GENETIC DIVERSITY MEASURES FOR THE CHEDDAR GORGE SORBUS TAXA Taxon Ni Xo S. aria 10 2 S. anglica 6 3 S. cheddarensis 10 3 S. eminens 6 4 S. eminentoides l 4 S. rupicoloides ] 4 Xe AG Ng 2 14 10 4 12 1 9 1 4 12 ] 13 15 Ni = number of individuals sampled; Xo = observed ploidy, based on the maximum number of displayed alleles at a single locus; Xe = expected ploidy, based on published cytological work; A’ = total number of different alleles seen across all loci; Ng = number of multilocus phenotypes detected. TABLE 3. MULTILOCUS PHENOTYPES FOR THE CHEDDAR POLYPLOID SORBUS TAXA AND THE TOTAL ALLELE POPULATION FOUND AT EACH LOCUS FOR S. ARIA Taxon Alleles at locus: F02 MSS5 D11 MSS16 S. aria 196,198,200,206 126,132,134,144,146 163,177,193 166,172 S. anglica 192,202,206 1261365142 157,159,205 162,166,170 S. cheddarensis 196,206,208 132,140 159,193 166,178 S. eminens 196,202,206 120,126,140,142 159,163,169 166,178 S. eminentoides 196,202,208 120,132,136,140 159,161,205 166,168,178 S. rupicoloides 200,204,208,214 126,132,138,180 159,169,175,189 166,168,170 SORBUS IN CHEDDAR GORGE multilocus genotype, as would be expected of an obligate out-crossing diploid species. All individuals genotyped for S. aria displayed either one or two alleles at each locus, confirming them to be diploid. In contrast, each individual genotyped among S. anglica, S. cheddarensis, S. eminens, S. eminentoides and S. rupicoloides displayed more than two alleles (usually three or four) at a minimum of one locus, confirming that they were all polyploid (Table 3). The triploid ploidy level of S. anglica and the tetraploid ploidy level of S. eminens indicated by these alleles are supp- orted by existing chromosome counts from Cheddar (Bailey et al. 2008). The alleles suggest that S$. cheddarensis is at least triploid, and S. eminentoides and S. rupicoloides tetraploid, which require cytological confirmation. Genetic similarity among the 15 multilocus phenotypes was calculated and the subsequent similarity matrix was used as the input for 0.65 2 0.37 e 4 Dim-3 0.09 o----------0 See Se =O) Nn Dim-2 -0.19 -0/38 -0.65 -0.55 -0.25 0.05 Dim-] 287 Principal Coordinate Analysis (PCO). In addition to the 15 multilocus phenotypes found during the current study the multilocus pheno- types of S. porrigentiformis and S. rupicola, found during a previous study of Sorbus taxa (Robertson et al. 2009), were included in the PCO analysis. These two taxa were included as they are possible putative parents of S. cheddarensis, S. eminentoides and S. rupi- coloides, and samples of these two taxa were not readily available from the sampled area of the current study. The PCO analysis revealed that the 17 multilocus phenotypes were divided clearly into two groups, group | consisting of S. aria and group 2 consisting of S. cheddarensis, S. eminentoides and S. rupicoloides and the other four polyploid apomicts (Fig. 2). The widespread nature of the S. aria group shows clearly the genetic diversity found among the individuals while the close grouping of the seven polyploid taxa Ar ‘ At’ O1 "ee An lf O Art 3 on ii 1 Po rly 1 ! O 1 ie | Ru a AHR orton etn Emit! OG ee O; 0.35 0.65 FIGURE 2. Principal Coordinate Analysis of the similarity relationships among the 15 Cheddar Gorge and two Avon Gorge Sorbus multilocus phenotypes, showing the first three dimensions. Two major groupings are easily identified; the grouping to the left of the figure consists solely of S. aria, and the grouping to the right consists solely of Sorbus polyploids. These three dimensions account for 57-44% of the total variation found among the phenotypes. Ar = S. aria, An = S. anglica, Ch = S. cheddarensis, Em = S. eminens, Emt = S. eminentoides, Po = S. porrigentiformis, Ru = S. rupicola and Rud = S. rupicoloides. 288 L. HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S.J. HISCOCK & T. C. G. RICH demonstrates how closely related they are and how likely it is that they share similar origins. Further sampling is required to determine the exact parentage of the new taxa. DESCRIPTIONS OF NEW SPECIES Sorbus cheddarensis L. Houston & Ashley Robertson, sp. nov. HOLOTYyPUS: Cheddar Gorge, v.c. 6, North Somerset, England, ST472544, 19 September 2007, T. C. G. Rich & L. Houston, Cheddar survey number 601 (NMW, accession number VEZ 007 AIz13 0): Vernacular name: Cheddar Whitebeam Frutex vel arbor parva ad 7 m ut minimum. Gemmae ovoideae, virides, ad apicem pilis albidis pilosae. Folia lata brachyblastorum (8:5—)9-0-12-0(-12:5) x 6-0-8:0 cm, (1:35-) 1-4-1-7(-1-8)plo longiora quam _latiora, elliptica, ad longitudinis suae 48—58(—60) partes centensimas latissima, apice acuto vel obtuso, basi cuneata, in centro laminae (Q0—)2— 11(-13) partibus centensimis ad costam plerumque leniter lobata, marginibus uniserratis dentes parvos acutos prorsum projectos ferentibus vel leniter biserratis, sed 1—2 centimetris infimis prope petiolum fere integris; pagina superior viridis glabra; pagina inferior viridula tomentosa, venis (19—)20—24 ad angulum (24—)27—37° a costa tentis. Petioli 13—24 mm. Inflorescentia ad 8 cm lata, ramulis viridialbis tomentosis instructa. Flores 15— 17 mm _ lati, fragrantes. Sepala anguste triangularia, viridula sed ad basin tomentosa, in marginibus glandibus nullis vel paucis. Petala 7-10 X 5—7 mm, anguste elliptica, cupulata, alba. Antherae cremeae, nonnunquam dilutissime roseotinctae. Styli 2, in triente infimo colligati, virides, basi pilosa. Fructus maximi (10—-)10°5-13°5-14) x (d0O-)l1— 13) mm, 0O-9-1-1(-1-:2)plo longiores quam latiores, plerumque aspectu globosi, in medio latissimi, in maturitate rubri, lenticellis paucis parvis vel mediocribus praediti. Shrub or small tree to at least 7 m. Buds ovoid, green, pilose at tip with whitish hairs. Broad leaves of short shoots (8:5—)9-0—12-0(—12°5) x 6:0-8:0 cm, (1:35—)1-4—1-7(—1-8) times as long as wide, elliptic, widest at 48-58(—60)% along leaf length, with apex acute to obtuse and base cuneate, usually shallowly lobed (Q—)2—11(— 13)% of way to midrib at centre of leaf: margins uniserrate with small acute, forwardly- directed teeth to weakly biserrate, but with the lowest c. 1-2 cm near the petiole nearly entire; upper surface green, glabrous; lower surface greenish tomentose, with (19—)20—-24 veins held at an angle of (24—)27—37° to the midrib. Petioles 13—24 mm. Inflorescence to 8 cm across, with branchlets greenish-white tomentose. Flowers 15-17 mm across, fragrant. Sepals narrowly triangular, greenish but tomentose at base, without or with a few glands on margins. Petals 7-10 x 5-7 mm, narrowly elliptic, cupped, white. Anthers cream, sometimes with vaguest hint of pink. Styles 2, joined for the lowest third, green, hairy at base. Largest fruits (10—)10-5—13-5(-14) x (10-)11-13 mm, 0-9- 1-1(—1-2) times as long as wide and usually looking globose, widest at the middle, red at maturity (RHS colour 42A, 44A, 45B), with a few small to medium-sized lenticels. Sorbus cheddarensis is a difficult species to characterize in words, but is distinct in the field once known. It has elliptic leaves which are widest just above the middle (but not enough to be regarded as obovate), which curve to the base and apex symmetrically in the shape of an eye, with the shallow lobes on the margins cutting in to join the lamina at right angles, small, neat, acute, forward-directed teeth, and greenish-tomentose underneath with the prominent, parallel, whitish veins standing out noticeably (Fig. 3). It is a member of Sorbus subgenus Aria Pers. Sorbus cheddarensis differs from S. aria in the greener underside to the leaves, the regular, neat toothing, and globose fruits (white undersides to the leaves with variable teeth and lobing, and fruits usually longer than wide in the very variable S. aria). Sorbus porri- gentiformis differs in having obovate leaves and fruits wider than long. Sorbus cheddarensis is readily separated from all other Cheddar taxa; S. anglica and S. intermedia have deeply lobed leaves, S$. eminentoides has broadly elliptic leaves with large, coarse toothing, and S. rupicoloides has much narrower leaves. Two plants collected in 2001 by T. Rich and reported as S. aria and S. porrigentiformis, now redetermined as S. cheddarensis, had the widespread Aria group ‘A’ chloroplast type (Chester et al. 2006). The pollen stainability of the type tree was 71% (82 grains counted; A rn terri i ei i At ta i SORBUS IN CHEDDAR GORGE 289 FIGURE 3. Leaves and fruits of Sorbus cheddarensis. Scale bars 1 cm. A. Patto, pers. comm. 2007). Fruit set in 2006 varied between trees. Acute leaf lobes in S. aria may be a sign of introgression from S. torminalis in the Wye Valley (Price & Rich 2007), but there is no evidence of S. torminalis in this species. Sorbus cheddarensis occurs on Carbon- iferous Limestone slopes and rocks in Cheddar Gorge, Somerset (v.c. 6), England. Soil from under the type tree was pH 7-2. It was first recognised as a distinct taxon by L. Houston in 2005 whilst surveying the whitebeams, and was given the nickname ‘Pinstripe’ on account of the prominent parallel veins as seen against the greenish lower surface of the leaf. It mainly grows associated with S. aria. 290 L. HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S. J. HISCOCK & T. C. G. RICH FIGURE 4. Leaves and fruits of Sorbus eminentoides. Scale bars 1 cm. Sorbus eminentoides L. Houston, sp. nov. HOLOTYPUS: Cheddar Gorge, v.c. 6, North Somerset, England, ST466539, 19 September 2007, T. C. G. Rich & L. Houston, Cheddar survey number 261 (NMW, accession number V2007 2a: ISOTYPUS: BM, CGE, E, OXF. Vernacular name: Twin Cliffs Whitebeam Frutex vel arbor parva ad 9(-15) m ut minimum; cortex cinereobrunneus. Gemmae ovoideoconicae, virides, ad apicem pilis albidis pilosae. Folia lata brachyblastorum (7-0—)8-0- 11-0(-11-5) x 5:5—8-5(—9-0) cm, (1-1—)1-2-1-4 (—1-5)plo longiora quam latiora, late elliptica, SORBUS IN CHEDDAR GORGE 291 raro late ovata, ad longitudinis suae (35—)41-— 58(-68) partes centensimas latissima, apice acuto vel raro acuminato, basi cuneata vel raro margine extrorsus arcuato, 11 partibus centensimis ad costam leniter lobata vel dentibus longis (c. 2-3 mm) latis acuminatis exstantibus valde biserrata, sed 2—3 centimetris infimis prope petiolum fere integris; pagina superior viridis glabra; pagina inferior cinereoviridis tomentosa, venis 16-24 ad angulum 30—40(—41)° a costa tentis. Petioli 10— 19 mm. Inflorescentia ad 8 cm lata, aggregata, leniter tholiformis, ramulis viridialbis tomentosis instructa. Flores 16-20 mm lati, fragrantes. Sepala anguste triangularia, viridula sed ad basin tomentosa, in marginibus sine glandibus. Petala 8-10 x 65-8 mm, late elliptica, cupulata, alba. Antherae cremeae, lenissime roseotinctae. Styli 2(—3), usque ad basin discreti, virides, basi pilosa. Fructus maximi (11-:5—)12—-14:5(-15) x = 12-15(- 15:5) mm, 0-9-1-Iplo longiores quam latiores, plerumque aspectu globosi vel longiores quam latiores, in maturitate rubri, lenticellis paucis parvis vel mediocribus praediti. Shrub or small tree to at least 9(-15) m; bark greyish—brown. Buds _ ovoid-conical, green, pilose at tip with whitish hairs. Broad leaves of short shoots (7-0—)8:0-—11-0(—11-5) x 5-5—8-5(— 9-0) cm, (1-1—)1-2—1-4(—1-5) times as long as wide, broadly elliptic, rarely broadly ovate, widest at (35—)41—58(—68)% along leaf length, with apex obtuse or rarely acuminate and base cuneate or rarely with its margin arching outwards, weakly lobed to 11% of way to midrib or strongly biserrate with long (c. 2- 3. mm) broad acuminate teeth directed outwards, but with the lowest c. 2—3 cm near the petiole nearly entire; upper surface green, glabrous; lower surface greyish-green tomentose, with 16—24 veins held at an angle of 30-40(-41)° to the midrib. Petioles 10- 19 mm. Inflorescence to 8 cm across, crowded, weakly domed, with branchlets greenish-white tomentose. Flowers 16—20 mm across, fragrant. Sepals narrowly triangular, greenish but tomentose at base, without glands on margins. Petals 8-10 x 65-8 mm, broadly elliptic, cupped, white. Anthers cream with a hint of pink. Styles 2(—3), split to base, green, hairy at base. Largest fruits (11-5—)12—14-5(-15) x 12- 15(-15:5) mm, 0-9-1-1 times as long as wide and usually looking globose or longer than wide, red at maturity (RHS colour 44A, 45A, 46B), with a few small to medium-sized lenticels. Sorbus eminentoides is characterised by the large, broadly elliptic to roundish leaves with large teeth or weak lobes (the point at which strongly biserrate becomes weak lobing is debatable) and the large fruits which are about as long as wide (Fig. 4). It 1s a member of Sorbus subgenus Aria. The broadly elliptic leaves separate it from most other British taxa with the exception of S. eminens and S. aria. It most closely resembles S. eminens in leaf shape (hence the specific epithet) but has large, coarse toothing whereas in S. eminens the teeth are fine and regular. Sorbus aria differs in having narrower, often ovate leaves with whitish undersides and fruits usually longer than wide. Sorbus eminentoides is readily separated from all other Cheddar taxa; S. anglica and S. intermedia have deeply lobed leaves, S. cheddarensis has elliptic leaves with fine toothing, S. porrigentiformis has obovate leaves and fruits wider than long, and S. rupicoloides has much narrower leaves. Sorbus eminentoides occurs on Carbon- iferous Limestone cliffs at the south-west end of Cheddar Gorge, Somerset (v.c. 6), England. It was first found and collected in 2006 by L. Houston whilst surveying the Whitebeams. It grows associated with S. anglica, S. aria, S. eminens, S. porrigentiformis sensu lato and S. rupicoloides. The soil under the type tree had pH 7-6. The pollen fertility of the type tree assessed using Alexander’s Stain was 94% (54 grains counted) and of another tree, no. 138, 88% (77 grains counted). Sorbus rupicoloides L. Houston & T. C. G. Rich, sp. nov. HOLOTYPUS: Cheddar Gorge, v.c. 6, North Somerset, England, ST467539, 19 September 2007, T. C. G. Rich & L. Houston, Cheddar survey number 254 (NMW, accession number WOOT IS): ISOTYPUS: BM. Vernacular name: Gough’s Rock Whitebeam Frutex vel arbor parva ad 7 m ut minimum, ramis saepe deorsum = arcuatis; cortex cinereobrunneus. Gemmae _ ovoideoconicae, virides, ad apicem pilis albidis pilosae. Folia lata brachyblastorum (6:5—)7-0—12-0(-12°5) x (4-0-)4-5-7:5(-8:5) cm, (1:5—)1-6-1-8plo longiora quam latiora, obovata, ad longitudinis suae 55— 65 partes centensimas latissima, apice acuto vel acuminato, basi cuneata, plerumque non lobata, marginibus apicem versus leniter biserratis FIGURE 5. Leaves and fruits of Sorbus rupicoloides. Scale bars 1 cm. dentes exstantes ferentibus, sed triente infimo fere integro; pagina superior atroviridis glabra, grosse rugosa, convexa; pagina inferior cinereoviridis tomentosa, venis 16-21 ad angulum 25-35° a costa tentis. Petioli 10- 21 mm, crassi. Inflorescentia aggregata, leniter tholiformis, ramulis albis tomentosis, ubi fructiferis glabrescentibus, instructa. Flores 15— 18 mm lati, fragrantes. Sepala anguste triangularia, tomentosa, 1n marginibus sine glandibus. Petala 8-10 x 6—8 mm, late elliptica vel magis minusve orbicularia, cupulata, alba. Antherae dilutissime roseae. Styli 2, in triente infimo colligati, virides, basi pilosa. Fructus maximi 11-J4(15) x (2—)13—17(18) mm, 0-73-0:97plo longiores quam _latiores, plerumque aspectu latiores quam longiores, in medio latissimi, in maturitate rubri sed inaequaliter maturescentes, lenticellis paucis vel aliquot parvis vel magnis _ praediti, lenticellis magnis praecipue fructus basin versus locatis. Shrub or small tree to at least 7 m, with branches often arching downwards; bark greyish-brown. Buds _ ovoid-conical, green, pilose at tip with whitish hairs. Broad leaves of short shoots (6:5—)7:0—12:0(—12:°5) x (4:0—)4-5— 7:5(—8:5) cm, (1:5—)1-6—-1-8 times as long as wide, obovate, widest at 55-65% along leaf length, with apex acute to acuminate and base cuneate, usually unlobed, with the margins towards the apex weakly biserrate with teeth directed outwards, but with the lowest third nearly entire; upper surface dark green, glabrous, coarsely rugose, convex; lower surface greyish-green tomentose, with 16-21 veins held at an angle of 25—35° to the midrib. Petioles -10-21 mm, stout. Inflorescence crowded, weakly domed, with branchlets white tomentose, glabrescent in fruit. Flowers 15— 18 mm across, fragrant. Sepals narrowly triangular, tomentose, without glands on mar- gins. Petals 8-10 x 6-8 mm, broadly elliptic to more or less orbicular, cupped, white. Anthers very pale pink. Styles 2, joined for lowest third, green, hairy at base. Largest fruits 1 1—14(—15) «x, (12—)13-17€E18)) mm: 0:73=0-97 stimesmas long as wide and usually looking wider than long, widest at the middle, red at maturity (RHS colour 44A, 45A, 46A) but ripening unevenly, with a few to a moderate number of small to large lenticels, the large lenticels mainly towards the base of the fruit. Sorbus rupicoloides is characterized by the narrow, obovate, acute leaves which are greenish-white underneath and the large, broad fruits (Fig. 5), which distinguish it from most other British taxa with the exception of the SORBUS IN CHEDDAR GORGE 293 narrow-leaved §. rupicola (Syme) Hedlund, S. vexans E. F. Warb. and S. margaretae M. C. F. Proctor (Taxon D sensu Proctor et al. 1989). It is a member of Sorbus subgenus Aria. At first glance, S. rupicoloides looks like S. rupicola (hence our epithet) but differs in having acute leaves with biserrate toothing (obtuse, oblong-obovate and with more or less uniserrate teeth which curve towards the leaf apex in §. rupicola), and has generally larger ism (averace 2:8 Boe ye” Nie pe o 2 5 = = = —- =~ pe FIGURE 8. Distribution of Sorbus eminens at Cheddar Gorge in areas surveyed. 296 LL. HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S. J. HISCOCK & T. C. G. RICH N | 500 m ~“. FIGURE 9. Distribution of Sorbus eminentoides at Cheddar Gorge in areas surveyed. N 500 m FIGURE 10. Distribution of Sorbus porrigentiformis sensu lato at Cheddar Gorge in areas surveyed. SORBUS IN CHEDDAR GORGE 297 FIGURE 11. Distribution of Sorbus rupicoloides at Cheddar Gorge in areas surveyed. extent or another (20% overall), and the high proportion of dead trees recorded on _ the National Trust side which were due to goat browsing (7 out of 9 trees), indicate the scale of the problem. It is not certain if the relatively high damage to S. cheddarensis and S. eminens (Table 4) is due to selective browsing of these species, or is related to their occurrence in areas the goats frequent; a specific study is required. Trees particularly affected are those beside the paths regularly used by goats, and the effects are most marked on the sunnier, more open slopes on the National Trust side, at Commoners’ Spur and Heart Leaf Bluff, where the goats prefer to reside. The main impact of animals on the Longleat side appears to be prevention of regeneration, which may be due to a combination of goats and Soay sheep. The goats have proved capable of destroying some small trees altogether, in a gradual process, by breaking off branches and crown to leave the remains of the stem as a moribund stick. Young shoots produced at the base have then been vulnerable to selective browsing. Smith (2005) reported that goats will browse to a height of up to 2 m in woodland, and we noted that in more open areas trees with slender stems up to 3 m tall were vulnerable to being pulled or broken down. Some goats were even observed climbing in trees. One S. anglica tree at Horseshoe Bend cut down during conservation work in 2002 did survive initially by re-sprouting from the base, but the young shoots were selectively browsed and it was dead by 2005. In a few cases, photographs taken by T. Rich in 2001 provide clear evidence of damage to individual trees (e.g. S. anglica no. 76), and other trees watched between 2005 and 2007 were seen to become progressively more damaged. -DISCUSSION Eight Sorbus species were recorded in Cheddar Gorge, of which the three new taxa S. cheddar- ensis, S. eminentoides and S. rupicoloides are endemic to the gorge, and S. eminens and S. porrigentiformis are endemic to Britain. This is clearly a nationally important site for Sorbus, ranking with the Avon Gorge, Wye Valley, Craig-y-Cilau and the north Devon-Somerset coast as centres of diversity. It is perhaps surprising that the importance of Cheddar for Sorbus has been so relatively poorly known, though this reflects the difficulties both of surveying the gorge and of identifying Sorbus 298 L. HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S. J. HISCOCK & T. C. G. RICH FIGURE 12. Photographs showing scrub invasion and woodland development at Cheddar. A. Old postcard, pre- Second World War. B, Approximate same view, 2005. (the recognition of the new taxa here aided by DNA), but it is possible that the post-war removal of grazing has allowed the rarer taxa to spread onto the screes and slopes so that they are now more easily found in accessible places. The survey has also shown many more individuals of the rarer species than had previously been known. Many of the rarer trees are also concentrated around the massive cliffs and slopes at the west end of the gorge, but not exclusively so. The conservation statuses of S. cheddarensis, S. eminentoides and S. rupic- oloides are ‘Critically Endangered’ following the IUCN (2001) guidelines. The diversity in Cheddar, as in the Avon Gorge (cf. Robertson et al. 2009), is partly attri- buted to presence of S. aria and the ease with which it hybridizes with S. porrigentiformis. The genetic and morphological diversity within S. aria is also important, this site containing one of the more morphologically variable populations of S. aria seen by T. Rich. Our survey data show that there is damage to the Sorbus populations by goats and other animals. Goats were introduced to Cheddar to control scrub, at which they are very effective and an important conservation tool (Smith & Bullock 1993). The National Trust herd is currently five, free-ranging, neutered males whose preference is for the north side of the gorge at the west end. The goats were initially confined to an enclosure, but soon escaped and continued to do so when corralled back into the enclosure. On the Longleat Estate side of the gorge, a large breeding herd has _ been introduced recently to an extensive enclosure, but these have also escaped. While some goats have been reported from as far away as Hutton, near Weston-super-Mare 20 km to the west (D. Bullock, pers. comm. 2008), others have crossed to the sunnier north side of the gorge; there were 25 feral goats in the gorge by 2007 (D. Bullock, pers. comm., 2008). Smith & Bullock (1993) found that goats selectively targeted Sorbus species in summer, and Smith (2005) noted that it may be the bark stripping during the winter which actually kills the trees. Goats tend to prefer sheltered, sunny, open areas, and will browse in some areas more than others, locally affecting the vegetation; this may cause problems where it coincides with the occurrence of rare species. The herd of feral Soay sheep have been in the gorge since 1992; their population has varied between 34 and 124 re SORBUS IN CHEDDAR GORGE 299 between 2000 and 2007 (D. Bullock, pers. comm. 2008). The sheep also browse saplings, and have an impact on the Sorbus species. A recent study of the impact of sheep and goats in Cheddar by Wheatley (2009) has confirmed that significant damage to the whitebeams is taking place. There is a nature conservation dilemma at Cheddar Gorge. On the one hand, it is important to maintain the open grassland for rarities such as Dianthus gratianopolitanus; on the other hand, it is important to conserve the Sorbus which prefers the intermediate stages of scrub succession to woodland. A comparison of modern day views with those shown in old photographs shows the extent of scrub recolon- ization in the gorge (Fig. 12). Allowing further development to closed woodland will not be acceptable for either rare grassland species or Sorbus, but continued heavy grazing would probably result in restriction of most of the Sorbus species to cliffs or places inaccessible to animals. Quite how all these conflicting priorities can be resolved remains to be seen, but clearly both need accommodating perhaps by compartmentalization of parts of the gorge. Finally, we strongly advise anyone wishing to see these whitebeams to take extreme care due to the number of loose rocks on the slopes and cliffs, and the large number of people who visit the gorge. Many of the Sorbus taxa can be seen safely from the public paths, but permission must be obtained to access the area of Gough’s Cave where S. rupicoloides occurs as it is above buildings in an area with restricted public access. ACKNOWLEDGMENTS We would like to thank Liz McDonnell for sharing her knowledge, Mark Courtiour and the National Trust for funding the 2005 survey and Hugh Cornwell and the Longleat Estate for funding the 2006 survey, David Bullock and Jude Smith for help regarding the impacts of goats, and Philip Oswald for Latin translations. REFERENCES AAS, G., MAIER, J., BALTISBERGER, M. & METZGER, S. (1994). Morphology, isozyme variation, cytology and reproduction of hybrids between Sorbus aria (L.) Crantz and S. torminalis (L.) Crantz. Botanica ‘Helvetica 104: 195-214. ALEXANDER, M. P. (1969). Differential staining of aborted and non-aborted pollen. Stain Technology 44: 117- I BROWNSTEIN, M. J., CARPTEN, J. D. & SMITH, J. R. (1996). Modulation of non-templated nucleotide addition by Taq DNA polymerase: Primer modifications that facilitate genotyping. Biotechniques 20: 1004-1010. CHESTER, M., COWAN, R. S., FAY, M. F. & RICH, T. C. G. (2007). Parentage of endemic Sorbus L. (Rosaceae) species in the British Isles — evidence from plastid DNA. Botanical Journal of the Linnean Society 154: 291-304. Dice, L. R. (1945). Measures of the amount of ecological association between species. Ecology 26:297—302. FITZGERALD, R. & MCDONNELL, E. J. (1997). Cheddar Gorge rare plants survey 1997. Unpublished report to English Nature. GIANFRANCESCHI, L., SEGLIAS, N., TARCHINI, R., KOMJANC, M. & GESSLER, C. (1998). Simple sequence repeats for the genetic analysis of apple. Theoretical and Applied Genetics 96: 1069-1076. GORNALL, R. J. (1999). Population genetic structure in agamospermous plants. pp. 118-138 in HOLLINGSWORTH, P. M., BATEMAN, R. M. & GORNALL, R. J., eds. Molecular systematics and plant evolution. Taylor and Francis, London. GREEN, R. P., GREEN, I. P. & CROUCH, G. A. (1997). The atlas flora Uf Some rel Crewkerne. HOUSTON, L. (2006). Cheddar Gorge Sorbus survey 2005. Unpublished report to the National Trust, March 2006. HOUSTON, L. (2009). Cheddar Gorge Sorbus survey 2006. Unpublished report to the Longleat Estate, 2009. IUCN (2001). IUCN Red list categories and criteria. Version 3.1. 1UCN, Gland. ODDOU-MURATORIO, S., ALIGON, C., DECROOCQ, S., PLOMION, C., LAMANT, T. & MUSH-DEMESURE, B. (2001). Microsatellite primers for Sorbus torminalis and related species. Ecology Notes 1: 297-299. PRICE, D. T. & RICH, T. C. G. (2007). One-way introgressive hybridisation between Sorbus aria and S. torminalis (Rosaceae) in southern Britain. Watsonia 26: 419-432. PROCTOR, M. C. F. & GROENHOF, A. C. (1992). Peroxidase isoenzyme and morphological variation in Sorbus L. in South Wales and adjacent areas, with particular reference to S$. porrigentiformis E. F. Warb. Watsonia 19: 21-37. PROCTOR, M. C. F., PROCTOR, M. E. & GROENHOF, A. C. (1989). Evidence from peroxidase polymorphism on the taxonomy and reproduction of some Sorbus populations in south-west England. New Phytologist 112: 569-575. RICH, T. C. G. (2009). Pollen stainability in British Sorbus L. Plant Ecology and Diversity 2: 85-88. 300. ~L. HOUSTON, A. ROBERTSON, K. JONES, S.C. C. SMITH, S. J. HISCOCK & T. C. G. RICH ROBERTSON, A., RICH, T. C. G., ALLEN, A., HOUSTON, L., ROBERTS, C., HARRIS, S. A. & HISCOCK, S. J. (2009). Hybridization and polyploidy as drivers of continuing evolution and speciation in Sorbus (in preparation). ROE, R. G. B. (1981). The flora of Somerset. Somerset Archaeological and Natural History Society, Taunton. ROHLF, F. J. (2002). NtSYSpc, numerical taxonomy and multivariate analysis system. Version 2.11a, User guide. New York: Exeter Software ROYAL HORTICULTURAL SOCIETY (1966). R. H. S. colour chart. Royal Horticultural Society, London. SMITH, J. (2005). Grazing for the Avon Gorge? A feasibility study. Unpublished report to the Avon Gorge and Downs Wildlife Project. SMITH, J. E. & BULLOCK, D. J. (1993). A note on the summer feeding behaviour and habitat use of free- ranging goats (Capra) in the Cheddar Gorge SSSI. Journal of Zoology, London 231: 683-688. WHEATLEY, H. R. (2009). Are feral goats (Capra) and Soay sheep (Ovis aries) having an impact on tree species in Cheddar Gorge, in particular towards rare Whitebeams (Sorbus spp.)? Unpublished BSc. Conservation Biology thesis, University of the West of England. (Accepted April 2009) Watsonia 27: 301—313 (2009) 301 The diversity of Sorbus L. (Rosaceae) in the Lower Wye Valley ier G. RICE Department of Biodiversity & Systematic Biology, National Museum of Wales, Cardiff CF 10 3NP L. HOUSTON School of Biological Sciences, University of Bristol, Woodland Road, Bristol BSS 1UG C. A. CHARLES 14 Bath Street, Frome, Somerset, BAII IDN and A. C. TILLOTSON 69 Leighton Road, Southville, Bristol BS3 INR ABSTRACT The diversity of Sorbus in the Lower Wye Valley on the border between England and Wales is described. Sixteen Sorbus taxa occur in the Lower Wye Valley, including two new species which are described as S. parviloba T. C. G. Rich and S. saxicola T. C. G. Rich. Diverse, important sites include the Symonds Yat to Great Doward area, and the Pen Moel to Shorn Cliff area. Overall, the Lower Wye Valley is the second most diverse site for Sorbus in the British Isles. There is evidence of a minor decline in diversity of a few species at a few sites. KEYWORDs: England, Wales, new species. INTRODUCTION Britain, southern Scandinavia and_ the mountains of south-east to central Europe are the three main centres of diversity of Sorbus L. (Rosaceae) in Europe. Within Britain, the six key areas of diversity are the North Devon and North Somerset coast, Cheddar Gorge, Avon Gorge, the Lower Wye Valley, the Brecon Beacons and Arran. A few other areas such as Morecambe Bay, the Mendips and South Devon also support some endemics. In this paper an account of the diversity and distribution of taxa within one of these, the Lower Wye Valley, is given based on surveys carried out between 1999 and 2006, with the description of two new species. The data contribute towards Target | “Understanding *e-mail: tim.rich@museumwales.ac.uk and documenting plant diversity’ under the Global Plant Conservation Strategy (Secretariat for the Conservation of Biodiversity 2002). The Lower Wye Valley is a large, meander- ing, north-south orientated valley c. 40 km long situated on the border between England and Wales extending from Ross-on-Wye to the Severn Estuary (Fig. 1). At the north end in the Symonds Yat area between English Bicknor and Monmouth, the river meanders between spectacular, large, wooded Carboniferous Limestone cliffs. The river then straightens running south through a broad valley over sandstones, before entering a second series of meanders through another wooded Carbon- iferous Limestone gorge between Tintern and Bulwark. Three vice-counties are represented in the Lower Wye Valley; v.c. 34 West Gloucestershire and v.c. 36 Herefordshire in England, and v.c. 35 Monmouthshire in Wales. The natural history of the valley has recently been described by Peterken (2008). Significant sites are protected as Sites of Special Scientific Interest and Special Areas for Conservation, and it is an Area of Outstanding Natural Beauty. In 1999, T. Rich began a project looking at diversity and evolution in Sorbus and visited many areas in the Wye Valley. In 2000, the Wye Valley was included in the European Ravine Wood Life Project as part of the Natura 2000 initiative, and English Nature (now Natural England) commissioned abseil access 302 TC) G; RICH, LX HOUSTON GA’ CHAREES AND AY Gil rOrson ts V.c. 36 Herefordshire Si 8. Bowlers Hole Me to Symonds Yat \ 18. Little Doward \ and Dennis Grove ‘ -_ oe . 3 pao ee ™ 10. Rosemary Se Topping | 17. Far Hearkening Rock 9. Symonds Yat Rock EN, 16. Lady to Coldwell Rocks nw ¢ Ay Monmouth _——, Farle Wood ter 19. Seven Sisters Rocks 20. The Biblins to Symonds Yat West | | | R | i : V.c. 34 West | e Gloucestershire A W y ikon Ween BS: - Bigsweir Monmouthshire St Briavels Brockweir | Tintern 7. Shorn Cliff 15. Black Cliff — to Cockshoo I 14. The Wynd Cliff — 13. Piercefield Park to Lover's Leap | 12. Chepstow Castle 3. Woodcroft Quar =~ 2. Pen Moel 1 km 11. Bulwark Cliffs 5 , 6. Offa's Dyke, Tidenham 5. Ban-y-gor Rocks | ; : 4. hie agrees) to Aleove Woods NC Tutshill Severn \| Estuary , LEN Chepstow \ 4% NY 1. Sedbury - Tutshill Cliff t ry FIGURE 1. Location of Sorbus sites surveyed in the Lower Wye Valley. SORBUS IN THE LOWER WYE VALLEY 303 surveys of Sorbus on the major cliffs to provide information for conservation and management (Houston et al. 2001, 2002, 2003, 2004). These abseil surveys provided a wealth of new Sorbus material, which significantly augmented what could be collected from cliff tops. The data from both these studies are combined in this paper to give an overview of the diversity of Sorbus in the Wye Valley. METHODS MORPHOLOGY AND DESCRIPTIONS OF NEW SPECIES Broad leaves from the short, vegetative shoots in sunlit situations, excluding the oldest and youngest leaf (Aas et al. 1994), were measured on herbarium material of the two new taxa in NMW. Fruits were measured on fresh material, and the colours matched against the Royal Horticultural Society colour charts (Royal Horticultural Society 1966). Conservation status assessments were made for the new species following the IUCN (2001) guidelines. HISTORICAL INFORMATION AND SURVEYS Historical information was compiled from herbaria (BM, BRISTM, GLR, K, NMW), the voucher material of Proctor & Groenhof (1992) now held at NMW, and from the literature (e.g. Evans 2007; Riddelsdell et al. 1948). Surveys were carried out between 1999 and 2008, mainly in the autumn when fruits were present helping identification. Two main methods were used: 1. ‘Look-see’ surveys (Rich et al. 2005) were carried out by walking along cliff tops and sometimes cliff bottoms, and noting taxa seen without consistently counting exact numbers of each species. Some material was collected using a long-handled pruner. 2. Abseil or rope access surveys were carried out by abseiling down rock-faces in a series of ‘drops’ along the cliffs at intervals of 3— 10 m according to the frequency of Sorbus or presence of hazards such as loose rock. Samples were collected directly from the ropes, or sometimes using a long-handled pruner. Numbers of each species were counted, but due to practicalities of identification some counts are minima. In certain areas only quicker cliff-top or cliff- base ‘look-see’ surveys were carried out. Full details are given in the survey reports (Houston et al. 2001, 2002, 2003, 2004). Collection of voucher specimens concen- trated on the rarer or more difficult to identify taxa. Identifications of voucher material were confirmed by T. Rich, with other material determined in the field. The 755 specimens collected are deposited in the National Museum of Wales (NMW). Most of the taxa found during the surveys were allocated to existing taxa, but the situation with the S. eminens E. F. Warb. group and S. porrigentiformis E. F. Warb. group was more complex (see also Proctor & Groenhof 1992). In the S. eminens group, plants with orbicular leaves matching the widespread Avon Gorge- Mendips form were called S. eminens sensu stricto, and plants with leaves subrhombic in outline with deeper teeth as S. eminentiformis T. C. G. Rich, following Rich & Proctor (2009). Proctor & Groenhof (1992) noted that these two species probably belonged to separate taxa, and regarded all the Wye Valley S. “eminens’ as belonging to S. eminentiformis. Our more extensive collections coupled with analysis of the plastid DNA (Chester ef al. 2007) indicate both taxa are present throughout the valley. Warburg’s type specimen was collected from Offa’s Dyke, Tidenham, v.c. 34 West Gloucester-shire (BM; Warburg 1952, 1957) but 1s a poor specimen which is not clearly referable to either form, but he stated that the description related to plants from the Avon Gorge and Lower Wye Valley. However, an unusual feature of Warburg’s description was that his S. eminens had fruits up to 2 cm long which were longer than wide; this does not equate to either our observations of S. eminens as currently interpreted, or those of Proctor & Groenhof (1992). Searches of Offa’s Dyke at Tidenham have failed to reveal any plants with such long fruits, but several plants were found at Shorn Cliff immediately below Offa’s Dyke with long fruits which had leaves like S. eminens but not identical (NMW); this problem is best dealt with by amending the description of S. eminens sensu stricto to having fruits wider than long. Warburg (1952, 1957) also described S. porri- gentiformis from Offa’s Dyke at Tidenham, and this probably represents the widespread, lowland form (Proctor & Groenhof 1992). Again we have failed to find any plants which clearly match the type, though at least one tree remains with relatively narrow leaves on the banks of Offa’s Dyke (this does not have leaves as narrow as S. saxicola described below however). As Warburg (1957) pointed out, plants from different locations differ, and we 304 T. C. G. RICH, L. HOUSTON, C. A. CHARLES AND A. C. TILLOTSON have recorded plants with small, obovate leaves and fruits wider than long as S. porri- gentiformis sensu stricto, and similar plants with larger leaves of more variable shape as S. porrigentiformis sensu lato. With both the S. eminens group and S. porrigentiformis group we suspect that there are a significant number of hybrids with S. aria in the Wye Valley, as also recently reported for the Avon Gorge (Rich ef al. 2009); con- firmation of these will have to await DNA analysis. Plants from Symonds Yat discussed as similar to S. graeca (Spach) Kotschy by Proctor & Groenhof (1992) are here included in S. porrigentiformis sensu lato, and are probably also locally derived apomictic hybrids of S. aria and S. porrigentiformis. The sites surveyed are as follows, ordered first by vice-county and second from south to north with approximate grid references (Fig. 1): 1. Sedbury — Tutshill Cliff (ST541939), v.c. 34 These 20-40 m high, W- to SW-facing cliffs are about | km long and fall directly to the River Wye with private gardens above. There are scattered S. aria trees visible with binoculars, but they have not been surveyed in detail due to difficult access. Pen Moel (ST539955), v.c. 34 Pen Moel cliffs form a series of WNW- facing crags about 30-50 m high and 300 m long above the River Wye, and much, except the very top of the cliffs, has been quarried below and is wooded to the top edge and from below on the old quarry workings. The top has been landscaped in places as a garden. The cliffs were surveyed by walking in autumn 2003 (Houston eft al. 2004) and using ‘look-see’ in 2006. 3. Woodcroft Quarry (ST541959), v.c. 34 Northwards from Pen Moel, Woodcroft Quarry has an inner section of W-facing, 40 m high quarried cliffs c. 100 m long, behind a WNW-facing out section of 30-50 m cliff (‘Fly Wall’) which drops down to secondary woodland and quarry spoil below. The top part of the outer cliffs has been completely removed, and _ what remains seems to be the only section of this line of cliffs left unquarried. The quarried inner section and part of the outer sections were surveyed by rope drops and parts by walking in autumn 2003 (Houston et al. 2004) and using ‘look-see’ in 2006. N 4. Wintour’s Leap (ST542964), v.c. 34 The huge, quarried, 100 m high, NW- to SW-facing cliffs of Wintour’s Leap cont- inue north for c. 500 m from Woodcroft Quarry to Lancaut. They have secondary woodland over quarry spoil below, and private gardens above with a short section of public footpath. The cliffs and woods are part of Gloucestershire Wildlife Trust’s Lancaut Nature Reserve. The cliffs were surveyed in a series of rope drops in autumn 2003 (Houston et al. 2004) and using ‘look-see’ in 2006. 5. Ban-y-gor Rocks — Cockshoot (ST547972), v.c. 34 Ban-y-gor Rocks and Cockshoot form a series of crags about | km long with a N- to WNW-aspect. The western crags are 15—20 m high and dip slightly eastwards to give increasingly shaded wooded slopes above until the obvious break of ‘Ladder Gully’. A clear path follows their base. The c. 60 m high eastern crags are crossed by a wide vegetated ledge (‘The Terrace’) with most botanical interest above, and are bounded at the top by private gardens. Nine days were spent surveying the cliffs in autumn 2000 by a systematic series of vertical rope drops (Houston ef al. 2001). Several subsequent visits were made in 2002 to collect further material. 6. Offa’s Dyke, Tidenham (including Plumweir Cliff and Devil’s Point) ST547985 to 542001), v.c. 34 Offa’s Dyke, an ancient earthwork c. 2—5 m high, has a series of small, SW- to NW- facing .limestone crags (such as Pulpit Rock) and rocky slopes associated with it in the Tidenham area, but most is in woodland and some in conifer plantation. Both S. eminens and S. porrigentiformis were originally described from this area, but it is not known exactly where the type trees were. Offa’s Dyke has been surveyed using ‘look-see’ in 2001 noting interesting taxa. 7. Shorn Chiff (ST541992), v.c. 34 The Shorn Cliff is an unquarried, natural, 20-35 m high, WNW-facing limestone crag about 600 m long overlooking Tintern Abbey. There is a series of open buttresses and rock faces with more forested sections between. There is dense woodland below the crag with some climbers’ paths up from the forest track, and a very steep, wooded slope above the crag beiow Offa’s Dyke. 8. SORBUS IN THE LOWER WYE VALLEY Four days were spent surveying the cliff in a series of rope drops in autumn 2001 (Houston eft al. 2002). One subsequent visit was also paid in 2002. Bowler’s Hole — Symonds Yat (SO562158), v.c. 34 This WNW-facing, 20-50 m high series of cliffs and rocky outcrops is about 500 m long. It is very popular with climbers and tourists and 1s partly fenced. There is steep woodland on scree below, and woodland above. The cliffs and rocks were surveyed by ‘look-see’ surveys in autumn 2002 (Houston et al. 2003), and have been visited many times using ‘look-see’ in 1999-2006. Symonds Yat Rock — Coldwell Rocks (SO566158), v.c. 34 This E- to N-facing broken cliff above a meander in the River Wye is about | km long, and contains a series of small (10 m) to large (25 m) and massive (50 m) cliffs, spurs and pinnacles. Parts have been quarried (Quarry Rock) and there was a popular Victorian walk along the top, ‘Coldwell Walks’, which is still visible in places. There is mainly secondary wood- land below and tall, dense woodland above. Some cliff top scrub is cut back around Symonds Yat Rock to maintain views. The cliffs were surveyed in detail in autumn 2002 using abseil drops (Houston et al. 2003) and several ‘look-see’ visits 1999-2006. . Rosemary Topping (SO579163), v.c. 34 There are a series of small, poorly access- ible, mostly shaded, E- and N-facing cliffs above English Bicknor in dense woodland, in total about 100 m long. The cliffs were visited using ‘look-see’ in 2003, but little was found. . Bulwark Cliffs (ST538927), v.c. 35 This is a small series of ENE-facing, partly wooded, limestone crags about 400 m long and 20-30 m tall, which have been quarried away on their west side to form the railway cutting, leaving thin, upright slabs of limestone. The east side drops directly into the River Wye. Access is very limited. The crags have been visited several times using the ‘look-see’ approach, including once in 2001 with T. G. Evans who has visited them regularly for over 30 years. Ie iG: 305 Chepstow Castle — Alcove Woods (ST532942), v.c. 35 The NE to NW-facing crags below Chepstow Castle are c. 400 m long and c. 30—40 m tall, and fall directly into the River Wye. They are partly vegetated. One ‘look-see’ visit in 2005 indicated only S. aria was present, so no further surveys were carried out. The record for S. domestica L. in Wade (1970) refers to S. torminalis (Evans 2007). . Piercefield Cliffs — Lover’s Leap (Sili537/963' 10522968) nv.c: 35 The NW to E-facing cliffs and very steep wooded slopes of Piercefield Cliffs rise from their low eastern end by the River Wye where they are from 5—20 m tall, to larger cliffs mid-slope, and then 40 m cliffs over woodland at their east end at Lover’s Leap, c. 2 km to the west. There are large areas of woodland with relatively limited open areas and a few outcrops. The cliffs have been surveyed using a ‘look-see’ approach on four occasions between 2000 and 2005. . The Wynd Cliff (ST528974), v.c. 35 This dramatic, mainly S-facing cliff c. 100 m high and 500 m long stands out above the river valley, with woodland below and wood- land sloping down to the cliffs at the top. The cliffs were surveyed using a ‘look- see’ approach in 2001. SBilacke @lihta(Silis32983) aAcers> This ENE-facing cliff is about 10-30 m high and 400 m long above the road. Parts have been quarried. The cliffs were surveyed using a ‘look- see’ approach in 2001; little of interest was found. Lady Park Wood (SO546145), v.c. 35 The E-facing cliffs of Lady Park Wood are c. 20 m high and 300 m long above the old railway line and the River Wye. At the top the cliffs are fairly accessible but are heavily wooded with Fagus. Some areas at the bottom have secondary woodland over old quarry spoil. The woodland is a National Nature Reserve. The cliffs were surveyed using system- atic rope drops in autumn 2003 (Houston et al. 2004), and were visited again in 2005 using the ‘look-see’ method. 306 T. C. G. RICH, L. HOUSTON, C. A. CHARLES AND A. C. TILLOTSON 17. Far Hearkening Rock (SO941151), v.c. 35 Far Hearkening Rock is the highest of a series of small (to 5 m) NW-facing rock outcrops about 100 m long in mature woodland, with very limited open rock. The crags were surveyed in autumn 2005 using the ‘look-see’ method. 18. Little Doward and Dennis Grove (SO541158), v.c. 36 These low outcrops to c. 5 m high face SW to NE and total about 400 m long. They are mainly set in woodland, with open areas and forestry on the top of the Little Doward. The crags were surveyed from the top, with one abseil drop in 2002 (Houston et al. 2003), and further visits in 2004 and 2005S. 19. Seven Sisters Rocks (SO545153), v.c. 36 These SW-facing, 10-35 m high crags form a series of outcrops about 400 m long, above steep wooded scree down to the River Wye, and with woodland above. The crags have been visited many times between 1999 and 2006, and most of them were systematically surveyed by abseil drops in autumn 2002 (Houston ef al. 2003). 20. The Biblins — Symonds Yat West (SO555150), v.c. 36 These crags form a series of S-, SE- and ESE-facing cliffs above the Wye about | km long, which vary from c. 5 to 25 m in height. Within this length there are a few outcrops with open cliffs, and much partly wooded cliff, and the cliffs have one to three levels. The cliffs were surveyed using abseil drops at the south end and ‘look-see’ methods in autumn 2002 (Houston ef al. 2003) and ‘look-see’ in 2003. RESULTS DESCRIPTIONS OF NEW SPECIES Sorbus parviloba T. C. G. Rich, sp. nov. HOLOTYPUS: Ship Rock, Coldwell Rocks, Symonds Yat, v.c. 34 West Gloucestershire, SO5/ 1-156; 4 October m9993 Ie Richise vz: Houston (NMW, accession number VEI999332523)) Vernacular name: Coldwell Whitebeam Arbor parva ad 8 m ut minimum, non rhizomatosa; cortex cinereibrunneus. Gemmae ovoideae, acutae. Folia lata brachyblastorum (6:0-)7:0-12:0 x (5-:0—)5:5—10-0 cm, (1-05-) 1-15—1-5(—1-58)plo longiora quam latiora, late elliptica vel elliptica, ad longitudinis suae (44—) 47—63(—68) partes centensimas latissima, apice obtuso vel acuminato, basi cuneata (angulo 42— 60°), in parte supera lobos manifestos sed non profundos (3—)5—14 partibus centensimis ad costam sectos praebentia vel valde biserrata, dentes longos (circa 3-4 mm) acutos vel acuminatos exstantes ferentia, sed parte infima prope petiolum circa 2—3 cm fere integra; pagina superior atroviridis glabra; pagina inferior cinereiviridis tomentosa, venis 17—23(— 24) ad angulum 29-38° a costa tentis. Petioli (8—)10-18(—21) mm. Inflorescentia ad 10 cm lata, aggregata, leniter tholiformis, ramulis viridialbis tomentosis instructa. Sepala anguste triangularia, viridula sed ad basin tomentosa, in marginibus sine glandibus. Petala 9-11-5 x 5- 7 mm, elliptica, alba. Antherae cremeae, lenissime roseotinctae. Styli 2(—3), discreti vel ad medium conjuncti, ad basin pilosi. Fructus maximi 9-13 x (10:5—)11—14 mm, 1-0—1-22plo latiores quam longiores, aspectu latiores quam longiores vel ut videtur subgloboso, in maturitate rubri, lenticellis paucis dispersis parvis vel mediocribus praediti. Small tree to at least 8 m, not rhizomatous; bark greyish-brown. Buds ovoid, acute. Broad leaves of short shoots (6-0—)7-0-12-0 x (5-0—) 5:5—10-:0 cm, (1:05—)1-15—1-5(-1-58) times as long as wide, broadly elliptic to elliptic, widest (44—-)47-63(—68)% along leaf length, with apex obtuse or acuminate and base cuneate (angle 42-60°), with distinct shallow lobes (3—)5-14% of way to midrib in upper part to strongly biserrate with long (c. 3-4 mm), acute or acuminate teeth directed outwards, but with the lowest c. 2-3 cm near the petiole nearly entire; upper surface dark green, glabrous; lower surface greyish-green tomentose, with 17—23(— 24) veins held at an angle of 29-—38° to the midrib. Petioles (8—)10—-18(—-21) mm. Inflorescence to 10 cm across, crowded, weakly domed with branchlets greenish-white tomentose. Sepals narrowly triangular, greenish but tomentose at base, without glands on margins. Petals 9-11:5 x 5-7 mm, elliptic, white. Anthers cream with a hint of pink. Styles 2(—3), free or joined to middle, pilose at base. Largest fruits 9-13 x (10:5—)11—-14 mm, 1-0-1-22 times as wide as long and looking SORBUS IN THE LOWER WYE VALLEY 307 FIGURE 2. Broad leaves of short shoots of Sorbus parviloba. Scale bar | cm. wider than long to subglobose, red at maturity (RHS colour 42a), with a few, scattered small to medium lenticels. Sorbus parviloba is characterized by the broadly elliptic to elliptic leaves with shallow but distinct acute lobes in the upper part of the leaf, a cuneate base and greyish-green under- sides, and the small red fruits just wider than long (Fig. 2). Itis a member of the S. aria group. Sorbus parviloba is a distinctive species, first found and collected in September 1999 by T. Rich and M. Kitchen. Further trees were found in 2002 (Houston et al. 2003). It is endemic to Coldwell Rocks, Gloucestershire (v.c. 34), England, where it occurs from Quarry Rock to Ship Rock on Carboniferous Limestone cliffs and cliff edges in open deciduous woodland with Fagus sylvatica L., Fraxinus excelsior L. and at least five other Sorbus taxa. The altitudes range from c. 100 to 125 m. Only eight trees of this species are currently known (Houston ef al. 2003), but it is likely that more occur on the cliffs and in the woodlands. The provisional IUCN (2001) conservation status is ‘Critically Endangered’. A similar-looking plant from the Wynd Cliff requires DNA investigation to see if it belongs to the same clone. Sorbus saxicola T. C. G. Rich, sp. nov. HOLOTYPUS: cliffs on west side of Symonds Yat, v.c. 34 West Gloucestershire, SO562.157, 21 May 2004, T. C. G. Rich (NMW,, accession number V.2004.3.116). Vernacular name: Symonds Yat Whitebeam Frutex ad 5 m ut minimum, non rhizomatosus; cortex dilute cinereibrunneus. Gemmae ovoideae, acutae. Folia lata brachyblastorum 7-0-11-0 x (4:5-)5-0-7:0 cm, (1:-47—)1:5-1-85 (—2:05)plo longiora quam latiora, anguste obovata, ad longitudinis suae (53—)55—61(—65) partes centensimas latissima, apice acuminato vel aliquando acuto, basi cuneata (angulo 30- 43°), in parte supera lobos non _ profundos obtusos Q—15(—17) partibus centensimis ad costam sectos praebentia vel non praebentia, marginibus dentes parvos acutos _ leniter biserratos prorsum projectos ferentibus, sed parte infima prope petiolum circa 1-4 cm fere integra; pagina superior atroviridis glabra; pagina inferior cinereialba tomentosa, venis 15-22 ad angulum 23—32(—36)° a costa tentis. Petioli (7—)9-17(—21) mm. Inflorescentia ad 9 cm lata, aggregata, leniter tholiformis, ramulis viridialbis tomentosis instructa. Sepala triangularia, viridula sed ad basin inque marginibus tomentosa, in marginibus sine glandibus. Petala 6-9 x 5-6 mm, elliptica vel late elliptica, cupulata, alba, ad basin pilosa. Antherae roseitinctae. Styli 2—3, fere ad basin fissi, ad basin pilosi vel glabri. Fructus maximi (8)9-11(-12) x (9-0-)9:5-11-5 mm, 0-9- 1:05plo longiores quam latiores, aspectu globoso, in medio latissimi, in maturitate rubri vel atrorubri, lenticellis paucis dispersis parvis vel mediocribus praecipue prope basin praediti. Shrub to at least 5 m, not rhizomatous; bark light greyish-brown. Buds ovoid, acute. Broad leaves of short shoots 7-0-11:0 x (4:5—)5-0- 7:0 cm, (1:-47—)1-5—1-85(—2:05) times as long as wide, narrowly obovate, widest (53—)55-61 ¢ Vall d I d 6 SIO Weaing “TT z Z d d Suiddo |, Areurasoy ‘OT o Il tr Sas ECC V1 G (de) eae Alc) al CI 8 (Wer) 1 FLY) 91 SYIOY Pee = — YOOY JA SPUOWAS °6 . Gl fe +8 VI I G (@ilse) i 9 d (Sa? ~ | +OC 6 yooy 1A spuowiks 4 JO ISOM — IOP SOMO °8 = Ol I +E¢ aC Ol (47+)01 9 alg 1S = 0G SI JFYD WIOYS “L VU ¢ d d [d] d d VI weyuapLL “OYA SPIO “9 < Ol G I a +OS +61 +11 +¢ ¢ S0CH asOl yOOYsYIO,) OQ — syooy Jos8-A-urg °¢ es 6 Vv LG ail, e (Cel (Gl ee ILC dea] S INOWIN “p n p ¢ I L : 99 ALIEN) YOINPOOM “€ a ¢ d kel eI) d v (474)8 lS Id] [POW Ud “C 3 I d JFNO ee ae — AINGPIS “T Seyee sf 3 5. ee “oh ie cS G G SS zZ z ae : = S 6 2 Y 5 O S) es AATIVA JAM NI SA9GNOS AO STAYOOTA AO AUVININDS “1 ATAVL 308 309 SORBUS IN THE LOWER WYE VALLEY "| OINSL Ul UMOYS ST Says dy} JO UONRIOT OY, ‘WUepuUNge AT[LIO] = WT “pauNod jou Inq JUdsoId JIOWI= + “APJUDIAI UDIS JOU SP1OS91 STIO}STY = [ ] SOS JULIXO JO IAqUINN ISOM IBEX spuowAg— sulTqrg OU SYOOY SIBISIS UIAIS IAOID) STUD pure preMog any] yooy SUIUDYyAvaH ey PooM WAV Apey JSNO WPA JJM PUAN OULL dea] S .10A0'T > SIFMD Pleo td SPOO MA 9AOTTV — apseg Mojsdayp ‘OC ‘6! SI ‘El Vit Cl yl! nel iG! ‘yuosoid=g ‘sjuvyd pourmyuooun jo Jaquinu = (9)9 ‘ZZ ‘{P) ‘syueyd pounnyuos Jo Joquinu = ‘9)0 Z ‘[ “BIep sANeIWUeNb pue sANeenb Jo omIxtuU kv SI dPqQuI UL, L 8 p ©7He OT ¢ 9 Ol SII I I Cl] 9 G OC 8 Ol ¢ 7 I VI L ({S+) 7 I (It) 1 (THE +06 8 C Val (ces 7 C] VI aris leer 9 +h 7 | (Et) 7 +9] I 7 I d I d (LI+)8 +7 (iT) VI I I +¢ v I SE G d d (ZIF)P Id] d I (iI) | Vial Id] L d d VI d d VI d 6 d VI o > = SS = 2 3 aS) ~ S ® S = S > oO Gog = = v S. S a oo oo = BS = Ss = ee “Cees 8 5 = = g s SS a8 5 = : é = > on Q > =e = S 2 Ss a 5 ese eS oS = = = S bn Ss 8 2 S S 3 ae oo S. =. a 8 g g Q, 5 5 3 ra c 4 > Si o S “LNOO AATIVA JAM NI S09YOS AO SGYOOTY AO AUVINNNS 1 ATAVL 310 T. C. G. RICH, L. HOUSTON, C. A. CHARLES AND A. C. TILLOTSON FIGURE 3. Broad leaves of short shoots of Sorbus saxicola. Scale bar 1 cm. (—65)% along leaf length, with apex acuminate or sometimes acute and base cuneate (angle 30-43°), with or without shallow, obtuse lobes O-—15(-17)% of way to midrib in upper part; margins with small, acute, weakly biserrate teeth directed forwards, but with the lowest c. 1-4 cm near the petiole nearly entire; upper surface dark green, glabrous; lower surface greyish-white tomentose, with 15-22 veins held at an angle of 23—32(—36)° to the midrib. Petioles (7—)9—17(—21) mm. Inflorescence to 9 cm across, crowded, weakly domed with branchlets greenish-white tomentose. Sepals triangular, greenish but tomentose at base and on margins, without glands on margins. Petals 6-9 x 5-6 mm, elliptic to broadly elliptic, cupped, white, pilose at base. Anthers pink- flushed. Styles 2-3, split nearly to the base, pilose at base or glabrous. Largest fruits (8)9- 11(-12) x (9:0—-)9-5—11:5 mm, 0-9-1-05 times as long as wide and looking globose, widest at the middle, red to dark red at maturity (RHS colour 45a), with a few, scattered small to medium lenticels mainly near base. Sorbus saxicola is characterized by the narrowly obovate leaves with or without shallow obtuse lobes in the upper part of the leaf, a usually acuminate apex and a cuneate base, greyish-green underneath with 15-22 veins, and the small, globose, red fruits (Fig. 3). It is a member of the S. aria group. Sorbus rupicola has larger leaves with fewer veins and obtuse apices, and larger fruits wider than long. Sorbus porrigentiformis has broader leaves 1:25-1:65 times as long as wide, and fruits wider than long. SORBUS IN THE LOWER WYE VALLEY 311 Sorbus saxicola was first found and collected in September 1999 by T. Rich and L. Houston, and has since been more widely found in surveys in the Symonds Yat and Doward areas (Houston et al. 2003). Sorbus saxicola is endemic to the Wye Valley around Symonds Yat and the Great Doward, in Gloucestershire (v.c. 34), Monmouthshire (v.c. 35) and Herefordshire (v.c. 36), England. It has been recorded from Dennis Grove (4 plants), Seven Sisters Rocks and King Arthur’s Cave (4, possibly 6, plants), Far Hearkening Rock (1 plant), on the cliffs above the west side of Symonds Yat (at least 9 plants), the Yat Rock area (at least 3 plants) and at Coldwell Rocks (1 plant). It occurs on Carboniferous Limestone cliffs and cliff edges in open, deciduous woodland with Fagus sylvatica, Fraxinus excelsior and many other Sorbus taxa. The altitudinal range is c. 75 to US) Ta, Only 23 trees of this species are currently known (Houston et al. 2003, with updates), but there are probably more on the extensive cliffs in this area. The provisional IUCN (2001) conservation status 1s “Critically Endangered’. SURVEYS The results of the surveys are summarized in Table 1. Collecting vouchers of the rare taxa, or those difficult to identify in the field, from parts of the cliffs not previously surveyed has resulted is a much more detailed knowledge about the distribution and frequency of each taxon, and allowed recognition of new taxa. The number of sites for each species, and number of species for each site, are still dependent on the quality and detail of the survey which is inconsistent between sites, but these are the best available data which are unlikely to be bettered in the foreseeable future. The commonest and most widespread species are the diploids S. aria and S. torminalis, the former often the commonest species present. The diploid S. aucuparia is much rarer on the limestone cliffs and thus infrequent in our surveys, but is common in the associated woodlands on sandstone (Peterken 2008). Their hybrids S. x thuringiaca (Ilse) Fritsch (=S. aria x S. aucuparia) and S. xX tomentella Gand. (=S. vagensis Wilm., S. aria x S. torminalis) are both scattered through the valley, the latter locally frequent in the Symonds Yat area where it backcrosses with S. aria (Price & Rich 2007). The next most widespread taxon is S. eminentiformis which is most abundant at the northern end of the Lower Wye Valley, but also occurs more locally at the southern end. Sorbus eminens sensu stricto is also scattered through the valley but is rare at the northern end. Sorbus porrigentiformis sensu stricto 1s quite rare with only small numbers of plants, whilst plants attributed to S. porrigentiformis sensu lato are more frequent and also more widespread. Sorbus anglica Hedl. is quite widespread, though it has not been recorded recently in two sites and has declined at another, and S. rupicola (Syme) Held. and S. whiteana T. C. G. Rich & L. Houston occur in a few sites. Sorbus parviloba and S. saxicola are both restricted to the northern end of the Lower Wye Valley as described above. The alien S. intermedia has only been recorded in one site, not too far from Tutshill where it grows as a Street tree. The sites vary in diversity, which is only partly related to their size (Table 1). The two richest sites are site 8 Bowler’s Hole — Symonds Yat, and site 9 Symonds Yat Rock — Coldwell Rocks (Fig. 1) which have extensive cliffs with good populations. Site 20 The Biblins — Symonds Yat West immediately west of this area on the opposite north side of the River Wye is also very rich with ten taxa, though the whitebeams are more scattered. Site 16 Lady Park Wood and site 19 Sevens Sister’s Rocks are very diverse for small areas. At the southern end of the Lower Wye Valley, site 4 Wintour’s Leap, site 5 Ban-y-gor Rocks and site 7 Shorn Cliff are diverse with ten taxa each. Not all cliffs are rich in species, for example Rosemary Topping and Black Cliff are poor, perhaps due to limited outcrops, tall woodland or quarrying. Both the northern area around Symonds Yat and the Doward, and the southern end from Chepstow to Shorn Cliff, have 14 Sorbus taxa recorded. DISCUSSION It is inevitable given the large areas of the Lower Wye Valley investigated, the scale of the cliffs and the difficulty of access to the cliffs with or without ropes, that the surveys are incomplete. It is also often difficult to see 32 T. C. G. RICH, L. HOUSTON, C. A. CHARLES AND A. C. TILLOTSON the trees for the wood. Nonetheless 16 Sorbus taxa have been recorded, of which three are endemic to the Lower Wye Valley (S. eminenti- formis, S. parviloba and S. saxicola) and three endemic to Britain (S. eminens sensu stricto, S. porrigentiformis and S. whiteana). Both the northern area around Symonds Yat and the Doward, and the southern end from Chepstow to Shorn Cliff have 14 taxa recorded. Overall, the Lower Wye Valley is the second most diverse site for Sorbus in the British Isles, second only to the Avon Gorge which hosts about 22 taxa. If hybrids and alien species are excluded from the comparison (the former are much better studied in the Avon Gorge), both the Wye Valley and Avon Gorge support 12 species. The key factor for, the diversity” is\ the presence of extensive areas of open or partly shaded, limestone rocks and cliffs with little disturbance or grazing. These open rocks provide ideal habitat for members of the S. aria group such as S. rupicola and S. porrigenti- formis which are small, light-demanding trees and shrubs. Historically the woodlands were managed to the cliff edges by coppicing as high forest, as shown in old photographs (Helme 1989; Rainsbury 1989; Peterken 2008), and this may have provided conditions for colonization along the cliff edges. Other areas were man- aged as wood pasture, the grazing probably limiting the regeneration except on the cliffs. Plantations rarely affected the cliffs directly. In some places these cliff edges are now overgrown, forcing normally small trees such as S. anglica to grow to 10 m or more tall to keep their leaves in sunlight. As these edges become shaded by trees once more, the whitebeams may become more restricted to the Chitfi tacesy “ihe? exception mitonsthisemismes: eminentiformis, which grows readily in closed woodland as well as on open rocks; for example, there are extensive colonies on the east side of the Great Doward in the wood around the old mines (not surveyed here). There are only minor threats to maintenance of the Sorbus diversity. Most sites are protected by legislation as Sites of Special Scientific Interest or Special Areas for Conservation, and specific conservation management is neither re- quired nor practical in most sites. At Symonds Yat, whitebeams are regularly cut back to maintain scenic views, but the trees regrow from the base. At least two whitebeam trees have been damaged and lost in an area heavily used by outdoor educational groups at Symonds Yat. In general there is little evidence from historical records for anything other than a minor decline in diversity of sites or species. Some losses have occurred at Pen Moel and Wintour’s Leap, probably due to the extensive quarrying. Losses at the Wynd Cliff may be due to increased shading by tall woodland. The failure to refind S. eminens on Offa’s Dyke could be due to forestry operations replacing broad-leaved woodland with conifers, but the location of the type tree is not known. ACKNOWLEDGMENTS The work was funded by English Nature (now Natural England) and the National Museum of Wales. We would like to thank Trevor Evans, Clare and Mark Kitchen, Philip Nethercott, David Price and Ashley Robertson, and land owners for access permission. Philip Oswald kindly provided the Latin translations. REFERENCES AAS, G., MAIER, J., BALTISBERGER, M. & METZGER, S. (1994). Morphology, isozyme variation, cytology and reproduction of hybrids between Sorbus aria (L.) Crantz and S$. torminalis (L.) Crantz. Botanica Helvetica 104: 195-214. CHESTER, M., COWAN, R. S., FAY, M. F. & RICH, T. C. G. (2007). Parentage of endemic Sorbus L. (Rosaceae) species in the British Isles — evidence from plastid DNA. Botanical Journal of the Linnean Society 154: 291-304. EVANS, T. G. (2007). Flora of Monmouthshire. Chepstow Society, Chepstow. HELME, A. (1989). Monmouth and the River Wye in old photographs from Goodrich to Brockweir: from the collections of Monmouth Museum. Alan Sutton Publishing, Gloucester. HOUSTON, L., TILLOTSON, A. & CHARLES, C. (2001). Wye Valley Sorbus project. 1. Ban-y-gor rocks Autumn 2000. Unpublished contract report to English Nature, Ledbury, March 2001. HousTON, L., TILLOTSON, A. & CHARLES, C. (2002). Wye Valley Sorbus project. 2. Shorn Cliff Autumn 2001. Unpublished contract report to English Nature, Ledbury, March 2002. HOousSTON, L., TILLOTSON, A. & CHARLES, C. (2003). Wye Valley Sorbus project. 3. Upper Wye Gorge Autumn 2002. Unpublished contract report to English Nature, Ledbury, March 2003. SORBUS IN THE LOWER WYE VALLEY 3113) HOUSTON, L., TILLOTSON, A. & CHARLES, C. (2004). Wye Valley Sorbus project. 4. Wintour’s Leap and Lady Park Wood Autumn 2003. Unpublished contract report to English Nature, Ledbury, March 2004. IUCN (2001). IUCN Red List Categories. Version 3.1. The World Conservation Union, Gland. PETERKEN, G. (2008). Wye Valley, New Naturalist Series 105. Harper Collins, London. Price, D. T. & RICH, T. C. G. (2007). One-way introgressive hybridization between Sorbus aria and S. torminalis (Rosaceae) in southern Britain. Watsonia 26: 419-432. Proctor, M. C. F. & GROENHOF, A. C. (1992). Peroxidase isoenzyme and morphological variation in Sorbus L. in South Wales and adjacent areas, with particular reference to S. porrigentiformis E. F. Warb. Watsonia 19: 21-37. RAINSBURY, A. (1989). Chepstow and the River Wye in old photographs from Severn to Tintern: from the collections of Chepstow Museum. Alan Sutton Publishing, Gloucester. RicH, T. C. G. & PRocToR, M. C. F. (2009). Some new British and Irish Sorbus L. taxa (Rosaceae). Watsonia 27: 207-216. RICH, T. C. G., HACK, V. & MCMEECHAN, F. (2005). Vascular plants, pp. 307-321 in HILL, D., FASHAM, M., TUCKER, G., SHEWRY, M. & SHAW, P., eds. (2005). Handbook of Biodiversity Methods. Survey, evaluation and monitoring. Cambridge University Press, Cambridge. RIDDELSDELL, H. J.. HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. T. Buncle & Co., Arbroath. ROBERTSON, A., RICH, T. C. G., HOUSTON, L., ALLEN, S., ROBERTS, C., HARRIS, S. A. & HISCOCK, S. J. (2009). Hybridization and polyploidy as drivers of continuing evolution and speciation in Sorbus (Rosaceae). (in preparation). ROYAL HORTICULTURAL SOCIETY (1966). R. H. S. colour chart. Royal Horticultural Society, London. SECRETARIAT FOR THE CONSERVATION OF BIODIVERSITY (2002). Global strategy for plant conservation. Secretariat for the Conservation of Biodiversity, Montreal. WADE, A. E. (1970). The flora of Monmouthshire. National Museum of Wales, Cardiff. WARBURG, E. F. (1952). Sorbus L., pp. 539-556. in CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. (1952). Flora of the British Isles. Cambridge University Press, Cambridge. WARBURG, E. F. (1957). Some new names in the British flora. Sorbus. Watsonia 4: 43—46. (Accepted April 2009) Watsonia 27: 315-321 (2009) BS The decline of Lady’s-mantles (Alchemilla vulgaris L. agg.) and other hay-meadow species in Northern England since the 1950s MARGARET E. BRADSHAW ‘Lady’s Mantle’, Hill Top, Eggleston, Barnard Castle, Co. Durham, DL12 OAU ABSTRACT In Teesdale (v.cc. 65, 66 & 70) and Weardale (v.c. 66), a considerable decline in the status of five species of the Alchemilla vulgaris L. aggregate, together with three other species of herbs character- istic of northern England hay-meadows, has been revealed by repeat surveys of meadows, and road and rail verges first visited between 1950 and 1967. As a consequence, these five Alchemilla species and the three herbs should be considered threatened and in urgent need of conservation action. KEYWORDS: Teesdale, meadows, conservation, agricultural changes, grassland management. INTRODUCTION Max Walters (Walters 1949), in his paper ‘Alchemilla vulgaris L. agg. in Britain’, clarified the identity of the various British species within the aggregate, thereby establish- ing some order out of the chaos created during the last century by the Swiss botanist, Firmin Jaquet. In 1951, Walters suggested that the author might like to look for them in Teesdale (v.cc. 65, 66 & 70) where he had seen several interesting species himself. It soon became clear that in Northern England there were eleven native agamospecies of the Alchemilla vulgaris aggregate (Lady’s-mantles) and nine of these were subsequently found in the north Pennine valleys of Teesdale and Weardale (lat. 54°40’ N, long. 2° 15’ W). It is now known that three: A. acutiloba Opiz (‘Starry Lady’s- mantle), A. monticola Opiz (Velvet Lady’s- mantle) and A. subcrenata Buser (Large- toothed Lady’s-mantle) are almost restricted to these two dales and that two more — A. glomerulans Buser (Clustered Lady’ s-mantle) and A. wichurae (Buser) Stefansson (Rock Lady’s-mantle) are here at the southern fringe 'See APPENDIX p. 321 for Vernacular names of their distribution and known to be centred in Scotland (Preston et al. 2002). The North Pennines is considered to contain approximately 40% of the surviving upland herb-rich hay meadows (National Vegetation Classification MG3 Anthoxanthum odoratum- Geranium sylvaticum grassland) in the U.K. These grasslands are characterised by boreal species such as Trollius europaeus L. (Globeflower), Geranium sylvaticum L. (Wood Crane’s-bill) and Cirsium heterophyllum (L.) Hill (Melancholy Thistle) as well as more typical grassland species such as Alchemilla spp., Rhinanthus minor L. (Yellow-rattle), Conopodium majus (Gouan) Loret (Pignut), Sanguisorba officinalis L. (Great Burnet) and Leontodon hispidus L. (Rough Hawkbit). It was fortuitous that the survey of the Alchemilla spp. took place in the early 1950s, shortly after the Second World War and about the time that farming in the dales was becoming mechanised and _ with artificial fertilisers readily available. In 1967 the author directed a survey of three other characteristic northern hay-meadow species, namely Trollius europaeus, Geranium sylvaticum and Cirsium heterophyllum in Teesdale. By this time, restriction to field margins and riversides of species such as Trollius europaeus indicated that a change in the diversity of the meadow flora had already begun. On the author’s return to Teesdale, a survey was carried out in the year 2000 to determine the status of these Alchemilla species after a fifty-year period, whilst in 2003 and 2004 the Upper Teesdale Botany Group searched the original fields for the three meadow species. This paper now describes the results of these two surveys, the causes of the observed declines, and the prospects for the future conservation of northern hay-meadow species. 316 M. E. BRADSHAW METHOD The five species of the Alchemilla vulgaris aggregate originally surveyed were the northern- montane meadow species: A. acutiloba, A. monticola and A. subcrenata together with the arctic-alpine species A. glomerulans and the arctic-subarctic species A. wichurae. The fine- scale distributions of these five species were mapped in the early 1950s as ‘dots’ on the Ordnance Survey 1: 25,000 maps, with a ‘dot’ representing one or more plants (Bradshaw 1962). Where a species was abundant the whole field was coloured-in on the maps and any contiguous presence on road verges was marked as a solid line. In the year 2000, approximately 80% of these sites were re- visited and the findings marked on similar maps and in a similar way. Since then, many of the remaining 20% of sites have also been inspected. A summary of these results is given in Table 1. In 1967 the three hay-meadow species, Trollius europaeus, Geranium sylvaticum and Cirsium heterophyllum, were surveyed in the hay-fields, roadsides, and river margins in two main areas: (i) the Newbiggin-in-Teesdale and Holwick area (altitude c. 260 m) to the west of Middleton-in-Teesdale, and (11) Forest-in-Teesdale (altitude c. 380 m) to the west of the High Force Hotel (Bradshaw 2001). These sites were then re- surveyed in 2003 and 2004 (Bradshaw 2006). RESULTS ALCHEMILLA SPECIES In the fifty-five year period since the first survey, the status of each of the five species of the A. vulgaris aggregate was found to have changed from being apparently stable (although in some species rare) to endangered, even in the most abundant species, A. monticola and A. acutiloba. A detailed assessment of each species is given below, see also Table 1. Alchemilla subcrenata A. subcrenata is the scarcest of the four Northern Montane Lady’ s-mantles in Britain. It was first collected in 1951 by the author near Newbiggin-in-Teesdale and _ identified by Walters (1952). Its main area of distribution is restricted to the Newbiggin-Holwick area but in 2007/2008 a few plants were also found in each of three meadows near Allendale, in Northumberland (v.c. 67), a considerable distance from its centre in Teesdale. Since the first survey, the number of fields in which A. subcrenata has been recorded in Teesdale has fallen from 22 to 10 (49 to 25 dots) and now there is only a single road verge site. Most populations are confined to meadows where it 1s a scarce species and to two pastures. Plants found in the 1950s in two fields in Weardale, north of Stanhope, were not re-found in 2000 nor in 2002, both fields having been agriculturally ‘improved’. TABLE 1. CHANGES IN THE FREQUENCY OF THE ‘DOT’ RECORDS FOR FIVE SPEGIES OF THE ALCHEMILLA VULGARIS AGGREGATE IN TEESDALE AND WEARDALE MADE IN THE 1950S AND BETWEEN 2000 AND 2007 ; me ‘Dots’ visited ‘Dots’ found Species Area Dots’ in 1950s 30100-2007 3000-2007 % lost Comment A. acutiloba Teesdale 27 20 16 20 Weardale 195% 164* 82 50 plus 10 fields* A. monticola Teesdale 498* 380* 56 85 plus 33 fields* Weardale 36 36 l 97 A. subcrenata Teesdale 49 49 25 49 Weardale 4 4 0 100 A. glomerulans Teesdale 81 74 14 81 Weardale 2 D | 50 A. wichurae Teesdale 34 Di 15 44 * In the 2000-2007 survey neither species was found to be as frequent in any of the fields as they were in the original survey, nor were either of them frequent along contiguous lengths of road-verge. Finds in these places were recorded as ‘dots’. ALCHEMILLA DECLINE Sil7I Only one site, a pasture, occurs within the perimeter of the Moor House-Upper Teesdale National Nature Reserve although several populations in fields on the south side of the River Tees are under ‘agri-environment’ agree- ments. On the north side of the Tees, the field where the species was discovered is not currently under any conservation management scheme, another field has a_non-statutory agreement between the landlord and the tenant, whilst a third is in the Higher Level Steward- ship Scheme (Natural England). A. subcrenata is classified as ‘endangered’ in the new Red List (Table 3) (Cheffings & Farrell 2005). The current population probably comprises no more than 300 plants. Alchemilla acutiloba and A. monticola A. acutiloba has been known from Teesdale since 1946, but was not collected in Weardale until 1951 when the author found a dense colony on disturbed ground near the new Burnhope Reservoir dam. In the 1950s it was found to be very frequent in meadows and on road- and rail-verges in Weardale but not as common in similar habitats in Teesdale. It has also been recorded from a scattering of sites, mostly on roadsides, to the east and north of the area. It was only in 2006 that it was found to occur south of the Tees with the location re- confirmed in 2008. A. monticola was first recorded in Teesdale in 1922. Walters (1949) found it to be frequent in meadows and on roadsides in 1947 and this was also the case in the 1950s. It is rare in upper Weardale where the author found only one colony on a verge in a side valley. Many plants were also found to the east of this site growing around Hamsterley village green on the south side of the valley. Since the 1950s both species have become much scarcer in both dales. In Teesdale the number of meadow populations of A. acutiloba has fallen by 20% from 20 visited-sites (of the 27 originally identified) to 16; in Weardale the decline has been much greater (50%) from 164 (of the 195 originally identified) to 82. In Teesdale the number of A. monticola popula- tions has fallen by 80% from 330 (of the 448 originally identified) to 63 and in Weardale all the 35 ‘records’ at Hamsterley have been destroyed due to the ‘tidying-up’ of the village green. However, the very small colony of roadside plants in the upper Dale was still present in 2008. There have been even greater losses of both species from road- and rail-side habitats and from certain meadows. No longer are there long stretches of verge where one or both species are frequent as they were in the 1950s°. A. acutiloba has been lost from a 2:9 km length of verge in Teesdale and from a 2-1 km length in Weardale, whilst A. monticola has dis- appeared from a 4°8 km length of verge in Teesdale; this is a total length of 9-7 km (..e. almost 6 miles). Of the thirty-three small fields’ near Middleton where A. monticola was formerly very frequent, about half of them are now built on with houses and a school whilst, in the remainder, the species have been reduced to individual ‘dots’. Similarly, A. acutiloba is now scarce in the fields in Weardale where it was formerly very frequent. Alchemilla glomerulans and A. wichurae Alchemilla glomerulans and A. wichurae are respectively arctic-alpine and arctic-subarctic species which, in the UK, occur mainly above 615 m in Scotland. Populations of A. wichurae in northern England are at the southern limit of their UK distribution and also of their world distribution (Kurtto et al. 2007) and are unusual in occurring at the relatively low altitude of 250 m. Walters (1949) discovered A. glom- erulans when visiting Teesdale in 1947 and it is now known to occur in meadows, meadow- type vegetation near the Tees, and in well- drained and flushed hill grasslands extending up to 676 m on the Pennine escarpment in v.c. 69 (Westmorland). It has also been recorded on a road verge and a bank in a meadow in v.c. 70 (Cumberland). In Weardale it was discovered by the author in the 1950s growing in a single meadow and in a nearby railway cutting. A. wichurae was collected by Walters in 1946 on Ingleborough (v.c. 64) and in Teesdale in 1947 (v.cc. 66 & 69). It has been found to be very thinly scattered, usually occurring as single plants, in a few meadows, upland pastures, and on limestone outcrops. Several records occur by waterfalls over the whinsill (dolerite) rock. In Teesdale, by 2007, the number of A. glomerulans populations has declined by 82% from 74 (of the original 81) ‘dots’ in the 1950s to 14 (ncluding 4 recent finds). Some of the losses have been from meadows; one of these was due to ploughing but the majority have been of isolated single plants in close-grazed hill pastures. In Weardale it has been lost from its meadow site leaving just one plant by 2007. 318 M. E. BRADSHAW In Teesdale, the number of ‘dots’ for A. wichurae has declined by 44% from 27 (of the original 34) to 15, most losses having been from meadow sites. In hill pastures, A. wichurae has survived better than A. glomerulans but is still vulnerable to extinction where only single plants remain. Nowadays, the only concen- tration of plants of A. glomerulans is in four meadows near Middleton-in-Teesdale. SPECIES CHARACTERISTIC OF NORTHERN HAY-MEADOW The survey of three hay-meadow species, Cirsium heterophyllum, Geranium sylvaticum and Trollius europaeus, has revealed a decrease in the numbers of fields in which each species can now be found (the slight increase of Geranium sylvaticum in Forest-in-Teesdale is very likely due to the more thorough search made in 2004) (Table 2). In the Newbiggin- Holwick area, Trollius europaeus is now almost exclusively confined to the banks of the Tees. Geranium sylvaticum is the only species that is widely dispersed throughout any individual fields and is most frequent in Forest- in-Teesdale. It is clear that these characteristic and conspicuous species are less common than they once were and that there has been a general decrease in the species diversity of the hay meadows. This is supported by an observable change in general colour of most dales meadows which at one time were mauve, white and shades of yellow to nowadays being only buttercup-yellow, white and brown. In 2006-2007, John O’Reilly searching for herb- rich meadows in the North Pennines Area of Outstanding Natural Beauty found that only 8% of the 388 fields examined could be classified as species-rich and that even this figure might be an _ over-estimate when considering the total number of meadows in existence (pers. comm., J. O’ Reilly, 2008). CAUSES OF DECLINES The decline of the five species of Alchemilla and the other meadow species probably began in the 1960s and, based on the author’s experience, most had occurred by the 1980s. The primary causes in meadow habitats have been due to changes in farm management and animal stocking. Heavy continental-cross suckler cows have replaced the dairy cows and larger numbers of sheep are being ‘wintered’ in the meadows. Artificial fertilisers have supplemented farmyard manure, which itself has changed from a sweet-smelling dung to an acrid, ammonia-laden, semi-raw compost created from bought-in straw used for bedding; this most probably has had a significant effect on the soil micro-flora and fauna as well. At the same time, there has been a decrease in the application of lime and/or basic-slag. Some hill farms stocked with fell-sheep continue to make small bales of hay, but large-bale silage and haylage has generally become the norm and a surer way of conserving the grass for winter fee@s Although during the second half of the last century, a few fields had been ploughed and re- seeded or slot-seeded with more productive rye-grass/clover (Lolium/Trifolium) mixtures, the recent relaxation of Environmental Impact Assessment (Agriculture) (England) Regu- lations 2006 has resulted in several herb-rich meadows being destroyed. The sum of all these developments has been a marked loss in species-diversity formerly evident in herb-rich meadows. Road verge habitats have also been affected. There has been a great increase in the amount and weight of motorised traffic which has necessitated strengthening the road-edges by the addition of imported ballast. As a result, seeds and parts of weedy plants have accidentally been introduced onto the verges and become established. Eutrophication, TABLE 2. NUMBER OF FIELDS CONTAINING THE THREE MEADOW SPECIES IN THE SURVEYS OF 1967 AND 2003-2004 Nos. fields Nos. fields with Nos. of fields with Nos. of fields with resurveyed Trollius europaeus — Geranium sylvaticum — Cirsium heterophyllum in 2003/2004 1967 2003/2004 1967 2003/2004 1967 2003/2004 Newbiggin / Holwick 80 26 oy 44 yp 15 Forest in Teesdale Sy) 36 28 36 26 Zs ee ee ee ee ALCHEMILLA DECLINE 319 TABLE 3. CONSERVATION STATUS OF THE RARE ALCHEMILLA VULGARIS AGGREGATE SPECIES IN TEESDALE Species (1) Red List GB (2) Scarce (3) DRPR (4) BAP Alchemilla acutiloba VU A2c RD V + Alchemilla monticola EN A2c RD E + Alchemilla subcrenata EN A2c Blab(v) + 2ab(v) C2a(I) 4 <800 RD E + individuals Alchemilla glomerulans* VU A2c E Alchemilla wichurae* EN A2c SPIB E (1) Vascular Plant Red Data List for Great Britain. (Cheffings & Farrell 2005). (2) Scarce Plants in Britain. (Stewart et al. 1994). (3) Durham Rare Plant Register (Durkin, unpublished) (pers. comm. 2008). (4) Biodiversity Action Plan UK BAP List 2007 * Natural England Common Standards Monitoring of Rare Plant Species, Local Distinctiveness Attributes (LDA), e.g. scarce in England but frequent in Scotland or at edge of the species range. caused by exhaust fumes, spray and salt from the roads, and seepage of nutrients from adjacent fields, have also had an adverse effect. Successive changes in road verge management have involved the use of herbicides; mowing by County Council machines with wide flails which create a heavy mulch has favoured the stronger, coarser grasses, broad-leaved plants and imported weeds. In addition, re-alignment of, and other work on, many minor roads has added to the toll. Collectively these changes have decimated the colourful meadow plant community. Pacha (2004) found a similar decline over the last two decades in the distribution of Geranium sylvaticum in the meadows within the Yorkshire Dales National Park. This species has disappeared from 40% of meadows where it was known to exist in the 1980s with surviving populations having become increas- ingly isolated over the survey period. Also, even fields managed under agri-environment schemes have become more impoverished with a general decrease in diversity and richness, especially among species characteristic of MG3 Anthoxanthum-Geranium meadows. The survey results also highlighted the increasingly impor- tant role that road verges and river margins play for the survival of northern hay-meadow species such as G. sylvaticum, especially in areas where the species has disappeared from the adjacent meadow habitats. CONSERVATION All five species of the Alchemilla vulgaris aggregate mentioned in this paper are classified as ‘endangered’ or ‘vulnerable’ in the new Red List for Great Britain (Cheffings & Farrell 2005), and the three meadow species (A. acutiloba, A. monticola, A. subcrenata) are included in the new Biodiversity Action Plan (BAP) List of priority species (Table 3). At the time of writing, in 2008, all of these species are very vulnerable to further reductions in their numbers. This is particularly so because of the uncertainties in the future of farming and of the management of the upland meadows and fells for food production, game birds and recreation. Already, more sheep as well as cattle are housed in winter on imported straw, thus producing larger quantities of straw-based manure (see earlier) which will have unknown effects on the meadow biota. Other changes, for example, might include the reduction in the numbers of cattle and hence their availability for the desirable autumn grazing of the meadows; also a possible increase in the number of sheep, including breeds with different fodder requirements, followed by an increase in winter grazing of the meadows; these would all have an effect. On the fells, the reduction in the number of sheep that has already begun might prove beneficial in the short-term by allowing heavily grazed plants to flower (and so be identified) at least in those areas where rabbits are absent or controlled. In the longer term, under-grazing could produce a different threat, especially to light-demanding plants. Of the five Alchemilla species, A. wichurae would be the most likely to be adversely affected by this. Finally, it should be remembered that global warming is predicted to affect many more species than the Alchemilla species and meadow plants covered in this paper. 320 M. E. BRADSHAW ACKNOWLEDGMENTS Most importantly, I thank the late Max Walters for introducing me to the genus Alchemilla and encouraging and supporting my _ work thereafter. Thanks are due to the Raby Estates, Staindrop and Strathmore Estates, Barnard Castle, and to all the farmers who allowed access to their fields, both to the author and to members of the Upper Teesdale Botany Group, the latter whom [ also thank. Also I am grateful to the referee for making many helpful suggestions and to Kevin Walker for his comments and contributions to the manuscript. REFERENCES BRADSHAW, M. E. (1962). The distribution and status of five species of the Alchemilla vulgaris L aggregate in Upper Teesdale. Journal of Ecology. 50: 681-707. BRADSHAW, M. E. (2001). A survey of three meadow plants in Teesdale, Teesdale Record Society Journal 3rd series 9: 10-16. Barnard Castle. BRADSHAW, M. E. (2006). The survey of three meadow plants in Teesdale revisited, Teesdale Record Society Journal 3rd series 14: 25-30. Barnard Castle. CHEFFINGS, C. M. & FARRELL, L. (eds) (2005). The Vascular Plant Red Data list for Great Britain. J.N.C.C., Peterborough. DEFRA (2006). Environmental Impact Assessment (Agriculture) (England) Regulations. DEFRA (2007). The UK Biodiversity Action Plan (UKBAP). KURTTO, A., FROHNER, S. E. & LAMPINEN, R. (eds.) (2007). Atlas Florae Europaeae. 14: Rosaceae (Alchemilla and Aphanes). The Committee for Mapping the Flora of Europe & Societas Biologica Fennica Vanamo. Helsinki. PACHA, M. J. (2004). Fragmentation of northern hay meadows and populations of Geranium sylvaticum in the Yorkshire Dales National Park. Unpublished Ph. D. thesis, University of Lancaster. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. (2002). Atlas of the British and Trish Flora. Oxford University Press, Oxford. STACE, C. A. (1997). New Flora of the British Isles. 2nd ed. Cambridge University Press, Cambridge. STEWART, A., PEARMAN, D. A. & PRESTON, C. D. (1994). Scarce Plants in Britain. J.N.C.C., Peterborough. WALTERS, S. M. (1949). Alchemilla vulgaris L. agg. in Britain. Watsonia 1: 6-18. WALTERS, S. M. (1952). Alchemilla subcrenata Buser in Britain. Watsonia 2: 277-278. (Accepted January 2009) ALCHEMILLA DECLINE a2 APPENDIX VERNACULAR NAMES FOR ALCHEMILLA VULGARIS S.L. SPECIES The following vernacular names are proposed for the agamospecies of the Alchemilla vulgaris aggregate found in Britain. In non-scientific papers it is recommended that the English name be followed by the Latin in brackets. A. glabra A. acutiloba A. micans A. filicaulis ssp filicaulis A. filicaulis ssp. vestita A. wichurae . glaucescens . xanthochlora A A A. subcrenata A. glomerulans A . monticola minima alpina . conjuncta . tytthantha mollis SSS Sf venosa Smooth Starry Shining Slender Hairy Rock Silky Pale Large-toothed Clustered Velvet Least Crimean Soft Crisp Lady’s-mantle Lady’ s-mantle Lady’ s-mantle Lady’ s-mantle Lady’ s-mantle Lady’ s-mantle Lady’s-mantle Lady’ s-mantle Lady’ s-mantle Lady’ s-mantle Lady’ s-mantle Lady’s-mantle Lady’ s-mantle Lady’ s-mantle Lady’s-mantle (belongs to the ‘Sub-glabrae’ and can have hairs and so belie the name glabra and confuse learner botanists) as used in Scandinavia as used in Scandinavia, the species frequently grows on rocky outcrops in the UK I frequently describe the upper surface of the leaf as looking like ‘velvet’— dense, short erect hairs Alpine Lady’ s-mantle is already in use Silver Lady’ s-mantle is already in Stace (1997) nthe Watsonia 27: 323-338 (2009) S28 Studies in the floristic diversity of Durham walls, 1958-2008 D. W. SHIMWELL* High House Farm, High Stoop, Satley, Bishop Auckland DL13 4HL ABSTRACT A study of the flora of the walls of Durham City over the period 2006—2008 indicated that the floristic diversity had increased by between 39% and 42% since a survey undertaken fifty years ago by Woodell & Rossiter (1959). In addition, the increased frequency of occurrence of twenty-four of the thirty- four key species from the first survey is apparent and there is group of seven species not recorded in the first survey that has become especially prevalent. The increases may be interpreted in terms of the amelioration of the atmospheric environment which has offset any potential reduction of diversity associated with gentrification of the city. An influx of native species has been mainly responsible for the rise in diversity, though the number of neophytes has increased by almost a third of the original total. The floristic diversity of the urban wall habitat is higher than that of rural settlements of the Durham Dales, emphasising the importance of the former as a refuge for many species which have been perceived as declining at the national scale. The walls of each rural settlement have a distinct floristic signature, but the main chasmophyte communities show close phytosociological affinities with the alliance Cymbalarion-Asplenion. KEYWORDS: floristic diversity, walls, urban, rural, Durham City, Durham Dales. INTRODUCTION This paper takes its stimulus from an article by Woodell & Rossiter (1959) published in the Proceedings of the Botanical Society of the British Isles which reported the findings of a survey of the flora of the walls of Durham City in the period 1955-1958. Their study was based upon a survey of Cambridge walls by Risbeth (1948) and in turn, stimulated extensive research on the flora of the urban walls of Middlesex (Kent 1961), and more specific projects such as surveys of the churchyard walls in Middlesex (Kent 1964) and Norfolk (Silverwood 1965). Woodell continued his interest in urban habitats and became recognised as a leading authority in the genre (Woodell 1979). In the meantime, Segal (1969) published his classic volume on the ecology of *e-mail: shimwell44 @tiscali.co.uk European wall vegetation and this encouraged rural research projects such as those of Holland (1972) on the old walls of western Ireland, and Payne (1978) in south-eastern Essex. As interest in urban ecology increased, so did an appreciation of the importance of walls as modified cliff-like habitats with a special group of ecological characteristics. The popular text by Darlington (1981) entitled Ecology of walls brought the value of the habitat to public attention and the subject area was admirably developed by Gilbert (1989, 1996), through to the work of Larsen eft al. (2000). Related studies of ruderality in a variety of urban habitats (Crowe 1979; Kent et al. 1999; Hill et al. 2002; Lundholm & Marlin 2006, inter alia) have all shown the value of walls in terms of microhabitat preference of certain species in the context of an environment which is subject to frequent disturbance. Other studies have placed a greater emphasis on the cryptogamic flora of walls, with examples ranging from a transect across County Durham by Wright (1984) to the recent Flora of Dry Stone Walls conservation project (Presland 2007, 2008a, b). As a result of various ecological investigations, the recognition of the ecological value of the built environment and walls has led to their inclusion in several Local Biodiversity Action Plans (vide those of Surrey County Council (2001) and Hull (2002), for example). The history of the construction of the defences of the key military and ecclesiastical site of Durham, on its peninsula formed by the River Wear, is well known (Page 1928). Probable Anglo-Saxon walls surrounding the cathedral and castle were rebuilt by Bishop Flambard (1099-1128) and the defensive walls extended to encompass most of the peninsula in 1173-1174. In 1315 and again in 1337, the townsfolk petitioned the King for murage grants to construct walls to protect the market place, the focal trading point of the Bishop’s Borough. The key aspect of these early details is that the wall habitat may be viewed as a relatively ancient artificial ecosystem and that, although many of the defences from the early historical periods do not survive, their 324 D. W. SHIMWELL continual alteration and augmentation have provided a constant sequence of mural habitats for populations of species to colonise. The greater proportion of the walls in the city, the heirs of this long history of urban evolution, are of eighteenth and nineteenth century origin and this fact is implicit in the statement by Woodell & Rossiter (1959) that ‘factors con- tributing to the high number of species present are the great age of many of the buildings and walls in the city, and the present state of neglect of many of them.’ In the half century since this statement, there have been many aspects of urban development which would appear to have caused subtle ecological changes to the city environment and thus, the flora of Durham walls. The full effects of the Clean Air Act 1956 would not have been realised at the time of the 1955-1958 survey and atmospheric conditions have been further ameliorated by the Smoke Control Area provisions of the 1968 Act, reinforced by the Clean Air Act 1993. The Rivers (Prevention of Pollution) Act 1951 was in force and its repeal by the Control of Pollution Act 1974, no doubt contributed to increases in the floristic diversity of the riverside retaining walls. To a certain extent, these increases may have been negated by the construction of new flood defences of a structure less amenable to plant colonisation. Further, the perceived general gentrification of the city in association with the designation of the Cathedral and Castle environs as a World Heritage Site in 1986 (extended 2008) would seem to suggest a means for the potential rectification of the neglect described by Woodell & Rossiter. The basic hypothesis for this resurvey of the flora of Durham walls fifty years later, is that the combination of a variety of general environmental change and urban developmental processes have had a cumu- lative effect to reduce the floristic diversity of this particular habitat. SURVEY AND METHODOLOGY The methods of the investigation undertaken by Woodell & Rossiter were difficult to replicate based on the rather minimal information published in their 1959 paper. They ‘listed the flora of 66 walls from Durham City and its environs’ on ‘a series of visits over the whole year’. The walls were mainly of the simple type, but the survey also included bridges, old buildings and stone buttresses. The com- position and aspect of each wall and _ its function was recorded, but such data not included in the paper. Details are also lacking on the criteria by which the sample walls were chosen or what was the size of the sample. Some enlightenment was provided to the author by Dr Stan Woodell through personal communications in 1968, but unfortunately much of the raw data had been discarded at that time. Some original data was lodged with the Durham Colleges Natural History Society, founded by Woodell in 1953, to which the author had access as editor of the society journal in 1967-1968. These sources indicated that the survey was unstructured in terms of geography and that a wall was sampled initially when it had a complement of more than eight species in a fifteen metres length, and there- after, whenever a new or different species not previously listed was noted. The walls chosen fell mainly within the urban core of the city around the castle and cathedral, radiating out to the Victorian terraces to the south of the city, perhaps within ‘a radius of one mile from the castle’. Out-of-town sites at Shincliffe village, Old Durham and Finchale Priory, which fell within the purview of Durham City Council, were included in the survey on account of their age and heritage importance. The lists included only those species rooted in the fabric of the wall and the archive provided records of the occurrence of all the species recorded in four broad groups, i.e. the 14 species with 15 or more records (published in Woodell & Rossiter and identified in this paper as ‘Category 1 species’); the 8 species recorded between 10 and 14 times (‘Category 2 species’); the 21 between 5 and 9 times (‘Category 3 species’); and the 129 species recorded less than 5 times (‘Category 4 species’). The recent research began with a feasibility study in September 2006 on the variety of fabric types and potential differences in floristic complement due to aspect and shade. This study indicated that all the walls were mortared, that the greater proportion was of local sandstone and that little significant diff- erence between the floras of brick as opposed to sandstone fabric could be determined. Similarly, few variations due to either aspect or shade were apparent within the context of the selection of samples by the minimum species number criterion. The preliminary survey also concentrated on determining the general FLORISTIC DIVERSITY OF DURHAM WALLS 325 potential for replication of the survey by Woodell & Rossiter, particularly the iden- tification of potential sites for analysis, with attempts to identify specific sites mentioned in their paper and archive. Direct comparisons with a few sites were possible. For example, their ‘brick kiln, in large meadow near river’ is in fact the grain/root vegetable silo at Kepier Farm (site 66, NZ284434). Other sites could be identified as the only locality for a particular species in both surveys, such as Campanula rotundifolia on the wall opposite the Seven Stars Inn, Shincliffe (67, NZ293406), Milium effusum on Cathedral Banks, Pimlico (20, NZ271419) and Stellaria holostea in Margery Lane, Durham School (4, NZ268420). It soon became apparent that it would be possible to find a similar number of sites, with eight or more species in a minimum required length of fifteen metres, in a similar geographical spread, for analysis through 2007 and 2008. The 68 sites (cf 66 of Woodell & Rossiter) fell unevenly within ten monads_ with a concentration of 29 in NZ 2742, the World Heritage Site and old urban core of the city (see Appendix 3 for details). Each site was visited on three occasions in both 2007 and 2008 during the last fortnight of April, the last week of June/first week of July, and in the first fortnight of September. The total complement of chasmophytes actually rooted in the fabric of the wall was recorded. Mention was also made in Woodell & Rossiter (1959) of the wall flora of towns and villages to the west of the city in Weardale and Teesdale, though with little specificity of locality. As a second part of the present survey, it was decided to attempt an overall com- parison of urban/rural wall floras and feas- ibility studies were undertaken in the period from September 2007 to January 2008. Their aims were twofold. The first sought to examine the potential for the generation of a compare- ative data set from the towns and villages of Weardale (Frosterley & Wolsingham and Stanhope) and Teesdale (Barnard Castle and Middleton-in-Teesdale). Thirty sample sites were selected from four localities, two in each dale, from a total of thirteen monads in relatively close comparison with the similar number in the city (see Appendix 3 for details). The methods used and the frequency of visit during 2008 were the same as for the urban survey. | RESULTS A. INDICATIONS OF URBAN FLORISTIC CHANGE, 1958-2008 After nomenclatural corrections since 1958, the total complement of species recorded in the two surveys in Durham City was 260, though this figure would certainly have been higher had identification of the microspecies of Taraxacum officinale and Rubus fruticosus been attempted. On the advice of Dr A.J. Richards, the former was recorded as Taraxacum Section Ruderalia, since most urban dandelions fall into this group and many may be identified to either 7. exacutum or T. ekmanii. The mature specimens of Rubus fruticosus which were encountered usually keyed out to R. dasyphyllus, but most were too immature to be positively identified. A similar situation applied to many seedlings of Cotoneaster — C. simonsii and the C. x wateri complex — but C. horizontalis was usually distinct. No attempt was made to distinguish taxa and hybrids in Rosa Section Caninae, Subsection Caninae. The full list of species is presented in Appendix | in which three categ- ories are recognised: those species recorded in both surveys; those recorded in 1959, but not seen as chasmophytes in 2006—2008; and those not recorded in 1959 but noted in 2006-2008. Of the 159 species of pteridophyte, gymno- sperm and angiosperm recorded in the 1959 paper, 34 were not noted as chasmophytes in 2006-2008, but all, with the exception of Deschampsia flexuosa, Galium saxatile and Genista rigida, were recorded in a variety of other habitats, notably wall bases and in overhanging garden shrubbery. The 2006—2008 survey recorded 226 species, including 86 chasmophyte species that were not recorded in the original 1959 paper. The primary indication of change from comparison of the two data sets is thus that the overall diversity of the urban wall flora has not been reduced in the intervening fifty years, rather, that it has increased by between 39% and 42% using the total species difference between Woodell & Rossiter and the 226 species of the 2006-2008 survey, and the 260 total recorded species for the two surveys. A second indication of the nature of change may be derived from a closer look at the patterns of distribution of the 226 species in the 326 D. W. SHIMWELL 2006-2008 survey. Basically, the following data indicate that the mural vegetation has become more prominent, better developed and that the likelihood of encountering a species- rich community is far greater in 2008 than in 1958. a. the range of species number was from 12 to 46 with a mean number of 22-4 per sample; thirty-eight samples had 20 or more species; b. 68 species occurred in seven (>10%) or more samples and 29 species in 17 (25%) or more samples; c. 35 species occurred in 15 or more samples compared with 14 of Woodell & Rossiter (eemablen)); d. 85 species occurred in five or more samples compared with 45 of Woodell & Rossiter; e. 137 species occurred in four samples and less, 75 of these in only one sample; the relative numbers from Woodell & Rossiter were 129 and 74; the two surveys had only 72 and 39 species in common respectively in these two categories (see Appendix 2 for details). Reference to Table | provides an insight into the changes in percentage frequency of individual species in the four categories recognised by Woodell & Rossiter. Of the fourteen species in Category | (Table la), only four (Senecio jacobaea, Rubus fruticosus, Taraxacum Section Ruderalia and Epilobium montanum) have increased in frequency. Dactylis glomerata and Senecio vulgaris are the two of the remaining ten species which show the greatest decline. Direct comparisons with species falling into Categories 2, 3 and 4 are not possible due to the lack of detail in the 1959 paper, but taking the maximum possible percentage frequency for each category (1.e. 21. 14 and 6) it is possible to see that none of the species achieving an overall frequency >20% in 2008 has declined. Category 2 is notable for the increase of Hedera helix and Urtica dioica, in Category 3 Cymbalaria muralis, Galium aparine and Geum urbanum stand out, whilst the three fern species Phyllitis scolopendrium, Dryopteris filix-mas and Asplenium ruta- muraria are of interest in Category 4. The final group of seven species in Table le, those not recorded in the earlier survey, are of particular interest. Neophyte species such as Epilobium ciliatum, Centranthus — ruber, Cotoneaster horizontalis and Senecio squalidus would fall into the common perception of invasive weed species, but perhaps not so the other three native species. In this context, an analysis of the status categories of the total complement of species, as defined by Preston, Pearman & Dines (2002) — native, archaeophyte and neophyte — provides some indication of the types of invading species since 1958. This must be viewed with a certain latitude for species such as Clematis vitalba, Euphorbia amyg- daloides and Meconopsis cambrica, inter alia, considered native in a part of their range are clearly of horticultural origin on the city walls. With this reservation in mind, it is interesting to note that the main increase in diversity has been due to the addition of an increment of 54 native species in contrast with a mere ten neophytes. B. ACOMPARISON OF URBAN AND RURAL FLORAS AND THE NATURE OF RURAL SIGNATURES W&R = 2006-2008 Both Total 159 226 260 Native 106 160 173 Archaeophyte 7 20 23) Neophyte 36 46 64 Perhaps the most interesting feature of the Durham Dales data set from 120 samples of rural walls was the lower diversity when compared to that of Durham City, specifically, a total of 162 species as opposed to 226. The Durham Dales data set had 135 species in common with that of Durham City and also a total of 27 species not recorded on the urban walls. The difference may in part be explained by the absence of samples from riverside retaining walls in all rural localities except Barnard Castle. This locality did, however, have the lowest total of the four sub-sets, totals for which ranged between 95 and 107 species. Other differences may be seen in the summary data presented below, particularly in that the range and mean number of species per sample are considerably lower in the rural situation. FLORISTIC DIVERSITY OF DURHAM WALLS SEI TABLE |. SIGNIFICANT CHANGES IN PERCENTAGE FREQUENCY OF KEY SPECIES, 1958-2008 [1957/58 (a) and 2006—2008 (b)] a. Changes in the Percentage Frequency of 1958 Category 1 species (>15 records/>23% ) a n Taraxacum Section Ruderalia 42 Chamerion angustifolium 35 Sambucus nigra Di Dactylis glomerata 26 Poa annua 26 Epilobium montanum US Acer pseudoplatanus 23 Senecio vulgaris 18 Lolium perenne 15 Plantago lanceolata NS Poa pratensis agg. 15 Rubus fruticosus Lis) Rumex obtusifolius 15 Senecio jacobaea 15 b +/-% +15 -3 -9 -18 -15 +8 -10 -18 -10 -] -5 +20 -10 +28 b. 1958 Category 2 species (10—14 records/15—21%) achieving >20% frequency in 2008 (5 species) Hedera helix 14 Urtica dioica 14 Agrostis stolonifera 14 Festuca rubra 14 Ranunculus repens 14 % n % 64 54 79 aye) 34 50 41 22 ay) 39 14 21 39 16 24 38 Sil 46 35 177 DS Dil 6 9 23 9 13 23 15 2D D3 1 18 8) 29 43 DS 9 13 23 35 oil D)\ 42 62 21 36 58) 21 22 a2 | 19 28 Dal 16 24 +4] +32 +11 +7 +3 c. 1958 Category 3 species (5—9 records/8-14% ) achieving >20% frequency in 2008 (10 species) 14 14 Cymbalaria muralis Galium aparine Geum urbanum Fraxinus excelsior Lapsana communis Sonchus oleraceus Arrhenatherum elatius Geranium robertianum Sonchus asper © 0 OO O O OO O OO Sisymbrium officinale 14 34 30 30 28 27 ZS 50 44 44 41 40 37 34 31 28 2 +36 +30 +30 +27 +26 +23 +20 +17 +14 +7 d. 1958 Category 4 species (1-4 records/max 6% ) achieving >20% frequency in 2008 (5 species) Phyllitis scolopendrium 4 Buddleia davidii + Dryopteris filix-mas 4 Tanacetum parthenium + Asplenium ruta-muraria 4 e. Percentage frequency of species not recorded in 1958, >20% in 2008 (7 species) Epilobium ciliatum - Holcus lanatus - Cotoneaster horizontalis . Centranthus ruber - Senecio squalidus - Hieracium vulgatum : Msi 31 19 28 18 26 18 26 15 Dp 42 62 28 4] 24 35 I) 28 7 DS 16 24 +25 +22 +20 +20 +16 Oo N (oe) D. W. SHIMWELL a b G d e if g Durham City 2008 226 7 Bi, 68 158 12-46 22-40 Durham Dales 2008 162 4 22 42 120 8-33 15-00 Barnard Castle 95 V 26 38 47 10-24 14-00 Middleton-in Teesdale 98 5) 29 49 49 10—23 14°75 Stanhope 103 4 26 45 58 9-26 14-00 Fosterley & Wolsingham 107 4 26 61 46 8-33 16:26 [a. Total number of species; b. number of species in >50% samples; c. number of species in >20% samples; d. number of species in >10% samples; e. number of species in <10% samples; f. range of species numbers per sample; g. mean species number per sample. ] The differences in frequency of individual species recorded in >20% samples in either data set between the urban and rural situations are shown in Table 2 in which three groups are recognised: a. species more common on urban walls by >20% difference; b. species more common on urban walls by 10-19% difference; and c. species more common on rural walls. The greater proportion of species falling into categories a and b is the most interesting immediate feature of the table, whilst the percentage difference may be interpreted as a simple index of urbanity or rurality. This latter concept is rather tenuous when applied to the data set as a whole for the four data sub-sets provide an implication of specific floristic signatures for the dale villages. Reference to Table 3, in which the number of records for those species achieving an overall frequency >10% are presented for the four sub-sets, indicates local concentrations of certain indicator species combinations. The key species of the signatures are as follows: Barnard Castle Linaria purpurea, Phyllitis scolopendrium, Centranthus ruber Middleton-in-Teesdale Geranium lucidum, Geranium robertianum, Dryopteris filix-mas, Poa pratensis Stanhope Erinus alpinus, Poa nemoralis, Arabis caucasica, Aubretia deltoidea, Mycelis muralis, Ceastium tomentosum Frosterley/Wolsingham Cotoneaster horizontalis, Alliaria petiolata, Tanacetum parthenium, Fraxinus excelsior, Arabis caucasica, Aubretia deltoidea DISCUSSION AND CONCLUSIONS Four major discussion points emerge from the results presented above. First, the overall floristic diversity of the walls of Durham City has increased by between 39% and 42% in the period 1958-2008, suggesting that the prime factors of the amelioration of atmospheric conditions has enabled the colonization of the mural habitat with an increasing number of mainly native, as opposed neophyte species. The hypothesis that general urban gentri- fication has had a deleterious effect on bio- diversity would thus seem to be unfounded. As a corollary to these findings, it 1s interesting to note that a comparison of urban and rural walls indicates that the flora of the former is more diverse than the latter, although the proviso must be made that the rural data set did not include samples from riverside retaining walls, Whereas the urban sample did. Both these features provide a strong case for the integ- ration of biodiversity into urban planning, a major field of largely theoretical research and development for the past twenty years which has begun to manifest itself in the twenty-first century mainly through the agency of Local Urban Biodiversity Action Plans. A second discussion topic concerns the importance of urban walls as a refuge habitat for certain groups of species which have been suggested to be in decline on a national scale. Indices for change in the national distribution of individual species have been generated in recent times by Preston, Pearman & Dines (2002) and by Braithwaite, Ellis & Preston (2006), the latter providing strong evidence that many species of their category BHI17 (Built-up areas and gardens) have increased their frequency in recent times. The authors point out (p. 228) that ‘plants with low fertility requirements and those of climates charac- terised by low rainfall and warm summers have done especially well.’ The species in_ this category are those from a variety of urban habitats, not specifically those growing on walls, and conversely, many of those found on the walls of Durham City are more typical of other broad habitat types as defined by FLORISTIC DIVERSITY OF DURHAM WALLS TABLE 2. A COMPARISON OF SPECIES FREQUENCIES IN THE URBAN (DC) AND RURAL (DD) DATA SETS (species with >20% frequency in one or both data sets) 329 a. Species more common in urban walls (>20% difference) Lapsana communis Geum urbanum Holcus lanatus Chamerion angustifolium Arrhenatherum elatius Fraxinus excelsior Galium aparine Epilobium montanum Rubus fruticosus Acer pseudoplatanus Senecio squalidus Epilobium ciliatum Buddleia davidii b. Species more common in urban walls (10-19% difference) Cotoneaster horizontalis Agrostis stolonifera Phyllitis scolopendrium Sonchus asper Sisymbrium officinale Ranunculus repens Senecio jacobaea Plantago lanceolata Hieracium vulgatum Taraxacum Section Ruderalia Sambucus nigra Centranthus ruber Hedera helix Dryopteris filix-mas Sonchus oleraceus Urtica dioica Geranium robertianum Tanacetum parthenium c. Species more common on rural walls Asplenium ruta-muraria Sedum acre Cardamine hirsuta Asplenium trichomanes Poa annua Cymbalaria muralis Dactylis glomerata Poa compressa Poa pratensis Festuca rubra DC DD Diff 40 9 31 44 iS 29 4] Wg) 28 50 23 7H 34 7 DA 42 Ite) Dal 44 18 26 46 23 ye} 43 Zl DD PES 8 22 JES) 4 21 62 42 20 28 8 20 Sy) 18 19 Sy 13 II 3) IZ 19 28 10 18 20 3 17 24 df 7) SI 35 16 Dag 6 16 24 9 hs) 79 67 12 BY) 20 12 28 16 12 62 Sl 11 26 15 11 37) 28 y) 53 45 8 Sil Mp) 8 26 DD, + Hips 62 AO 4 28 24 My 45 Wp) 16 36 20 24 36 1 50 60 10 20 29 9 12 20 8 18 21 3 28 30 2 330 D. W. SHIMWELL TABLE 3. CHARACTERISTICS OF RURAL DATA SETS: FLORISTIC SIGNATURES FOR INDIVIDUAL SETTLEMENTS [numbers of records for species achieving >10% total frequency for T (Total), BC (Barnard Castle), MT (Middleton-in-Teesdale), S (Stanhope), FW (Frosterley and Wolsingham)] T BC MT S FW Taraxacum Section Ruderalia 80 17 18 20 De Asplenium ruta-muraria 74 DD 2 a) 18 Cymbalaria muralis YP D3 18 18 13) Hedera helix 61 20 8 ial 22 Cardamine hirsuta 54 NS 16 8 15 Urtica dioica 54 11 10 13 20 Epilobium ciliatum 50 16 US) 7 1) Poa annua 44 7 7 9 11 Asplenium trichomanes 43 11 13 13 6 Senecio jacobaea 42 16 9 4 13 Senecio vulgaris 40 13 18) 8 6 Festuca rubra 36 7 10 10 9 Dactylis glomerata 35 8 8 9 10 Sedum acre 34 3 14 9 8 Sonchus oleraceus 34 10 6 10 8 Epilobium montanum 28 5) 13 4 6 Geranium robertianum 28 3 14 7 4 Chamerion angustifolium 28 8) 8 8 9 Tanacetum parthenium 26 8 3 5 10 Poa pratensis 25 5 10 5 5 Rubus fruticosus | 25 9 6 6 4 Poa compressa 24 + 3 9 8 Sambucus nigra 24 7 2 6 9 Alliaria petiolata 23 6 ©) 2 10 Erinus alpinus 23 3 0 16 4 Linaria purpurea PB) 10 3 5 5 Cotoneaster horizontalis DD, l + 6 11 Galium aparine Dl 6 4 3 8 Poa nemoralis 21 l 5) 14 1 Centranthus ruber 19 8 2 4 5) Geranium lucidum 19 1 14 y) yy Dryopteris filix-mas 18 4 10 + 0 Fraxinus excelsior 18 2 4 | 11 Geum urbanum 18 >) 7 1 5 Agrostis stolonifera 16 + + >) S) Arabis caucasica 16 0 l 1 8 Digitalis purpurea 16 6 6 3 3) Aubretia deltoidea 15 0 3 7 5 Holcus lanatus 15 3 6 1 5 Ulmus glabra 15 6 4+ 3 2 Polypodium vulgare 14 2 9 2 1 Phyllitis scolopendrium 14 9 2 2 1 Arabidopsis thaliana 3) 0 8 y) 3 Cerastium tomentosum 12 l 4 6 1 Mycelis muralis 12 Dp 0 1 3) FLORISTIC DIVERSITY OF DURHAM WALLS TABLE 4. TYPES OF NATIONAL CHANGE INDICES AND URBAN CHANGE FREQUENCIES FOR SELECT SPECIES [CI — Change Index from Preston, Pearman & Dines (2002); CF — Change Factor from Braithwaite, Ellis & Pearman (2006); %C — Durham City data 1958-2008] a. All 3 measures positive Buddleia davidii Agrostis stolonifera Festuca rubra Sonchus asper Phyllitis scolopendrium Tanacetum parthenium Asplenium ruta-muraria Senecio jacobaea b. One national change index negative, urban positive Ranunculus repens Arrhenatherum elatius Urtica dioica Dryopteris filix-mas Cymbalaria muralis Sisymbrium officinale Geranium robertianum Sonchus oleraceus Lapsana communis Geum urbanum c. Both national change indices negative, urban positive Galium aparine Rubus fruticosus Epilobium montanum Hedera helix Fraxinus excelsior d. One national change index positive, urban negative Plantago lanceolata Poa annua Rumex obtusifolius Poa pratensis Acer pseudoplatanus e. All three measures negative Chamerion angustifolium Dactylis glomerata Lolium perenne Sambucus nigra Senecio vulgaris 33] Cl +3-73 +3:-66 +2:96 +0:78 +0-45 +0-23 +0-15 +0-11 +0-55 +0:°37 +0:28 +0-03 -0-10 -0-21 -0-41 -0-42 -0:47 -0:53 -0-09 -0:29 -0:39 -0-65 -0-73 +1:35 +0-83 +0-66 +0-60 -0-40 -0-01 -0-06 -0:29 -0-75 -1-08 CE +70 +77 +22 +28 +33 +47 =13} +10 +13 +18 +11 JoC +22 +1] +7 +14 +25 +20 +16 +28 +3 +20 +32 +20 +36 +7 +17 +23 +26 +30 +30 +20 +8 +4] +27 -1 -15 -10 -5 -10 -3 -18 -10 -9 -18 382 D. W. SHIMWELL Braithwaite, Ellis & Pearman. In spite of these differences in the two categories, some comparisons are possible. Using the change index (CI) of Preston, Pearman & Dines (2002) and the Change Factor (CF) from Braithwaite, Ellis & Pearman (2006), Table 4 categorises the Durham City wall species into five groups: those with positive or negative values (a and e) for all three indices; those with conflicting national indices, positive for urban wall habitats (b); those with negative national values, positive for urban wall habitats (c); and those with conflicting national indices and negative values for urban walls (d). Many interpretations of these data are possible, but the over-riding impression that emerges is the importance of urban wall habitats as refuges for species which have been otherwise viewed as being in decline. The third aspect of the research concerns the general representativeness of the data at the regional and national scales, the primary indication of which must come from _ plant sociological considerations. The accepted classification of rock crevice and_ wall vegetation in western Europe is that published in Rodwell (2000) and Rodwell ef al., (2000), and it is into the two alliances Centrantho- Parietarion Rivas-Martinez 1960 and Cymbalario-Asplenion Segal 1969 that most chasmophyte wall communities of Great Britain and Ireland may be classified. The city and county data for Durham indicate that stands of the former thermophilous comm- unities are poorly represented and although indicator species such as Centranthus ruber, Parietaria judaica, Antirrhinum majus and Erysimum cheiranthoides are generally un- common, it would seem to be of importance to flag these species as potential indicators of future climatic amelioration. Thus, the data primarily reflect the geographical realm of the Cymbalario-Asplenion. Whilst the combination of Cymbalaria muralis with either or both Asplenium ruta-muraria and A. trichomanes 1s a sound indication of such communities, combi- nations which are to be found quite commonly in both urban and rural data sets, it is important to realise that many wall samples do not fall into this typical wall community. Many stands combine species typical of weedy vegetation dominated by phospho-nitrophilous perennials (Galio-Urticetea, e.g. Galium aparine and Urtica dioica) with disturbance indicators such as Chamerion angustifolium (Epilobietea angustifoli1) and, falling into the category of ‘communities of open habitats’, await closer definition. In terms of evidence for the boreal influence and the presence of stands referable to the alliance Cystopteridion fragilis Richard 1972, the presence of Cystopteris fragilis alone, or in combination with Polypodium vulgare and species of Asplenium is also a fringe feature, the type species being found in only ten of the Durham Dales samples, notably in the more humid atmosphere of Middleton- in-Teesdale. The fourth point concerns the identification of potential climate change indicators as a basis for future monitoring. In addition to the group of Centrantho-Parietarion species and Cystopteris fragilis mentioned above, two other species are worthy of interest, namely Ceterach officinarum and Erinus alpinus. The former was not recorded in either data set, but was formerly known in one site, in Durham City (Graham 1988, Dr M. Smith pers. comm.). It was seen during the present survey in two rural localities in Weardale and one in_ lower Teesdale and appears to have spread since the publication of the County Flora in 1988. This member of the Submediterranean-Subatlantic element (Preston & Hill 1997) is presently known in only five tetrads, but may be following the trend noted by Rumsey (2002) of benefiting from the built environment in its spread in eastern England. According to The Flora of North-east England (2009), Erinus alpinus 1s only known in two tetrads post-2000. The present survey, however, added a further two tetrads to the east in Weardale and the first record for Teesdale at Barnard Castle. Being a plant of Subatlantic-Montane distribution, its liking for the cooler climates of northern England and Scotland was noted by Ellis (1993), whilst Horsfall (2002) records a change index of +1-52. It 1s a chasmophyte which 1s capable of colonizing the smallest of mortared cracks in walls wherein the small seeds germinate freely. It has a strong association with both Asplenium ruta-muraria and Cymbalaria muralis and the stands of wall vegetation in which it is dominant may be viewed as a humid, northern variant of the widespread OV42 community. The basic question to be posed in this context is whether changes in the distribution of representative species from these and other geographical elements, involving their spread into urban wall habitats, is a reliable reflection of a reaction to general climate change, or whether spread is encouraged by the peculiar microclimatic characteristics of the built environment? FLORISTIC DIVERSITY OF DURHAM WALLS 333 ACKNOWLEDGMENTS I am grateful to the Reverend Gordon and Mrs Paddy Graham for stimulating discussions in the early stages of the project and for advice on the incorporation of records for the genera Rosa and Rubus; to Dr John Richards for encourage- ment and advice on the genus Taraxacum; to Dr Margaret Bradshaw and the Upper Teesdale Botany Group for supplemental data from Barnard Castle; and to Dr Malcolm Smith for information on the changing fern flora of Durham in the past twenty-five years. The fieldwork for the urban section of the project was made possible by the requirement of Rosemary, Meredith and Joseph to attend the University of Durham Nursery on three days a week for the first two years of the project. REFERENCES BRAITHWAITE, M. E., ELLIS, R. W. & PRESTON, C. D. (2006). Change in the British Flora 1987-2004. Botanical Society of the British Isles, London. CROWE, T. M. (1979). Lots of weeds: insular phytogeography of vacant urban lots. Journal of Biogeography 6: 169-181. DARLINGTON, A. (1981). Ecology of walls. Heinemann, London. ELLIS. R. G. (1993). Aliens in the British Flora. British Plant Life, Number 2. National Museum of Wales, Cardiff. GILBERT, O. L. (1989). The Ecology of Urban Habitats. Chapman & Hall, London. GILBERT, O. L. (1996). Rooted in Stone: The Natural Flora of Urban Walls. English Nature, Peterborough. GRAHAM, G. G. (1988). The Flora & Vegetation of County Durham. Durham Flora Committee and Durham County Conservation Trust, Durham. HILL, M. O., Roy, D. B. & THOMPSON, K. (2002). Hemeroby, urbanity and ruderality: bioindicators of disturbance and human impact. Journal of Applied Ecology 39: 708-720. HOLLAND, P. G. (1972). The pattern of species density of old stone walls in western Ireland. Journal of Ecology 60: 799-805. HORSFALL, A. (2002). Evinus alpinus, in Preston, C. D., PEARMAN, D. A. & DINES, T. D. New Atlas of the British & Irish Flora. Oxford University Press. HULL LOCAL BIODIVERSITY ACTION PLAN (2002). http://www.hull.ac.uk/HBP/ActionPlan ‘The Built - Environment’ and ‘Wall Ferns’. KENT, D. H. (1961). The flora of Middlesex walls. London naturalist 40: 29-43. KENT, D. H. (1964). Notes on the flora of churchyard walls in Middlesex. London naturalist 43: 13-16. KENT, M., STEVENS, R. A. & ZHANG, L. (1999). Urban plant ecology patterns and processes: a case study of the flora of the City of Plymouth, Devon, U.K. Journal of Biogeography 26: 1281-1298. LARSON, D. W., MATTHES, U. & KELLY, P. E. (2000). Cliff Ecology. Cambridge University Press, Cambridge. LUNDHOLM, J. T. & MARLIN, A. (2006). Habitat origins and microhabitat preferences of urban plant species. Urban Ecosystems 9: 139-159. PAGE, W. (1928). (ed.) A History of the County of Durham, Volume 3; City of Durham, pp16—28. Victoria County History Series, London. PAYNE, R. M. (1978). The flora of walls in south-eastern Essex. Watsonia 12: 41-46. PRESLAND, J. L. (2007). Conserving the flora of limestone dry stone walls. Wiltshire Natural History Publications Trust. PRESLAND, J. L. (2008a). The flora of walls: dry stone versus mortared. BSBI News 108: 7-11. PRESLAND, J. L. (2008b). Is there a limestone dry stone walls community? BSBI News 109: 9-12. PRESTON, C. D. & HILL, M. O. (1997). The geographical relationships of British and Irish vascular plants. Botanical Journal of the Linnaean Society 124: 1-120. PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. (2002). (eds.) New Atlas of the British & Irish Flora. Oxford University Press. RISHBETH, J. (1948). The flora of Cambridge walls. Journal of Ecology 36: 136-148. RODWELL, J. S. (2000). (ed.) British Plant Communities, Volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press. RODWELL, J. S., DRURY, J. C., AVERIS, A. B. G., PROCTOR, M. C. F., SCHAMINEE, J. H. J. & DARGIE, T. C. D. (2000). Review of the coverage of the National Vegetation Classification. JNCC Report No. 302, Peterborough. RUMSEY, F. J. (2002). Ceterach officinarum, in PRESTON, C. D., PEARMAN, D. A. & DINES, T. D. New Atlas of the British & Irish Flora. Oxford University Press. SEGAL, S. (1969). Ecological notes on wall vegetation. Junk, The Hague. SILVERWOOD, J. H. (1965). Ferns on church walls in Norfolk. Proceedings of the Botanical Society of the British Isles 6: 120. 334 D. W. SHIMWELL SURREY COUNTY COUNCIL (2001). Urban Biodiversity Action Plan. http://www.surreycc.gov.uk THE FLORA OF NORTH-EAST ENGLAND (2009). http://www.botanicalkeys.co.uk/northumbria WOODELL, S. R. J. (1979). The flora of walls and pavings, pp 135-156 in Laurie, I.C. (ed.) Nature in Cities. John Wiley & Sons, New York. WOODELL, S. R. J. & ROSSITER, J. (1959). The flora of Durham walls. Proceedings of the Botanical Society of the British Isles 3: 257-273. WRIGHT, P. J. (1984). The flora and vegetation of walls in County Durham. MSc Dissertation, University of Durham. (Accepted May 2009) APPENDIX 1. FLORA OF DURHAM CITY WALLS, 1959-2008 [regular — recorded in both the 1959 and 2006-2008 surveys; italic — recorded in the 1959 survey but not noted as a chasmophyte in the 2006-2008 survey; bold/italic — not recorded in the 1959 survey but recorded as a chasmophyte in 2008; * nomenclature changes 1959-2008] 001 Acer plat 002 Acer pseu 003 Achi mill 004 Aego poda 005 Aeth cyna 006 Agro capi* 007 Agro stol 008 Alch moll 009 Alli peti 010 Alli ursi O11 Alnu glut 012 Alop prat 013 Alys saxa 014 Ange sylv 015 Anis ster* 016 Anth odor 017 Anth sylv 018 Anti maju 019 Aqui vulg 020 Arab thal 021 Arct minu 022 Arrth elat 023 Arte vulg 024 Aste lanc 025 Aste novi 026 Athy fili 027 Atri patu 028 Aspl adia 029 Aspl ruta 030 Aspl tric 031 Ball nigr 032 Bell pere 033 Betu pube 034 Brac sylv 035 Brom ramo* (036 Brom hord 037 Budd davi 038 Cale offi 039 Caly sepi 040 Caly silv 041 Camp patu ‘042 Camp port 043 Camp rapu —_— 044 Camp rotu 045 Caps burs 046 Card flex 047 Card hirs 048 Care pend 049 Care remo 050 Cata rigi 051 Cent nigr 052 Cent rube 053 Cera font* 054 Cera glom 055 Cera tome 057 Chae temu 056 Cham angu* 058 Chel maju 059 Clem vita 060 Cirs arve 061 Cirs vulg 062 Conv arve 063 Cory avel 064 Coto frig 065 Coto hori 066 Coto spec 067 Coto sple 068 Crat mono 069 Cruc laev 070 Cymb mura 071 Cyno cris 072 Cyst scop* 073 Dact glom 074 Dauc caro 075 Desc cesp 076 Desc flex 077 Dian cary 078 Digi purp 079 Dips full 080 Dryo fili 081 Elym cani 082 Elyt repe* 083 Epil cili 084 Epil hirs 085 Epil mont 086 Epil parv 087 Equi arve 088 Erys chei* 089 Euph amyg 090 Euph heli 091 Euph pepl 092 Fagu sylv 093 Fall bald* 094 Fall conv* 095 Fall japo* 096 Fest ovin 097 Fest prat 098 Fest rubr 099 Ficu cari 100 Fili ulma 101 Foen vulg 102 Frag vesc 103 Frag xana* 104 Frax exce 105 Gali apar 106 Gali palu 107 Gali saxa 108 Geni radi 109 Gera luci 110 Gera robe 111 Geum urba 112 Glec hede 113 Hebe xlew 114 Hede heli 115 Hera mant 116 Hera spho 117 Hesp matr 118 Hier saba* 119 Hier vulg 120 Holc lana 121 Hole moll 122 Hord muri 123 Hype andr 124 Hype caly ‘125 Hype perf 126 Hypo radi 127 Iber semp 128 Ilex aqui 129 Impa glan 130 Iris germ 131 Labu anag 132 Lami albu 133 Lami purp 134 Lapscomm. 1/35 Lath odor 136 Leon autu 137 Leon hisp 138 Leyc form 139 Ligu vulg 140 Lina purp 141 Loh mult 142 Loli pere 143 Loni peri 144 Lupi noot 145 Lyci barb 146 Lyco euro 147 Lysi vulg 148 Maho aqui —- 149 Maly sylv 150 Matr disc 151 Meco camb 152 Medi lupu 153 Meli offi 154 Ment aqua 155 Ment spic 156 Merc pere 157 Mili effu 158 Myce mura 159 Myos sylv 160 Myrr odor 161 Oena croc 162 Oxal corn 163 Papa rhoe 164 Papasomn 165 Pari juda 166 Pent semp 167 Peta hybr 168 Phal arun 169 Phyl scol 170 Pilo aura 171 Pilo offi* 172 Pisu sati 173 Plan lanc 174 Plan majo 175 Plan medi 176 Poa annu 177 Poa comp 178 Poa nemo 179 Poa prat 180 Poa triv 181 Poly avic 182 Poly seti 183 Pote anse 184 Prun vulg 185 Prun aviu 186 Prun padu 187 Prun spin 188 Pseu lute 189 Pter aqui 190 Quer petr 191 Ranu acri 192 Ranu repe 193 Rese lute 194 Rhod pont 195 Ribe rubr 196 Ribe uvac 197 Rosa cani* 198 Rubu frut 199 Rubu idae 200 Rume acet 201 Rume acel 202 Rume cris 203 Rume long 204 Rume obtu 205 Rume sang —. 206 Sagi apet 207 Sagi proc 208 Sali capr 209 Sali cine* 210 Sali vimi 211 Samb nigr 212 Saxi xumb* 213 Sedu acre 214 Sedu albu 215 Sedu refl 216 Semp tect 217 Sene cine 218 Sene jaco 219 Sene squa 220 Sene visc 221 Sene vulg 222 Sile vulg 223 Sina arve 224 Sisy offi 225 Sola dulc 226 Soli cana* 227 Sonc aspe 228 Sonc oler 229 Sorb aucu 230 Sorb inter 231 Stac sylv 232 Stel holo 233 Stel medi 234 Stel nemo 235 Symp albu 236 Symp offi 237 Tana part* 238 Tana vulg 239 Tara rude 240 Taxu bacc 241Trag prat 242 Trif dubi 243 Trif prat 244 Trif repe 245 Trip inod* 246 Tuss farf 247 Ulmu glabr =. 248: Urti dioi 249 Urti uren 250 Vale offi 251 Verb thap 252 Verb virg 253 Vero arve 259 Vici sepi 254 Vero cham 260 Vinc mino 255 Vero serp 256 Vici crac 257 Vici hirs 258 Vici sati FLORISTIC DIVERSITY OF DURHAM WALLS 335 APPENDIX 2. RECORDS OF SPECIES OCCURRING IN LESS THAN 20% (14) SAMPLES IN DURHAM CITY (Species only recorded in 2006-2008 denoted in bold italic) No. Species 13. Epilobium hirsutum (1) 12 Alliaria petiolata, Poa pratensis, Rosa canina, Ulmus glabra (4) ll Artemisia vulgaris, Asplenium trichomanes, Plantago major (3) 10. Betula pubescens, Heracleum sphondylium, Stellaria media (3) 9 Cirsium vulgare, Lolium perenne, Rumex obtusifolius (3) 8 Anisantha sterilis, Digitalis purpurea, Linaria purpurea, Meconopsis cambrica, Poa compressa (5) Alnus glutinosa, Anthriscus sylvestris, Capsella bursa-pastoris, Cardamine flexuosa, Cotoneaster spp, Elymus caninus, Festuca ovina, Myosotis sylvatica, Poa trivialis, Sagina procumbens, Taxus baccata, Trifolium repens (12) Antirrhinum majus, Cirsium arvense, Hypocharis radicata, Impatiens glandulifera, Prunus avium, Senecio vulgaris, Stachys sylvatica. (7) Achillea millefolium, Arabidopsis thaliana, Crataegus monogyna, Hieracium sabaudum, Hypericum androsaemum, Ilex aquifolium, Papaver rhoeas, Phalaris arundinacea, Rumex sanguineus, Veronica arvensis. (10) Aegopodium podagraria, Athyrium filix-femina, Brachypodium _ sylvaticum, Chaerophyllum temulentum, Euphorbia peplus, Fagus sylvatica, Festuca pratensis, Filipendula ulmaria, Rumex longifolius. (9) Ballota nigra, Calystegia sepium, Cerastium fontanum, Elytrigia repens, Equisetum arvense, Fragaria vesca, Hebe x lewisti, Lamium album, Oenanthe crocata, Parietaria judaica, Poa nemoralis, Pseudofumaria lutea, Rhododendron ponticum, Salix capraea, Sedum acre, Solanum dulcamara, Trifolium dubium, Verbascum thapsus, Veronica chamaedrys. (19) Aethusa cynapium, Agrostis capillaris, Arctium minus, Asplenium adiantum-nigrum, Calystegia_ silvatica, Campanula patula, Campanula rapunculus, Carex pendula, Cotoneaster splendens, Dipsacus fullonum, Fallopia japonica, Hesperis matronalis, Holcus mollis, Hypericum perforatum, Laburnum anagyroides, Lamium purpureum, Lolium multiflorum, Malva sylvestris, Mycelis muralis, Oxalis corniculata, Papaver somniferum, Pentaglottis sempervirens, Polygonum aviculare, Ranunculus acris, Reseda luteola, Ribes uva-crispa, Sagina apetala, Sedum album, Senecio viscosus, Sorbus aucuparia, Sorbus intermedia, Tanacetum vulgare, Vicia cracca, Vicia hirsuta. (34) Acer platanoides, Alchemilla mollis, Allium ursinum, Alyssum saxatile, Angelica sylvestris, Anthoxanthum odoratum, Aquilegia vulgaris, Aster novi-belgi, Bellis perennis, Bromopsis ramosa, Bromus hordaceus, Campanula portenschlagiana, Campanula rotundifolia, Carex remota, Catapodium rigidum, Centaurea nigra, Cerastium glomeratum, Cerastium tomentosum, Chelidonium majus, Clematis vitalba, Corylus avellana, Cotoneaster frigida, Cruciata laevipes, Cynosurus cristatus, Cystisus scoparius, Daucus carota, Erysimum cheiranthoides, Euphorbia amygdaloides, Euphorbia helioscopia, Fallopia baldschuanica, Foeniculum vulgare, Galium palustre, Geranium lucidum, Hordeum murinum, Hypericum calycinum, Iberis sempervirens, Leontodon autumnalis, Leontodon hispidus, Leycesteria formosa, Ligustrum vulgare, Lonicera periclymenum, Lycium barbarum, Lycopus europaeus, Lysimachia vulgaris, Mahonia aquifolium, Medicago lupulina, Melissa officinalis, Mentha aquatica, Mentha spicata, Milium effusum, Myrrhis odorata, Petasites hybridus, Pilosella aurantiaca, Pilosella officinarum, Polystichum setiferum, Prunella vulgaris, Prunus spinosa, Quercus robur, Ribes rubrum, Rumex acetosa, Rumex crispus, Salix cinerea, Salix viminalis, Sedum reflexum, Sempervivum tectorum, Senecio cinerea, Silene vulgaris, Solidago canadensis, Stellaria holostea, Symphoricarpos albus, Tragopogon pratensis, Tussilago farfara, Valeriana officinalis, Verbascum virgatum, Veronica serpyllifolia, Vicia sativa. (75) 336 D. W. SHIMWELL APPENDIX 3. SUMMARY OF MONAD DISTRIBUTION OF SAMPLE SITES | km grid square reference (number of samples); full details are available on request DURHAM CITY NZ2642 (12); NZ2643 (2); NZ2741 (8); NZ2742 (29); NZ2743 (1); NZ2841 (1); NZ2842 (7); NZ2843 (2); NZ2940 (2); NZ2946 (1) DURHAM DALES Barnard Castle: NZ0416 (15); NZ0516 (15); Middleton-in-Teesdale: NY9425 (23); NY9426 (2); NY9525 (5). Stanhope: NY9940 (5); NZ9839 (3); NZ9939 (22); Frosterley & Wolsingham: NZ0236 (12); NZ0336 (2); NZ0736 (2); NZ0737 (12); NZ0837 (2) FLORISTIC DIVERSITY OF DURHAM WALLS S57) APPENDIX 4. FLORA OF DURHAM DALES WALLS, 2006-2008 (bold/italic — flora of Durham Dales walls; italic — flora of Durham City walls 1958-2008) OO] Acer plat 007 Agro stol 013 Alys saxa 019 Aqui vulg 025 Aste novi 031 Ball nigr 037 Budd davi 043 Camp rapu 049 Care remo 055 Cera tome 061 Cirs vulg 067 Coto sple 073 Dact glom 079 Dips full 085 Epil mont 091 Euph pepl 097 Fest prat 103 Frag xana* 109 Gera luci 115 Hera mant 121 Holc moll 127 Iber semp 133 Lami purp 139 Ligu vulg 145 Lyci barb 151 Meco camb 157 Mili effu 163 Papa rhoe 169 Phyl scol 175 Plan medi 181 Poly avic 187 Prun spin 193 Rese lute 199 Rubu idae 205 Rume sang 211 Samb nigr 217 Sene cine 223 Sina arve 229 Sorb aucu 235 Symp albu 241Trag prat 247 Ulmu glabr 253 Vero arve 259 Vici sepi 002 Acer pseu 008 Alch moll 014 Ange sylv 020 Arab thal 026 Athy fili 032 Bell pere O38 Cale offi 044 Camp rotu 050 Cata rigi O57 Chae temu 062 Conv arve 068 Crat mono 074 Dauc caro 080 Dryo fili 086 Epil parv 092 Fagu sylv 098 Fest rubr 104 Frax exce 110 Gera robe 116 Hera spho 122 Hord muri 128 Ilex aqui 134 Laps comm 140 Lina purp 146 Lyco euro 152 Medi lupu 158 Myce mura 164 Papa somn 170 Pilo aura 176 Poa annu 182 Poly seti 188 Pseu lute 194 Rhod pont 200 Rume acet 206 Sagi apet 212 Saxi xumb* 218 Sene jaco 224 Sisy offi 230 Sorb inter 236 Symp offi 242 Trif dubi 248 Urti dioi 254 Vero cham 260 Vince mino 003 Achi mill 009 Alli peti O15 Anis ster* 021 Arct minu 027 Atri patu 033 Betu pube 039 Caly sepi 045 Caps burs O51 Cent nigr 056 Cham angu* 063 Cory avel 069 Cruc laev 075 Desc cesp O81 Elym cani 087 Equi arve 093 Fall bald* 099 Ficu cari 105 Gali apar 111 Geum urba 117 Hesp matr 123 Hype andr 129 Impa glan 135 Lath odor 141 Loli mult 147 Lysi vulg 153 Meli offi 159 Myos sylv 165 Pari juda 171 Pilo offi* 177 Poa comp 183 Pote anse 189 Pter aqui 195 Ribe rubr 201 Rume acel 207 Sagi proc 213 Sedu acre 219 Sene squa 225 Sola dulc 231 Stac sylv 237 Tana part* 243 Trif prat 249 Urti uren 255 Vero serp 135 in common 004 Aego poda O10 Alli urst 016 Anth odor 022 Arrh elat 028 Aspl adia 034 Brac sylv 040 Caly silv 046 Card flex 052 Cent rube O58 Chel maju 064 Coto frig 070 Cymb mura 076 Desc flex 082 Elyt repe* 088 Erys chei* 094 Fall conv* 100 Fili ulma 106 Gali palu 112 Glec hede 118 Hier saba* 124 Hype caly 130 Iris germ 136 Leon autu 142 Loli pere 148 Maho aqui 154 Ment aqua 160 Myrr odor 166 Pent semp 172 Pisu sati 178 Poa nemo 184 Prun vulg 190 Quer petr 196 Ribe uvac 202 Rume cris 208 Sali capr 214 Sedu albu 220 Sene visc 226 Soli cana* 232 Stel holo 238 Tana vulg 244 Trif repe 250 Vale offi 256 Vici crac OOS Aeth cyna O11 Alnu glut O17 Anth sylv 023 Arte vulg 029 Aspl ruta 035 Brom ramo* 041 Camp patu 047 Card hirs 053 Cera font* O59 Clem vita 065 Coto hori O71 Cyno cris 077 Dian cary 083 Epil cili 089 Euph amyg 095 Fall japo* 101 Foen vulg 107 Gali saxa 113 Hebe xlew 119 Hier vulg 125 Hype perf 131 Labu anag 137 Leon hisp 143 Loni peri 149 Maly sylv 155 Ment spic 161 Oena croc 167 Peta hybr 173 Plan lance 179 Poa prat 185 Prun aviu 191 Ranu acri 197 Rosa cani* 203 Rume long 209 Sali cine* 215 Sedu refl 221 Sene vulg 227 Sonc aspe 233 Stel medi 239 Tara rude 245 Trip inod* 251 Verb thap 257 Vici hirs Species recorded on Durham Dales walls and not in Durham City 001 Alli oler 007 Coni macu 013 Euph esul 019 Ribe nigr 025 Syri vulg 002 Arab cauc 008 Crep capi 014 Leuc vulg 0Q20Ribe sang 026 Vero hede 003 Arab hirs 009 Cyst frag 015 Orig vulg 021 Sedu spur 027 Vinc majo 004 Aren serp 010 Erig karv 016 Poly vulg 022 Sher arve 005 Aubr delt 011 Erin alpi 017 Pote ster 023 Sile dioi 006 Agro capi* 012 Alop prat 018 Anti maju 024 Aste lanc 030 Aspl tric 036 Brom hord 042 Camp port 048 Care pend 054 Cera glom 060 Cirs arve 066 Coto spec 072 Cyst scop* 078 Digi purp OS4 Epil hirs 090 Euph heli 096 Fest ovin 102 Frag vesc 108 Geni radi 114 Hede heli 120 Holc lana 126 Hypo radi 132 Lami albu 138 Leyc form 144 Lupi noot 150 Matr disc 156 Merc pere 162 Oxal corn 168 Phal arun 174 Plan majo 180 Poa triv 186 Prun padu 192 Ranu repe 198 Rubu frut 204 Rume obtu 210 Sali vimi 216 Semp tect 222 Sile vulg 228 Sonc oler 234 Stel nemo 240 Taxu bacc 246 Tuss farf 252 Verb virg 258 Vici sati 006 Camp pers 012 Erop vern 018 Ranu bulb 024 Symp orie 338 D. W. SHIMWELL APPENDIX 5. RECORDS OF SPECIES OCCURRING IN LESS THAN 10% (12) SAMPLES IN THE DURHAM DALES Species Cotoneaster sp., Rosa canina, Sedum album Cystopteris fragilis, Meconopsis cambrica, Pseudofumaria lutea, Ribes uva-crispa, Taxus baccata Hieracium vulgatum, Veronica arvensis Ranunculus repens Myosotis sylvatica, Plantago lanceolata Anthriscus sylvestris, Capsella bursa-pastoris, Leucanthemum vulgare, Sempervivum tectorum Antirrhinum majus, Cerastium fontanum, Cirsium vulgare, Crataegus monogyna, Erophila verna, Euphorbia peplus, Parietaria judaica, Ribes sanguineum, Sagina procumbens Achillea millefolium, Arenaria_ serpyllifolia, Erysimum cheiranthoides, Sagina apetala, Sedum spurium, Stellaria media, Symphoricarpos albus Agrostis capillaris, Alchemilla mollis, Anisantha_ sterilis, Cardamine flexuosa, Elytrigia repens, Euphorbia esula, Lolium perenne, Medicago lupulina, Pilosella aurantiaca, Plantago major, Rubus idaeus, Rumexs obtusifolius, Senecio viscosus, Trifolium repens, Veronica hederifolia Alyssum saxatile, Aquilegia vulgaris, Asplenium adiantum-nigrum, Betula pubescens, Bromus hordaceus, Calystegia sivatica, Campanula persicifolia, Campanula portenschlagiana, Conium maculatum, Crepis capillaris, Centaurea nigra, Fragaria vesca, Lamium album, Poa angustifolia, Prunella vulgaris, Ribes nigrum, Rumex acetosa, Salix capraea, Sorbus aucuparia, Symphytum orientale, Syringa vulgaris, Verbascum thapsus, Veronica chamaedrys, Veronica serpyllifolia, Vinca major Allium oleraceum, Anthoxanthum odoratum, Arabis hirsuta, Asplenium adiantum-nigrum, Athyrium filix-femina, Bromopsis ramosa, Catapodium rigidum, Cerastium glomeratum, Chaerophyllum temulentum, Cirsium arvense, Dipsacus fullonum, Elytrigia canina, Equisetum arvense, Erigeron karvinskianus, Euphorbia amygdaloides, Fagus sylvatica, Ilex aquifolium, Laburnum anagyroides, Leycesteria formosa, Lonicera periclymenum, Malva sylvestris, Myrrhis odorata, Origanum vulgare, Papaver rhoeas, Pilosella officinarum, Potentilla_ sterilis, Prunus avium, Prunus laurocerasus, Prunus spinosa, Ranunculus bulbosus, Reseda luteola, Sherardia arvensis, Silene dioica, Stachys sylvatica, Stellaria graminea, Trifolium dubium, Trifolium pratense, Vicia sepium je eee————e——eee————eEeEeEsee Watsonia 27: 339-353 (2009) 339 Distribution, status and ecology of Blysmus compressus (L.) Panz. ex Link on the Sefton Coast sand-dunes, Merseyside P. H. SMITH* 9 Hayward Court, Watchyard Lane, Formby, Liverpool L37 3QP. ABSTRACT This paper describes a 2008 survey of the nationally declining Blysmus compressus on the Sefton Coast sand-dunes (v.c. 59), where it has been known since 1801. The plant was recorded in 18 sites, occupying over 3000 m’, with an estimated total population of 15,000-—20,000 individuals. Evidence suggests B. compressus has increased here in recent decades, relatively few known sites having been lost. The plant was mainly found in calcareous wet-slacks, though three sites were described as damp grassland. Most localities are of fairly recent origin and have a history of recreation disturbance and _ grazing, especially by rabbits. A few are winter-grazed by livestock and some are mown annually. Ninety associated vascular. taxa were recorded. NVC analysis gave mainly poor to very poor fits to known dune-slack communities (SD15, SD16 and SD17) and two mesotrophic grassland types (MG8 and MGI11). The conservation requirements for B. compressus are discussed. KEYWORDS: communities, conservation manage- ment, dune-slacks, Ellenberg’s indicator values, history, National Vegetation Classification, mowing, population, rabbit-grazing, trampling. INTRODUCTION Blysmus compressus (L.) Panz. ex Link (Flat- sedge) (Cyperaceae) is a far-creeping rhizo- matous perennial of open vegetation in marshes and fens and in short, sedge-rich, damp grassland, calcareous flushes and _ stream borders that are subject to flooding, especially on base-rich rocks. The plant is a hemi- cryptophyte and a member of the European temperate flora with a continental distribution in Western Europe, extending as far north as southern Scandinavia and east to Central Asia (Foley & Porter 2002; Hill et al. 2004; Jermy et al. 2007; Walker 2008). Ellenberg’s indicator values in Hill et al. (2004) show that this species is a light-loving plant (L = §8), associated with constantly moist to wet soils (F = 8) that are base-rich (R = 8) and relatively infertile (N = 3) and that the plant is absent from saline sites (S = 0). *e-mail:philsmith1941 @tiscali.co.uk DISTRIBUTION IN BRITAIN B. compressus 1s widespread but localised in Britain. It 1s recorded for most English counties but in abundance only in the north and west, reaching a maximum elevation of 490 m in Co. Durham (v.c. 66). It is rare in lowland England, though there are still large populations on the Rivers Kennett and Thames. In Wales, this species is confined to a single site in Brecon (v.c. 42), while Scotland has no populations north of the Central Belt. B. compressus does not occur in Ireland (Foley & Porter 2002; Walker 2008). This is one of the most rapidly declining vascular plants in Britain, having been lost from more than half its pre-1930 hectads by 1962 and, subsequently, from 40% of its post- 1930 hectads. Its change index since 1962 is -]-28 (Foley & Porter 2002); as a consequence, it is a Red Data listed species in Britain, classified as Vulnerable (Cheffings & Farrell 2005). The plant has become extinct in South Devoni(wvee- 03). Kent (cn 5,116); Surrey (v:c: 17), Middlesex (v.c. 21), West Suffolk (v.c. 26), Staffordshire (v.c. 39), Shropshire (v.c. 40), Leicestershire (v.c. 55) and Nottinghamshire (v.c. 56). Similar declines have been reported elsewhere in Europe, for example, in Croatia and Finland (Walker 2008). DISTRIBUTION IN NORTH WEST ENGLAND There are no records of B. compressus for Cheshire (v.c. 58) (Newton 1971, G. M. Kay in litt. 2009) but in South Lancashire (v.c. 59), Savidge et al. (1963) record the plant as “occasional, locally common” in dune-slacks and marshes, at intervals along the coast between Hightown and Southport. They mention only one site away from what is now the Sefton Coast; a pasture near the canal at Nelson in 1926, where the plant does not seem to have been recorded since. The New Flora of South Lancashire Project (D. P. Earl in litt. 2008) has 23 records of 340 P. H. SMITH B. compressus from 1851 onwards in hectads SD20, 21, 31 and 39, these being situated along the Merseyside coast from Liverpool to Southport, and SD83 which includes Nelson in eastern Lancashire. In addition, there are 18 sheets of gatherings in the Manchester Museum herbarium (MANCH) for localities between Liverpool and Southport, most from the 19th century, the earliest on 17 June 1840 at Bootle, collected by J. B. Wood. The Liverpool Museum herbarium (LIV) has two earlier sheets, from 1821, from near Rimrose Bridge, Seaforth (Mr Shepherd; ex. J. Shillito) and 1801, north shore opposite Bootle (ex Liverpool Botanic Garden collection). Both these sites have been destroyed and are in the modern borough of Sefton. B. compressus has always been rare in West Lancashire (v.c. 60), being confined historically to two nearby sites on the bank of the R. Lune at Wrayton and at the junction of the R. Greta and the Lune. It was also found in dune-slacks at Lytham St. Anne’s on the north shore of the Ribble Estuary (Wheldon & Wilson 1907). More recently, the plant has been found in small quantity near these historical localities; in 1980 on Lune ox-bows north of Melling and at two adjacent dune sites at Fairhaven, Lytham, in 1998, one of these being lost to erosion by 2003 (E. F. Greenwood in litt. 2008). In the northern part of this vice- county, Livermore & Livermore (1987) describe it as “rare in a wet meadow”, where they found only two or three plants. It is not clear whether this is one of the Lune sites. There is also an un-localised record from the “Heysham Peninsula” in the north of the vice- county by G. W. Garlick in about 1957 (E. F. Greenwood in /itt. 2008) The Lancashire Biodiversity Action Plan (BAP) considers B. compressus to be critically endangered in v.c. 60, being threatened by nutrient enrichment, a lowered water-table and physical disturbance at its coastal locality and vulnerable to drainage, fertiliser run-off and river management works on the R. Lune (Greenwood 2002). The Lancashire Wildlife Trust’s Endangered Plants Project (E. F. Greenwood in /itt. 2008) recorded two colonies of the plant in 2003 on Lune ox-bows about 200 m apart, the total population being only about 60 spikes. Due to multiple private ownerships, this area is difficult of access and largely unexplored. Two small colonies were found on the dunes at Ansdell, Lytham St. Anne’s in 2003, the first being lost by 2005S. The second site had three spikes in 2007 and 100-200 in 2008 (E. F. Greenwood in litt. 2008). This seems to be the one described by Skelcher (2008) as “a good stand” in dune- slack vegetation north of Fairhaven Lake. B. compressus 1s much more widespread in Cumbria, with records for 69 tetrads, though Halliday (1997) states that it is nearly confined to a limited area of north Westmorland (v.c. 69), where it 1s quite frequent in calcareous flushes and marshes and by _beck-sides, particularly on ground subject to flooding. A few sites elsewhere include the edge of a salt- marsh near Bowness-on-Solway where some degree of tolerance to salinity is implied, despite the plant having an Ellenberg salinity (S) score of 0. OCCURRENCE ON THE SEFTON COAST (V.C. 59) B. compressus has been known from slacks in the Sefton Coast sand-dune system for over 200 years, 19th century herbarium records having already been mentioned. The earliest relevant flora is by Hall (1838) who states: “There are specimens of B. compressus in the Herbarium at the Botanic Gardens from the north shore of Liverpool, but I am not aware of it having been found there recently by any botanist.” However, only 13 years later, Dickinson (1851) remarks: “Grassy spots on Crosby sandhills, and near Bootle Land Marks, abundant. Also at Formby and Southport’. Despite the plant’s unassuming appearance, MeNicholl (1883) writes in his Handbook for Southport: “In parts of the sandhills ..... there are tracts such as are congenial to semi- amphibious plants. Here may be gathered ..... the Blysmus compressus.” More recently, it is described by Smith (1978) as a plant almost confined to dune slacks within this vice-county and listed as “very local, wet slacks” in a species list for Ainsdale Sand Dunes National Nature Reserve (NNR) (Payne 1982). In the early 1980s, B. compressus was known from under 10 sites in the dune system (Smith 1983). Edmondson et al. (1993a) mention its presence in a few slacks at Birkdale Sandhills Local Nature Reserve (LNR), Lifeboat Road and Cabin Hill NNR. In a more detailed study of 26 dune-slacks at Birkdale LNR, Smith (2006a) found B. compressus in ten of these slacks in 2003, an increase in occurrence from only two slacks twenty years earlier (Smith 1983). BLYSMUS COMPRESSUS ON THE SEFTON DUNES 34] SURVEY RATIONALE Reflecting its declining national status, B. compressus was added to the list of UK Biodiversity Action Plan (BAP) species in 2007 and highlighted for special study in the BSBI Threatened Plants Project in 2008 (Walker 2008). It is also listed as a Species of Conservation Importance in North West England (Regional Biodiversity Steering Group 1999). However, recent observations (e.g. Smith 2006a) suggest that, far from declining, this species may actually be increasing on the Sefton Coast. It was therefore thought opportune to investigate its status here in a comprehensive survey. As B. compressus 1s not mentioned in the UK National Vegetation Classification (NVC) and relatively little seems to be known in detail about its habitat requirements (Walker 2008) it was also decid- ed to collect relevant floristic and habitat data. METHODS Based on detailed personal knowledge of the dune system and previous reports of survey, all known and likely sites for B. compressus were visited in June, July and August 2008. Populat- ions were marked out using bamboo canes, their approximate areas measured by pacing and numbers of plants estimated. National Grid References of locations were determined using a hand-held GPS device. Distribution maps were prepared using the Sefton Coast Geographic Information System (GIS). For each site, depending on its size, between one and eight samples of representative vegetation were recorded in 2 m x 2 m quadrats using NVC methodology (Rodwell 1991). These were analysed using a modified TABLEFIT prog- ramme to determine the degree of fit to known NVC communities and sub-communities (Hill 1996). Notes were also taken on habitat type, condition and current management. The origin and past management of sites was investigated using literature sources, maps and_aerial photographs. Details of recently lost sites were investigated through contacts with local botanists and reference to the literature. RESULTS DISTRIBUTION OF SITES Eighteen sites for B. compressus were recorded in the dune system (Figure |). They are highly localised, falling into three distinct groupings; four at Altcar Rifle Range and Cabin Hill NNR in the south, five at Ainsdale-on-Sea, about 8 km to the north, and nine sites in the Birkdale sandhills (including Royal Birkdale Golf Course), another 2°5 km further north (Table 1). They occupy only three tetrads: SD20X, SD31B and SD31C. Despite extensive searches in 2008, no plants of this species were found in ostensibly suitable habitat elsewhere in the dunes, though a few additional historical sites are known and three came to light after the field work was completed (see below). TABLE |. BLYSMUS COMPRESSUS SITES ON THE SEFTON COAST, 2008 Site Central Grid Habitat Date of origin Area of Estimated Reference (SD) (where known) Colony (m’) population Altcar Rifle Range “I” range 2888 0462 Damp grassland After 1924 a2 1000+ Cabin Hill NNR Slack 1 2828 0524 Wet-slack 1970/71 1192 10,000 Slack 3 2835 0529 Wet-slack 1970/71 54 100+ Slack 20 2826 0520 Wet-slack 1945-50 336 1000+ Sands Lake, Ainsdale 3013282 Damp grassland After 1911 23 100 Ainsdale frontals Slack 49 3009 1330 Wet-slack Early/mid 1970s 145 1000+ Slack 50 800771315 Wet-slack Early/mid 1970s 384 100s Slack 51 3002 1309 Wet-slack Early/mid 1970s 24 100 Slack 52 3008 1309 Wet-slack Early/mid 1970s >) 10s Royal Birkdale pond 3159 1529 ~=Marshy pond edge 6 10s 3rd fairway 3162 1528 Damp grassland 180 100 Birkdale Sandhills Slack 26 S143 1537) Wet-slack 1929 7 100+ Birkdale frontals Slack 27 3134 1524 Wet-slack Early/mid 1970s 58 100s Slack 28 3142 1555 Wet-slack Early/mid 1970s 162 1000+ Slack 30 3202 1631 Dry-slack Pre-1970 56 100+ Slack 31 3196 1619 Wet-slack Pre-1969 100 100 Slack 32 3191 1614 Dry-slack Mid 1970s 63 100s Slack 33 3193 1614 Dry-slack 1976 170 100s 342 P. H. SMITH TABLE 2. WEIGHTED ELLENBERG’S INDICATOR VALUES FOR ASSOCIATES OF BLYSMUS COMPRESSUS Taxon Occ. in % F R N WE wR wN Sil quadrats Carex flacca 44 86:3 5 6 2 220 264 88 Trifolium repens 4] 80-4 5) 6 6 205 246 246 Agrostis stolonifera 36 70-6 6 7 6 216 Dy) 216 Salix repens 35 68-6 7 6 3 245 210 105 Mentha aquatica 30 58:8 8 7 5 240 210 150 Hydrocotyle vulgaris 29 56:9 8 6 3 SD 174 87 Festuca rubra Di SL) 5) 6 5 135 162 135 Holcus lanatus 27 a2;9 6 6 5 162 162 IB) Rubus caesius 24 47+] 7 7 6 168 168 144 Juncus articulatus | 41-2 9 6 3) 189 126 63 Dactylorhiza incarnata ssp. coccinea 7) 8)8)3) 9 6 2 153 102 34 Carex arenaria 16 31-4 3 5) D A8 80 a7 Equisetum palustre 16 31-4 8 6 3 128 96 48 Prunella vulgaris 16 31-4 5 6 4 80 96 64 Ranunculus repens 16 31-4 7 6 7 WZ 96 112 Cirsium arvense 15 29-4 6 d/ 6 90 105 90 Ranunculus acris IS) 29-4 6 6 4 90 90 60 Plantago lanceolata 14 PEE 5) 6 4 70 84 56 Eleocharis palustris 13 DD 10 6 4 130 78 52 Galium palustre (3) PISS 9 5) 4 la 65 SZ Ranunculus flammula 13 PIS 9 5 8) 117 65 39 Salix cinerea 13 DISS) 8 6 5 104 78 65 Lathyrus pratensis 12 235) 6 6 5 2 72 60 Carex hirta 1] 21-6 7 q] 6 TY Wi 66 Eleocharis quinqueflora ILI 21:6 9 7 Z 99 ii DD Leontodon autumnalis He PINS) 6 6 + 66 66 44 Lotus corniculatus jUgl 21-6 4 6 2 44 66 22 Senecio jacobaea 1] 21-6 4 6 4 44 66 44 Potentilla anserina 10 19-6 1 7 6 70 70 60 Rhinanthus minor 10 19-6 ) 6 4 50 60 40 Vicia cracca 10 1956 6 i 5 60 70 50 Bellis perennis 9 17-6 5 6 4 45 54 36 Juncus inflexus 9 17:6 i 7 5 63 63 45 Trifolium pratense 9 17-6 5 I >) 45 63 45 Epipactis palustris 8 Se7/ 8 if 5) 64 56 24 Euphrasia nemorosa 8 Py 5 6 + AO 48 ay) Cardamine pratensis if, 13-7) 8 5 4 56 35 28 Lotus pedunculatus 7 37 8 6 + 56 42 28 Lythrum salicaria if 137, 9 7 5 63 49 3)5) Taraxacum officinale 7 137 5) 7 6 35 49 42 Anthoxanthum odoratum 6 ies 6 4 3) 36 24 18 Glaux maritima 6 11-8 w 7 5) 42 42 30 Plantago major 6 11-8 5 6 7 30 36 42 Potentilla reptans 6 11-8 5) di 5 30 42 30 Equisetum variegatum 5 9-8 8 8 3 40 40 5 Lolium perenne 5 9-8 5) 6 6 YS) 30 30 Poa pratensis 5 9-8 5 6 5 Ts) 30 25) Cerastium fontanum 4 78 5 5 4 20 20 16 Cerastium semidecandrum + 7:8 3 6 3 12 24 2 Taxon Cynosurus cristatus Parnassia palustris Trifolium fragiferum Triglochin palustre Carex disticha Centaurea nigra Elytrigia repens Equisetum arvense Tris pseudacorus Juncus balticus Juncus gerardit Poa humilis Pulicaria dysenterica Veronica scutellata Bolboschoenus maritimus Carex nigra Carex viridula ssp. viridula Centaurium erythraea Dactylorhiza fuchsii Epilobium hirsutum Leontodon saxatilis Pastinaca sativa Phragmites australis Poa annua Sagina procumbens Tussilago farfara Agrostis capillaris Calystegia silvatica Cirsium vulgare Dactylis glomerata Dactylorhiza praetermissa Epilobium obscurum Epilobium palustre Equisetum fluviatile Eriophorum angustifolium Filipendula ulmaria Glechoma hederacea Lycopus europaeus Rosa rugosa Sonchus arvensis Totals BLYSMUS COMPRESSUS ON THE SEFTON DUNES 343 TABLE 2 CONTINUED Occ. in % F N WE wR wN 51 quadrats 4 7:8 5 6 4 20 24 16 4 78 8 7 3 By 28 12 4 7:8 7 7 6 28 28 24 4 7:8 9 6 D, 36 24 8 3 5:9 8 6 4 24 18 12, 3 5-9 5 6 5) IS) 18 15 3 5:9 5) 7 7 IS) 21 44 3 5:9 6 6 6 18 18 18 3) is) 9 6 6 Dif 18 18 3 5:9 8 5) D; 24 LS 6 3 5:9 7 7 6 Di Dil 18 3 5-9 6 6 4 18 18 12 3 5-9 7 7 4 Do] Dal 12 3 5:9 9 5) 3 Di 15 9 py) 3-9 10 8 q 20 16 14 yD 3-9 8 4 py) 16 8 4 2 3-9 7 7 3 14 14 6 2 3:9 5) 6 3 10 12 6 y, 3-9 8 7 3 16 14 6 2 3-9 8 i 7 16 14 14 D; 3-9 5) 6 3 10 1 6 Dy 3-9 4 7 5) 8 14 10 D 3-9 10 7 6 20 14 12 2 ay) 5 6 7 10 12 14 y) 3-9 6 6 > 12 Wy, 10 2 3-9 6 6 6 12 12 12 1 2:0 5 4 4 5 4 4 1 2:0 5 7 6 5 dl 6 1 2:0 > 6 6 5 6 6 ] 2:0 5) I 6 5) 7 6 ] 2:0 8 df 3 8 7] 8} l UG) 8 5) 5 8 5 5) 1 2:0 8 5 3 8 5) 3 1 D0 10 6 4 10 6 4 l 2:0 9 4 1 9 4 1 1 2:0 8 6 5 8 6 5 | 2:0 6 7 oF 6 7 7 I 2:0 8 7 6 8 7 6 ] 2:0 3 6 3) 3 6 3 I 2:0 6 7 6 6 7 6 798 5144 4905 3394 6.25) 10-250 4-3) O45 6:15 4:25 Means F = moisture; R = reaction; N = nitrogen; w = weighted 4 344 P: He SMM kilometres Birkdale Frontals FIGURE |. Locations of Blysmus compressus on the Sefton Coast, 2008. BLYSMUS COMPRESSUS ON THE SEFTON DUNES 345 POPULATION SIZES Estimated areas of colonies range from 1192 m° at Cabin Hill slack 1 to 5 m* at Ainsdale slack 52, the total extent being 3017 m* (Table 1). Due to time limitations, only approximate estimates were made of plant numbers. These range from under 100 spikes at Ainsdale slacks 51 and 52 to about 10,000 at Cabin Hill slack 1. Other large populations, with 1000 or more spikes, were recorded at Altcar Rifle Range, Ainsdale slack 50, Ainsdale slack 49 and Birkdale slack 28 (Table 1). Overall, a conservative estimate for the total Sefton Coast population of B compressus is 15,000—20,000 plants. HABITATS Most Sefton Coast sites for B. compressus are dune slacks that have formed secondarily by wind-erosion or by sand-winning and have a seasonally fluctuating water-table. Following Ranwell’s (1972) criteria, 11 are considered to be wet-slacks, flooded during the winter, and three dry-slacks, rarely holding water. All seem to accord with the Type B slack of Davy et al. (2006), this being a precipitation-fed slack in which water is lost by evaporation. A small colony at Royal Birkdale Golf Course is associated with the edge of a pond, while three sites (at Altcar Rifle Range, Ainsdale Sands Lake and Royal Birkdale) were categorised as damp grassland (Table 1). The ground profile of dune-slacks often leads to different degrees of wetness from dry around the edges to semi-aquatic in the centre; at some sites this seemed to influence the distribution of B. compressus. For example, at Cabin Hill slack 1, the plant occupies a zone between damp conditions to the north, with much Salix repens, and semi-aquatic vegetation to the south dominated by Equisetum fluviatile. It was also noted that, within several sites, B. compressus plants were often concentrated in areas where the vegetation was somewhat suppressed or altered by recreational trampling or rabbit (Oryctolagus cuniculus) -grazing or both; for example, on the surfaces or along the edges of lightly-used informal footpaths. Such areas often show a reduced frequency of S. repens, a dominant plant in many Sefton Coast dune- slacks (Gateley & Michell 2004). The flower spikes of B. compressus were often seen to stand proud of grazed vegetation, suggesting that rabbits may avoid the plant. At a few of the more heavily trampled sites, B. compressus plants were suppressed, being only a few cm tall; however, in most wet-slacks, the popula- tions consisted of vigorous, strongly flowering individuals, some stems attaining a height of 50 cm at Birkdale slack 28, compared to the 45 cm maximum cited in Jermy et al. (2007). Soil characteristics have been documented for some of the B. compressus sites. Smith (1983; 1984) studied soil profiles at eight of the Ainsdale-Birkdale slacks. Two are described as peaty-gleys, the others being ground-water gleys. pH values range from 7-0 to 8-8 at 3-5 cm depth and 6:9 to 8-9 at 10-15 cm, reflecting the base-rich nature of the parent material which has about 5% calcium carbonate derived from intertidal mollusc shells (Smith 2009). Such slack soils throughout this dune system have been included in the Greatstone Series defined by the Soil Survey of England & Wales, a general account of their characteristics on the Sefton Coast being given by James (1993). ORIGIN AND MANAGEMENT OF SITES At the southernmost site on Altcar Rifle Range, B. compressus 1s confined to wet grassland at the northern edge of “I range, where it was first found in 2008. “I” range seems to have been laid out when the ranges were rebuilt in 1923/24 (Cook 1989). This area was added to the Altcar Sand Dunes & Foreshore Site of Special Scientific Interest (SSSI) in 1986 and subsumed into the Sefton Coast SSSI in 2000. It has been managed by mowing, this being delayed until late summer (after 15 July) since SSSI designation. Some limited rabbit-grazing may occur but this species is controlled and has low populations at Altcar. At Cabin Hill, the slacks occupied by B. compressus are of fairly recent origin. Slack 20 was created by sand-winning, during the 1940s and early 1950s, which removed the largest sand-dune on this part of the coast down to the water-table. Because of a perceived threat to coast protection, a planning application for further sand removal was refused in 1952 (Crosby 2007). Slacks | and 3, the former now supporting the largest colony of B. compressus found during the study, were formed in 1970/71 when the then Mersey & Weaver River Authority constructed a 780 m-long barrier bank to act as a secondary sea-defence. The slacks are shallow borrow-pits, in the earlier sand-quarry floor, from which material to build the bank was extracted (Smith 2007a). B. compressus was first recorded at Cabin Hill in July 1979 in slack 20, from which it has apparently spread to the adjacent sites, being 346 P. H. SMITH noted in slack | in July 2003 and slack 3 in June 2008 (personal observations). Edmondson et al. (1988/89) also mention its presence in a Target Note for slack 20 in their compre- hensive NVC survey of the dune system. This area was leased as a National Nature Reserve in 1984 (declared 1991). During the 1970s and 1980s, sporadic grazing by small numbers of cattle and horses took place but, since 1992, the reserve has been managed by extensive winter-grazing using Herdwick sheep. There is also a large rabbit population, though only slack 3 is heavily rabbit-grazed. Slack 1 was mowed on four occasions in the 1990s and willow (Salix) scrub was removed in 1989 and 1990 (Smith 2007b). The Sands Lake, Ainsdale site is difficult to date accurately but this dune area developed after the Southport & Cheshire Lines Extension Railway was built along the foreshore in 1884 (Smith 1983). The lake itself was dug by hand in 1911, the surrounding area being subject to high levels of recreational disturbance there- after. The B. compressus site, of which the current population is a small relict, was first recorded in July 1979 as “a large colony of Blysmus at the north end of the lake” (personal observations). It was also. documented in a 1988 Target Note as “Rich strip along well- used area between Sands Lake and tall mixed scrub to east. Short, dense sward has Carex flacca, Blysmus compressus, Lolium perenne, Trifolium repens etc.” (Edmondson et a. 1988/89). Since then, dense scrub, especially of Hippophae rhamnoides, together with a tall herbaceous sward of Epilobium hirsutum etc., has overtaken much of the site, B. compressus, being restricted to a 1 m-wide zone of short, weedy turf on the edge of a carpark. Here, it is subject to trampling and, probably, some rabbit-grazing. There is no formal conservation management. To the west and north-west of Sands Lake, slacks 49, 50, 51 and 52 were formed by wind- erosion as recently as the mid 1970s when the water-table was exceptionally low and levels of recreational trampling high. Thus, they were not mapped as wet-slacks by Smith (1978) during a comprehensive survey of dune wet- lands. However, by 1983, they were well- established and mapped as “incipient” (1.e. immature) wet-slacks (Smith 1983). Twenty years later, they were listed as “mature wet- slacks” (Smith 2006a). B. compressus has been known in slack 50 since the 1990s and was recorded in slacks 49, 51 and 52 during Smith’s (2006a) study. In slacks 51 and 52 and, to a lesser extent in slack 50, B. compressus is largely confined to lightly-used informal footpaths. This area of dunes was added to the Birkdale Sandhills LNR in 1983. Conservation manage- ment has concentrated on control of H. rhamnoides which spread rapidly through the slacks in the late 1980s and 1990s. From 1995, this invasive shrub was progressively removed by cutting and stump treatment, followed by herbicide spraying of young regrowth. In addition, slack 49 has been mown annually since 1996, with removal of arisings. There is no livestock grazing but rabbits are present in moderate numbers. Royal Birkdale Golf Course was established in the sand-dunes in 1889, the pond on the southern edge of which B. compressus is currently found, has been present since at least 1945 (Smith 1983; 2009). The plant was first recorded here in a permanent 2 m xX 2 m quadrat established in 1998 at site “I” “east of an open water-body in a damp area with lush vegetation”, being still present in 2006 (Butcher ef al. 2006). The damp grassland east of the pond, which supports most of the B. compressus colony, was scrub-cleared in November 1996 and has been managed since by annual mowing in September using a tractor-mounted flail-mower with all arisings removed (C. Whittle in itt. 2009). There is little or no rabbit-grazing as this species is controlled on the course. Recently, P. S. Gateley (in litt. 2009) drew attention to the discovery of B. compressus at another site on Royal Birkdale Golf Course (site “K”, SD31061528), a large wet hollow on the western side of the course between the tee and green of the 12th hole. The plant was identified for the first time in July 2008 in a permanent -quadrat that had been recorded several times since 1998) SusingeeNNie methodology. This site has received the same management as slack “T’. Just to the west of the golf course, Birkdale slack 26 was formed after the above mentioned railway line was built in 1884, Smith (1984) suggesting that it probably dates to about 1929. Aerial photographs show that it was mature and fully vegetated in 1945. B. compressus was first noted here in 1982 during a study of Juncus balticus (Smith 1984). At that time, it was more abundant in the northern part of the slack than at present, the plant now being restricted to lightly- used footpaths through the generally rank vegetation, which is not grazed by rabbits (personal observations). This slack is within the Birkdale Sandhills LNR and was subject to scrub-control in 1987 and 2001 (Smith 2007b). BLYSMUS COMPRESSUS ON THE SEFTON DUNES Also at the northern end of the Birkdale Sandhills is a series of six frontal-dune slacks (nos. 27, 28, 30, 31, 32, 33) all of which have B. compressus colonies. Most of them were dated by Smith (1984) to the early/mid 1970s, though slacks 30 and 31 were thought to be a few years older. B. compressus was originally found here in July 1980, growing in slack 31 with Juncus balticus, etc. (personal observ- ations). During the period 1978-1982, most of these slacks were churned up by illegal motor- cycle scrambling (Smith 1984) but, in 1983, the frontal-dunes were added to the Birkdale Sandhills LNR and wardening introduced to control this activity. Edmondson et al. (1988/89) include Target Notes indicating the occurrence of B. compressus in slacks 27 and 31. By 2003, the plant was present in all six slacks (Smith 2006a). Conservation management of these slacks has mainly concentrated on the control of H. rhamnoides which became increasingly dominant during the 1980s and early 1990s. The slacks were scrub-cleared progressively from no. 27 in 1996 to no. 30 in 2003, with subsequent annual control of regrowth (Smith 2006a; 2007b). All the slacks have well-used informal footpaths, many of which support B. compressus. Slack 32, in particular, is heavily trampled and rabbit-grazed, with a very short sward. LOST SITES Herbarium records suggest that B. compressus was much more widespread on the Merseyside coast in the nineteenth and early twentieth centuries; however, building development in north Liverpool, Bootle, Crosby, Formby, Ainsdale, Birkdale and Southport destroyed about 40% of the dune system (Smith 2009) and no doubt many sites were lost. Within recent memory, only three dune sites for B. compressus are known to have disappeared, either wholly or in large part. One was a small wet-slack at Lifeboat Road, Formby Point (SD277066), created by post-Second World War sand-winning, where the plant was recorded as “abundant” in 1978. This site was soon lost to dense poplar (Populus) scrub (personal observations). The Ainsdale Sand Dunes NNR archive contains a “Rare Plant Species Location Map” drawn up by K. R. Payne in 1976. It records B. compressus in slack 4 (SD284105), part of the Massams Slack complex near the sea, this having been subsequently lost to coastal erosion and sand- blow (A. Kimpton in Jitt. 2008). Finally, the 347 Sands Lake, Ainsdale site (SD302129), described above, has largely been destroyed by vegetation maturation, including the develop- ment of dune-scrub and a tall herbaceous sward. Literature searches following the completion of field work identified two additional sites for B. compressus recorded in Target Notes by Gateley & Michell (2004). These are at slack 8, Altcar Rifle Range (SD285050), and slack 13, Birkdale Sandhills LNR (SD305135), in wet- slack vegetation on the margins of scrapes dug for nature conservation purposes in 1976 and 1977 respectively (Smith 1978). While both these sites have been visited since 2004 for purposes of biological recording, the presence of B. compressus was not suspected and it may therefore have been overlooked. Therefore, the continuing presence of the plant cannot be entirely ruled out. ASSOCIATED VASCULAR PLANTS Ninety vascular taxa were found associated with B. compressus in the 51 quadrats recorded for NVC analysis. These are listed, together with their frequency of occurrence and Ellenberg’s indicator values in Table 2. Equisetum x litorale, with eight occurrences, 1s omitted from the Table as no Ellenberg values are available for this taxon (Hill et al. 2004). The most ubiquitous species (with over 20 occurrences) are Agrostis stolonifera (36 occurrences), Carex flacca (44), Festuca rubra (27), Holcus lanatus (27), Hydrocotyle vulgaris (29), Juncus articulatus (21), Mentha aquatica (30), Rubus caesius (24), Salix repens (35) and Trifolium repens (34). All these taxa are characteristic of dune slacks on the Sefton Coast (Smith 2009; 2006a). Ten of the associates are notable, two (Dactylorhiza incarnata ssp. coccinea and Juncus balticus) being Nationally Scarce and eight (Carex disticha, C. viridula ssp. viridula, Eleocharis quinqueflora, Epipactis palustris, Equisetum variegatum, Parnassia palustris, Trifolium fragiferum and Triglochin palustre) ranked as Species of Conservation Importance in North West England (Regional Biodiversity Steering Group 1999). Table 2 gives Ellenberg’s indicator values for moisture (L), reaction (R) and nitrogen (N) for vascular taxa associated with OB. compressus. These have been weighted to take into account their frequency of occurrence. The weighted mean values (6°51, 6:21 and 4:30 respectively) indicate that the taxa favour constantly moist soils that are weakly acid to 348 P. H. SMITH TABLE 3. TABLEFIT ANALYSIS OF QUADRAT SAMPLES Site (no. of samples) NVC code Community Sub-community % fit Match Altcar (3) MGl la Festuca rubra-Agrostis Lolium perenne >)f/ Poor stolonifera-Potentilla anserina grassland Cabin Hill | SD15 Salix repens-Calliergon 46 Very poor (8) cuspidatum dune slack Cabin Hill 3 SD17 Potentilla anserina-Carex 49 Very poor (3) nigra dune slack Cabin Hill 20 (3) SD15 Salix repens-Calliergon Sil Poor cuspidatum dune slack Sands Lake MGlla Festuca rubra-Agrostis Lolium perenne 52 Poor @) stolonifera-Potentilla anserina grassland Ainsdale 49 SD16d Salix repens-Holcus Agrostis 55 Poor (3) lanatus dune slack stolonifera Ainsdale 50 SD15 Salix repens-Calliergon 47 Very poor (3) cuspidatum dune slack Ainsdale 51 No sample Ainsdale 52 SD16b Salix repens-Holcus Rubus caesius 58 Poor (1) lanatus dune slack Royal Birkdale pond No sample Royal Birkdale fairway MG8 Cynosurus cristatus- 60 Fair (3) Caltha palustris grassland Birkdale 26 (3) SD16b Salix repens-Holcus Rubus caesius 4] Very poor lanatus dune slack Birkdale 27 (3) SD16d Salix repens-Holcus Agrostis Dy Poor lanatus dune slack stolonifera Birkdale 28 (3) SD16d Salix repens-Holcus Agrostis 47 Very poor lanatus dune slack stolonifera Birkdale 30 (3) SD17b Potentilla anserina-Carex — Carex flacca 12) Very poor nigra dune slack Birkdale 31 (3) SD16b Salix repens-Holcus Rubus caesius 54 Poor lanatus dune slack ; or SD16d Salix repens-Holcus Agrostis 54 Poor lanatus dune slack stolonifera Birkdale 32 (3) SD17b Potentilla anserina-Carex ~ Carex flacca 36 Very poor nigra dune slack Birkdale 33 (3) SD16 Salix repens-Holcus 3 Very poor lanatus dune slack weakly basic in reaction and have intermediate fertility (Hill et al. 2004). Recalculation of weighted means, for ubiquitous taxa only, gives closely similar values (6:41, 6:29 and 4-36). NVC COMMUNITIES Table 3 lists TABLEFIT results for 51 quadrats supporting B. compressus. In almost all cases, the statistical fits to known NVC communities are “poor’ to “very poor’, only one (Royal Birkdale Golf Course) having a “fair” level of agreement with MG8 (Cynosurus cristatus- Caltha palustris grassland). As expected, vegetation at most sites equates to dune-slack communities, seven being SD16 (Salix repens-Holcus lanatus dune-slack), three SD15 (Salix repens-Calliergon cuspidatum dune-slack) and three SDI17_ (Potentilla anserina-Carex nigra dune-slack). The Salix- Holcus community is characteristic of older and dryer calcareous dune slacks, rarely flooded to any great extent, and often subject to grazing by livestock and/or rabbits. Two sub- communities, SD16d (Agrostis stolonifera sub- community) and SD16b (Rubus caesius sub- community) were indicated by the analysis, the BLYSMUS COMPRESSUS ON THE SEFTON DUNES 349 former being found in wetter conditions (Gateley & Michell 2004; Rodwell 2000). SD1I5 also occurs in older slacks but is associated with more prolonged flooding. As in the case of SD16, it is widely distributed nationally and is frequent on the Sefton Coast (Gateley & Michell 2004; Rodwell 2000). Being especially associated with northern dune systems, SD17 is found in damp or wet slacks in which the water is less base-rich, perhaps due to accumulation of organic matter. This slack type is relatively rare on the Sefton duneland (Gateley & Michell 2004; Rodwell 2000). Three sites apparently support mesotrophic grassland communities, two MGI11 (Festuca rubra-Agrostis stolonifera-Potentilla anserina grassland) (at Altcar Rifle Range and Sands Lake) and one MG8 (Cynosurus_ cristatus- Caltha palustris grassland) (at Royal Birkdale Golf Course). A rare community on the Sefton sand-dunes, MGI11 is a damp grassland type that seems to take the place of slack vegetation in a few disturbed and nutrient rich areas (Gateley & Michell 2004). MG8 is species-rich vegetation characteristic of periodically inundated land, usually managed as pasture and often associated with traditional water- meadows (Rodwell 1992). This community seems out of place here and was not recorded during Gateley & Michell’s (2004) NVC survey of the Sefton dunes. Indeed, the Royal Birkdale site was recorded as the rather similar MGS (Cynosurus cristatus-Centaurea nigra grassland) or MG5a (Lathyrus pratensis sub- community) by Butcher et al. (2006). This is a typical grassland of traditionally grazed hay- meadows on heavy textured lowland soils (Rodwell 1992) and is also an_ unlikely community to find on sand-dunes. DISCUSSION It seems clear that the status of B. compressus on the Sefton sand-dunes has changed in recent decades. It was recorded in 18 sites during the present study, having been known from under 10 sites in the early 1980s. This is, of course, a species that is easily overlooked but the fact that Smith (2006a) found an increase in occurrences from two to ten dune-slacks in careful studies at Birkdale over a 20-year period suggests that this apparent change is genuine. A continuing increase is suggested by the discovery of B. compressus for the first time in 2008 in a permanent quadrat on Royal Birkdale Golf Course that has been recorded regularly since 1998. In view of the plant’s rapid national decline, these are encouraging findings. Furthermore, only two known sites have been completely lost since the 1970s, one to coastal erosion and another to scrub development, while another (at Sands Lake) has shown a major decline in B. compressus due to maturation of the vegetation. Most sites for, and the largest colonies of, B. compressus are in calcareous wet-slacks (sensu Ranwell), soils being gleys or peaty gleys with a fairly high pH. This accords with habitat descriptions in the rather limited literature on the ecology of this species (e.g. Burkmar 2008; Foley & Porter 2002; Halliday 1997; Walker 2008). Marshy, seasonally-flooded, base-rich sites are said to be favoured, conditions which apply in these slacks. However, small to medium-sized populations were also found in dry-slacks and damp _ grassland, most surprisingly in the heavily disturbed, weedy fringe of a carpark at Sands Lake, Ainsdale, suggesting that the plant can persist for some time under sub-optimal conditions. There is a tendency for B. compressus populations to be associated with informal footpaths through slack basins, where vegetation height is much reduced, perhaps as a consequence of mechanical damage and, probably, enhanced rabbit-grazing. Plant composition may also be altered; for example, S. repens is often reduced in frequency or eliminated from such footpaths, a fact that may well be important for the survival of B. compressus, since dominance by S. repens is considered to reduce plant species richness (Edmondson et al. 1993b). These findings also imply that B. compressus is tolerant of grazing and trampling damage. A majority of the Sefton Coast sites for B. compressus appears to be of recent origin. This applies particularly to the wetland supporting the largest population at Cabin Hill and the slacks in the Ainsdale and Birkdale frontal dunes, all of which are less than 40 years old (Table 1). In some cases, it has been possible to date the first appearance of B. compressus to about 10 years after slack formation. There are a few ostensibly older sites but here the plant is favoured either by human trampling, grazing or mowing, these maintaining a shorter, more open sward. This supports the suggestion that the plant is a poor competitor (Walker 2008) and is therefore likely to decline, in the absence 350 of management, as slacks age and become colonised by ranker vegetation (Davy et al. 2006; Smith 2009). It is interesting that B. compressus survived and even increased in the northern Birkdale frontal slacks after these were severely damaged by motor-cycling in the late 1970s and early 1980s; this activity created large areas of bare ground and provided sites for colonisation, not only by B. compressus but also Juncus balticus at what was, by then, its only English location. Smith (1984; 2006b) concludes that J. balticus is a good coloniser of young, sparsely vegetated wet-slacks and may then persist for many years before declining as the habitat becomes dryer and more heavily vegetated. These conclusions seem also to apply to B. compressus. Although, B. compressus is currently well- established and locally abundant on the Sefton sand-dunes, its populations are highly localised, confined to only three tetrads and therefore vulnerable. Bearing in mind that this dune system is particularly well-endowed with wet-slacks (Smith 1978) and supports about 39% of all the dune-slack vegetation in England (Radley 1994), it is perhaps surprising that the plant is not more widespread here. Thus, the large expenses of Ainsdale Sand Dunes NNR (339ha) and Ainsdale Sandhills LNR (88ha), both with extensive slack habitat, appear to be devoid of B. compressus populations. However, many of the large slacks in these areas originated over a century ago (Smith 1978) and are densely clothed in S. repens. Such habitat seems to be unsuitable for B. compressus which is mainly found in young and/or disturbed sites. B. compressus grows here with a high diversity of associated vascular plants (90 taxa), many being characteristic of wet-slacks. However, analysis of Ellenberg’s indicator values shows that, on average, the associates are adapted to slightly dryer, less basic and more fertile sites that B. compressus. This suggests that requirements for the latter species may be more taxing than for other wet-slack vascular plants, a finding which may also help to account for its restricted distribution in the dune system. B. compressus is not mentioned by Rodwell (1991; 1992; 2000) and there is little published information on plant communities favoured by this species, though Walker (2008) gives some examples. Thus, in vice counties 80 (Roxburghs.) and 81 (Berwicks.), most colonies occur in a distinct habitat of short turf on a shelf below the level of stream banks. Here typical P. H. SMITH associates include Agrostis capillaris, Anthoxanthum odoratum, Cirsium palustre, Cynosurus cristatus, Eleocharis palustris, Festuca rubra, Bellis perennis, Geum rivale, Lychnis flos-cuculi, Poa sp., Ranunculus repens and Trifolium repens; nine of these also being associated with B. compressus in Sefton. A typical NVC community of B. compressus sites in the uplands of northern England is M10 (Carex dioica-Pinguicula vulgaris mire) (Walker 2008), a calcareous flush vegetation in which small sedges predominate and which is often maintained in an open condition by grazing and trampling of large herbivores (Rodwell 1991). In southern and eastern England, B. compressus occurs in species-rich fen meadows, including MG8 (Cynosurus cristatus-Caltha palustris grassland) and M23 (Juncus effusus/acutiflorus-Galium palustre rush-pasture) (Walker 2008). The former is found on seasonally-flooded, flat or slightly sloping land by rivers and streams and was managed traditionally as pasture for cattle and horses, sometimes as water-meadow (Rodwell 1992). MG8 was unexpectedly recorded at the Royal Birkdale site during the present study. The Juncus-Galium community is especially characteristic of ill-drained and relatively unimproved or reverted pasture in the cool and rainy lowlands of western Britain; here it is maintained by grazing or, much more rarely, by mowing (Rodwell 1991). Typical associates in both these types include Carex acutiformis, C. flacca, C. disticha, C. panicea, Cirsium dissectum,. Dactylorhiza. incarnata, D. praetermissa, Eleocharis uniglumis, Filipendula ulmaria, Lychnis flos-cuculi and Persicaria hydropiper (Walker 2008), five of these being listed as associates of DB. compressus on the Sefton Coast (Table 2). Almost all the samples taken during this study are poor or very poor fits to known NVC communities. Similar results were reported for several of the Birkdale sites by Smith (2006b). Samples taken in 1981/82 in slacks 26, 27, 30, 31 and 33 show mostly poor to very poor levels of fit, while those recorded in 2004 for slack 27 range from very poor to fair. Similarly, Butcher et al. (2006) report very poor to fair fits to MGS at the Royal Birkdale site (their slack “T’) between 1998 and 2006. A possible explan- ation is that the vegetation of these and most other Sefton sites for B. compressus has a recent origin and has been disturbed by several factors, including recreational activities, grazing, mowing and scrub removal. This might result in samples failing to match known BLYSMUS COMPRESSUS ON THE SEFTON DUNES 351 NVC communities because the vegetation is still maturing with respect to species ingress. However, low matches are likely to be as spurious as useful (P. S. Gateley in litt. 2009). For example, slack “I at Royal Birkdale is clearly neither a traditional water-meadow nor a hay-meadow but a form of dune-slack, albeit somewhat atypical. CONSERVATION All current sites for B. compressus on the Sefton Coast are protected by SSSI/Natura 2000 designations. Those at Cabin Hill lie within a National Nature Reserve, while apart from Royal Birkdale Golf Course, the Ainsdale and Birkdale sites benefit from Local Nature Reserve status. The plant is proposed for incl- usion in the North Merseyside BAP. The present study reinforces opinion (e.g. Burkmar 2007) that habitat management favouring the plant will include maintaining the hydrology of sites, avoiding eutrophication and preventing the over-maturation and coarsening of swards. The fact that the plant seems to have increased recently suggests that current Sefton Coast conservation management, including late-summer mowing (Altcar Rifle Range, Ainsdale slack 49 and Royal Birkdale), livestock-grazing (Cabin Hill) and scrub control (most sites) is being successful. Additionally, light to moderate recreational trampling and rabbit-grazing seem to be important in helping to maintain open swards and reduce dominance of slack vegetation by S. repens. Scrub control on the duneland has concentrated mainly on H. rhamnoides but attention may also need to be given to the larger willows, such as S. cinerea, which are slowly colonising some of the B. compressus sites. Long-term studies of Sefton dune hydrology provide no convincing evidence of water-table decline, though modelling the impact of climate change suggests adverse impacts could be seen within a few decades (Clarke & Sanitwong-Na- Ayutthaya 2006). Also of concern is eutrophic- ation by aerial deposition of nitrogen. Gateley & Michell (2004) report recent reductions in the areas of oligotrophic vegetation types on the Sefton duneland, implying enrichment is taking place, perhaps partly from this source. Over the last 30-40 years, many British dune systems, including the Sefton Coast, have become more stable and heavily vegetated. Meanwhile, atmospheric N_ deposition in Europe has approximately doubled (Jones ef al. 2004). The latter authors show a _ positive relationship between N inputs and above- ground biomass and a negative relationship with plant species-richness in mobile and fixed dunes, though the trends are less clear in dune- slacks. They suggest a critical load range of 10-20 kgNha’'yr™ for coastal dunes in the UK. As B. compressus seems to be largely absent from mature dune-slacks, management that promotes dune dynamics, including dune formation and localised blow-out development will assist the creation of suitable new habitat in the medium to long-term. The recent development of Birkdale Green Beach, a 60ha strip of salt-marsh, sand-dune, dune-slack and swamp communities on the foreshore between Birkdale and Ainsdale (Smith 2007c), provides potential future habitat for B. compressus, although only the related B. rufus has been recorded there to date. ACKNOWLEDGMENTS I am most grateful to Patricia Lockwood for help with field work and to Pauline Michell for assistance with TABLEFIT analysis. Thanks are also due to Michelle Newton who prepared a site map using the Sefton Coast GIS. Dave Earl kindly provided information from the New Flora of South Lancashire Project and Eric Greenwood from the Liverpool Museum herbarium. Natural England, Royal Birkdale Golf Club and Altcar Rifle Range authorities gave permission for access to their various land holdings, while Chris Whittle, kindly supplied details on the conservation management of slacks at Royal Birkdale. Mike Wilcox and Kevin Walker made helpful and constructive comments on a draft of the manuscript. REFERENCES BURKMAR, R. (2008). Blysmus compressus (Flat Sedge). Management Advice Note, Rare Vascular Plant Group. Wildlife Trust for Lancashire, Manchester and North Merseyside, Preston. BUTCHER, D., GATELEY, P., NEWTON, M. & SINNOTT, D. (2006). Royal Birkdale Golf Club, Survey of Dune Slacks: Vegetation and Topography, report on research completed in 2006. Department of Environmental Management, University of Central Lancashire, Preston. CHEFFINGS, C. M. & FARRELL, L. eds. (2005). The Vascular Plant Red Data List for Great Britain. Joint Nature Conservation Committee, Peterborough. 352 P. H. SMITH CLARKE, D. & SANITWONG-NA-AYUTTHAYA, S. (2006). Water balance at Ainsdale National Nature Reserve. Report of the Sefton Coast Partnership Research Meeting 14th July 2006, Edge Hill University College, Ormskirk, pp. 30-35. Sefton Council, Ainsdale. Cook, A. L. M. (1989). Altcar: the story of a rifle range. Territorial, Auxiliary & Volunteer Association for the North West of England & the Isle of Man. HMSO, London. CrosBy, A. G. (2007). The impact of sand extraction at Formby Point, Merseyside. Technical Services, Sefton Council, Bootle. DAvy, A. J., GROOTJANS, A. P., HISCOCK, K. & PETERSEN, J. (2006). Development of eco-hydrological guidelines for dune habitats — Phase 1. English Nature, Peterborough. DICKINSON, J. (1851). The Flora of Liverpool. Deighton & Laughton, Liverpool. EDMONDSON, S. E., EDMONDSON, M. R., GATELEY, P. S., ROONEY P. J. & SMITH, P. H. (1993a). Flowering plants (Angiospermae), in D. ATKINSON & J. HOUSTON eds. The Sand Dunes of the Sefton Coast, pp. 109-118. National Museums & Galleries on Merseyside, Liverpool. EDMONDSON, S. E., GATELEY, P. S., ROONEY, P. & STURGESS, P. W. (1993b). Plant Communities and Succession, in D. ATKINSON & J. HOUSTON eds. The Sand Dunes of the Sefton Coast, pp.65—84. National Museums & Galleries on Merseyside, Liverpool. EDMONDSON, S. E., GATELEY, P. S. & NISSENBAUM, D. A. (1988/89). National Sand Dune Vegetation Survey: the Sefton Coast. Report no. 917. Nature Conservancy Council, Peterborough. FOLEY, M. J. Y. & PORTER, M. S. (2002). Blysmus compressus Flat-sedge, in C.D. PRESTON, D. A. PEARMAN, & T. D. DINES eds. New Atlas of the British and Irish Flora, p. 709. Oxford University Press, Oxford. GATELEY, P. S. & MICHELL, P. E. (2004). Sand Dune Survey of the Sefton Coast. Report by TEP, Warrington to Sefton Metropolitan Borough Council, Bootle. GREENWOOD, E. F. (2002). Flat-sedge (Blysmus compressus). Lancashire BAP Species Action Plan (revised 2007). Lancashire Biodiversity Partnership, Wildlife Trust for Lancashire, Manchester & North Merseyside, Preston. HALL, T. B. (1838). The Flora of Liverpool. Whitaker & Co., London. HALLIDAY, G. (1997). A Flora of Cumbria. Centre for North-West Regional Studies, University of Lancaster. HILL, M. O. (1996). TABLEFIT V1.0, for identification of vegetation types. Institute of Terrestrial Ecology, Huntingdon. HILL, M. O., PRESTON, C.D. & Roy, D. B. (2004). PLANTATT. Attributes of British and Irish Plants: Status, Size, Life History, Geography and Habitats. Centre for Ecology & Hydrology, Monks Wood, Cambridge. JAMES, P. A. (1993). Soils and nutrient cycling, in D. ATKINSON & J. HOUSTON eds. The Sand Dunes of the Sefton Coast, pp.47—54. National Museums & Galleries on Merseyside, Liverpool. JERMY, A. C., SIMPSON, D. A., FOLEY, M. J. Y. & PORTER, M. S. (2007). Sedges of the British Isles. BSBI Handbook no. 1, edition 3. Botanical Society of the British Isles, London. JONES, M. L. M., WALLACE, H. L., NORRIS, D., BRITTAIN, S. A., HARIA, S., JONES, R. E., RHIND, P. M.., REYNOLDS, B. R. & EMMETT, B. A. (2004). Changes in vegetation and soil characteristics in coastal sand dunes along a gradient of atmospheric nitrogen deposition. Plant Biology 6: 598-605. LIVERMORE, L. A. & LIVERMORE, P. D. (1987). The Flowering Plants and Ferns of North Lancashire. Privately published, Preston. MCNICHOLL, E. D. ed. (1883). Handbook for Southport. Robert Johnson & Co., Southport. NEWTON, A. (1971). Flora of Cheshire. Cheshire Community Council, Chester. PAYNE, K. R. (1982). Ainsdale Sand Dunes National Nature Reserve Vascular Plant Species-list (1982 revision). Natural England, Ainsdale. RADLEY, G. P. (1994). Sand dune vegetation survey of Great Britain: Part 1 — England. Joint Nature Conservation Committee, Peterborough. RANWELL, D. S. (1972). Ecology of Salt Marshes and Sand Dunes. Chapman & Hall, London. REGIONAL BIODIVERSITY STEERING GROUP (1999). A Biodiversity Audit of North West England. 2 vols. Environmental Advisory Service, Maghull. RODWELL, J. S. ed. (1991). British Plant Communities Volume 2. Mires and heaths. Cambridge University Press, Cambridge. RODWELL, J. S. ed. (1992). British Plant Communities Volume 3. Grasslands and montane communities. Cambridge University Press, Cambridge. RODWELL, J. S. ed. (2000). British Plant Communities Volume 5. Maritime communities and vegetation of open habitats. Cambridge University Press, Cambridge. SAVIDGE, J. P., HEYwWoop, V. H. & GORDON, V. eds. (1963). Travis’s Flora of South Lancashire. Liverpool Botanical Society, Liverpool. SKELCHER, G. (2008). Fylde Sand Dunes Management Action Plan. A report to the Fylde Sand Dune Project Steering Group. Wildlife Trust for Lancashire, Manchester & North Merseyside, Preston. SMITH, P. H. (1978). The Ecological Evaluation of Wetland Habitats in the North- Merseyside Sand-dune System. Merseyside County Council, Liverpool. BLYSMUS COMPRESSUS ON THE SEFTON DUNES 353 SMITH, P. H. (1983). Ecology and Management of Birkdale Frontal Dunes, Merseyside. Sefton Metropolitan Borough Council, Bootle. SMITH, P. H. (1984). The distribution, status and conservation of Juncus balticus Willd. in England. Watsonia 15; 15-26. SMITH, P. H. (2006a). Changes in the floristic composition of sand-dune slacks over a twenty year period. Watsonia 26: 41-49. SMITH, P. H. (2006b). Revisiting Juncus balticus Willd. in England. Watsonia 26: 57— 65. SMITH, P. H. (2007a). Assessment of ecological impacts of sand extraction at Formby Point and opportunities for habitat maintenance, restoration or creation linked to targets within the North Merseyside Biodiversity Action Plan. Technical Services, Sefton Council, Bootle. SMITH, P. H. (2007b). An analysis of current information on the dune wetlands of the Sefton Coast. Sefton Coast Partnership, Sefton Council, Bootle. SMITH, P. H. (2007c). The Birkdale Green Beach — a sand-dune biodiversity hotspot. British Wildlife 19 (1): 11-16. SMITH, P.H. (2009). The Sands of Time Revisited. An Introduction to the Sand Dunes of the Sefton Coast. Amberley Publishing, Stroud, Gloucestershire. WALKER, K. (2008). Blysmus compressus (L.) Panzer ex Link. BSBI Species Accounts. Botanical Society of the British Isles, London. (Accepted May 2009) ——— ~-” weve eG Webic as en ee meets wen i _ rive t na eh toy 779) ‘ewe } ope. iy dy a rat ac a aaa We rie va) ved é ; - e-umee: ye net Se ee ee =. Ls ane a ere phn! th anasto (maibgalt es AAT vette Albee “1 1) ure O27 ee Ye qeeue hTk oe ae pts ie oo ay (ie hinge alga one neuLdangn rv tal : + ’ stems eT Bey) eee ae erred ganleonk “soll ee - ae. 1 aceite lode es ie seein aerial one ei » eye Wf, th ety othe eee libri, 4 my wee i? lalate 3 wT ye it | - Sianas W, 7)0Rbi hie Sate thn tide rh: aged app ere a = © - wee een E * bs } Pa v =(\fepepasqht Laura, animes — aid Horr Yt | patie eae rar. 7 oe ey, elena tal : ae on - Watsonia 27: 355—364 (2009) 3)3)5) Some localised early plant records from North-west England: then and now Wile Je Me IROMI ENC Faraday Building, Department of Biological Sciences, University of Lancaster, Lancaster LAI 4YA ABSTRACT The plant records of three late sixteenth/early seventeenth-century collectors made in two localised areas of north-west England are examined and the relevant plants’ continued survival there, or other- wise, is discussed. Apart from what at that time were already scarce plants or were aesthetically attractive and therefore over-collected, most of them still survive. KEYWORDS: Craven limestones, Great Harwood, botanical history, Thomas Hesketh, Thomas Penny, Thomasin Tunstall. INTRODUCTION In the late sixteenth and early seventeenth century, plants growing locally in north-west England were recorded in contemporary herbals, especially those prepared and published by John Gerard (1545-1612), John Parkinson (1567-1650) and Charles L’Ecluse (Clusius) (1526-1609). The latter were com- pilers of information but the detail came from local botanists who variously sent them specimens, drawings, descriptions, and locality details of any plants they considered interest- ing. In this particular area of the country at that time, the three most important and prolific were Thomas Penny (c. 1530-1588), Thomas Hesketh (1560-1613) and Thomasin Tunstall (fl. 1620- 1630) but, understandably, their areas of activity were usually close to their respective homes. Two limited geographical areas in which these three were active are considered here. The first is that part of the botanically-rich Craven limestone area (v.c. 64) which includes Ingleborough hill and the woods and ghylls on its south-west side close to the villages of Ingleton and Clapham (grid _ reference: SD67/77). This region was visited at different times by all three botanists. The second area appears to have been the sole preserve of Hesketh. This was close to his home near Great Harwood, Lancashire (v.c. 59), and included the nearby region around the confluences of the *email: m.foley @lancaster.ac.uk rivers Ribble, Calder and Hodder (v.cc. 59/60; SD63/73). Additional scattered records by Hesketh from elsewhere in northern England are also known including the west Lancashire (v.c. 59) moss-land near Rufford and Mawd- esley (SD41) where he owned a_ second property, as well as Rosgill, Crosby Ravens- worth and Walney Island (the latter three, all in Cumbria, v.cc. 69 & 70). These dispersed records are not discussed here. THE LOCAL BOTANISTS Thomas Penny was born in the small Lancashire village of Gressingham north-east of Lancaster within easy reach of Ingleborough hill. He apparently spent his early years in the area but, after graduating at Cambridge University, took up holy orders and moved to London. Later, between 1565 and 1569, he travelled on the continent, mainly to avoid the religious hostility present in England at that time and whilst abroad spent most of this period pursuing field botany and medical training. Despite having moved away from northern England, he nevertheless retained close links with his birthplace, returning there on occasion and visiting the Ingleborough area. In his will, he made a bequest to the poor of Gressingham “to be distributed at the discretion of Mr Fawcett minister...... cf Inetstiohnes Church, Gressingham, there is, to this day, a plaque listing previous ministers which in- cludes the said John Fawcett, the incumbent during the period 1562—1590. Penny’s interest in botany presumably originated from these early years in north Lancashire whereon some of his local finds were recorded by the contemporary herbal authors of the day, especially Gerard and Clusius. Later, during his time on the continent, he became an expert on the flora of the region around Geneva and was known to other eminent naturalists such as Conrad Gesner, Matthew de Jl Obel and Joachim Camerarius. On his return to England, he took up medical practice and also became an 356 ME NEO ENG eminent entomologist. He was a co-author, along with Thomas Mouffett and others, of the posthumously-published Jnsectorum — sive minimorum Animalium Theatrum (Mouffett 1634), a very early work on entomology. An account of Penny’s botanical work abroad together with additional details of his background, are given in Foley (2006). By coincidence, another person with a great interest in plants, but born rather later than Penny, lived within a mere five kilometres of Penny’s Gressingham birthplace. This was a lady, Thomasin Tunstall, whose plant collecting virtues were extolled by John Parkinson, herbalist to James I. In his book Paradisi in sole paradisus terrestris Parkinson (1629) described her as “a courteous Gentle- woman...... who dwelleth at Bull-banke, neare Hornby castle in those parts...”. Bull Bank, although now rebuilt, still exists as a farm, situated close to the River Greta, and with the village of Tunstall (as in her name) and Thurland Castle, little more than a kilometre away. The castle was once owned by the Tunstall family and it is possible that she was born there. An earlier Thomasin Tunstall, born around 1544, was the daughter of the then owner Marmaduke Tunstall and could have been Thomasin’s mother or grandmother, whilst Francis Tunstall, who sold Thurland Castle around 1605, was of an age to have possibly been her father. Little else is known of her background but there is a record in St John’s Parish Church register in the village of Tunstall, dated September 29, 1632, which may relate to her. This lists the marriage between Thomasin Smythies de Tunstall and Thomas Smyth. Also, Farrer & Brownbill (1914) state that, in 1629, Thomasin Tunstall and Alice Clopton, both of this area, compounded for two thirds of their estates liable to sequestration for recusancy and were each fined £2 10s. a year. This also may refer to her. She was apparently very familiar with the Ingleborough area (as was Penny) and sent plants collected there to John Parkinson in London. The third person under consideration 1s Thomas Hesketh who was born at Martholme Hall, Great Harwood, in 1560. An extant record in the local parish register for 15 May of that year records his baptism: Thomas Hesketh filius secundus Thome Hesketh militis (i.e. second son of the knight, also called Thomas Hesketh). On his father’s death in 1588, Thomas was left property at the village of Mawdesley in west Lancashire. Until that time, his early life would have been spent at Martholme, but he now moved to his inherited property whilst his widowed mother continued to live on at Martholme Hall. However, in 1593 he returned there to live with his mother again. This is confirmed by an entry into the Great Harwood parish records for October that year because, as a non-conforming catholic, he was considered to be a recusant, a penalty for which was the necessity to register his current place of residence. This entry, also extant, states: Thomas Hesketh gent a Recusant did /notifie/ his Cominge to the martholme to dwele (wth his mother/ to me Wm Herris Cur’ of the Chappell of moch Harwood the fowrte day of october A°® Regni Elizabeth nunc Regine xxxv° 1593. So, for much of his life prior to the publication of Gerard’s “Herball” (Gerard 1597), he lived at Martholme making excursions into the surrounding countryside as well as to the Ingleborough area. Around the turn of the century, he moved to nearby Clitheroe where he practised as a doctor, Parkinson (1640) referring to him as “a painefull Chirurgion [surgeon] and Simplist [plant collector] of Lancashire”. He was held in high regard by Gerard who considered him “a _ diligent searcher of Simples [plants]” who had dis- covered many plants “likewise, never before his time remembred or found out’. Also that he was “a curious [enquiring] gentleman in the searching forth of Simples ... [and] ...in the knowledge of plants”. It is quite possible that Gerard, prior to the publication of his Herbal in 1597, visited Hesketh at Martholme and met his mother (Lady Alice) as he refers to Martholme as “my Lady Heskiths house” (see Tragopogon porrifolius below); they may also have made an excursion together to Walney Island (Barrow-in-Furness, Cumbria). Matthias de L’Obel (1538=1616) ithe rekinees ‘botanographer’, also recorded some _ of Hesketh’s finds and Hesketh must have met both Gerard (see WNarthecium ossifragum, below) and de L’Obel (see L’Obel 1655) in London since both of them record his having brought plants to them. (A) THE CRAVEN LIMESTONES/INGLEBOROUGH AREA (V.C. 64) At the time of these early records, this area was variously considered to be either in Yorkshire or in Lancashire and Gerard, especially, was EARLY RECORDS FROM NORTH-WEST ENGLAND 3)5)// notoriously inconsistent in regards to geo- graphy. However, nowadays this area is firmly within the former county. CARDAMINE PRATENSIS L. Thomasin Tunstall sent a double-flowered plant of this to John Parkinson in London (Parkinson 1629: 389): “Cardamine flore pleno. Double Cuckow flowers .... in Lancashire, from whence I received a plant, which perished, but was found by the industrie of a worthy Gentlewoman, dwelling in those parts heretofore remembred, called M*istresse Thomasin Tunstall, a great lover of these delights”. C. pratensis is still very frequent here in damp grassland and elsewhere in the north although, understandably, the double variety appears to be ignored nowadays. CARDUUS HETEROPHYLLUM (L.) HILL This thistle, of quite striking appearance, was known to, and had been sketched by, Penny prior to 1581 where he must have seen it in the fields around Ingleborough. Clusius (1583: 655-657) stated “... cuius descriptionem & iconem mihi anno 1581. Londini communicavit C.V. Thomas Pennaeus Londin-ensis medicus”...."in pratis ad radices montis Englebrow totius Angliae celsissimi in Comitatu Eboracensi. Floret Iulio & Augusto”. It is still a frequent plant in these parts. CEPHALANTHERA LONGIFOLIA (L.) FRITSCH Parkinson (1629: 348) stated that “The second [i.e. C. longifolia] growth in many places of England, and with the same Gentlewoman [i.e. Thomasin Tunstall] also before remembred, who sent one plant of this kinde [this would be from Helks Wood, Ingleton] with the others [Cypripedium calceolus, see below]”. However, due to extravagant over-collecting, it appears to have met the same fate there as did the Cypripedium. In the early days it was also known from other woods within the area around Settle, Clapham and Ingleton and there is a localised reference to its presence near Brackenbrow, Ingleton in the eighteenth century. It appears to have persisted at the latter for some time and was seen there by F A Lees in 1869 (Lees 1888). However, no further records have been traced, from either there or elsewhere in this area. More recently it was known until the 1930s from a hazel coppice at Far Arnside (v.c. 69) but this locality is c. 25 kilometres to the west of the Helks Wood and the other Craven sites. The nearest occurrence for C. longifolia nowadays lies well to the north in Cumbria (v.c. 69/70) where it occurs at a few sites in open Fagus woodland COCHLEARIA PYRENAICA DC This plant of damp, bareish places on upland limestone was sent to Parkinson by Tunstall (Parkinson (1640: 286): “It has also been found growing upon Ingleborough hill in Lancashire, assured me by a worthy Gentlewoman Mrs Thomasin Tunstall remembred in my former booke for many things, found growing in those parts by her meanes which were not known to be growing in England, and thereof shee sent to me up some for a manifestation of the truth.”. It is still widespread here and common in suitable habitats. CYPRIPEDIUM CALCEOLUS Parkinson (1629: 348) recorded that the plant *...groweth likewise in Lancashire, neare upon the border of Yorkeshire, in a wood or plece called the Helkes [Helks Wood], which is three miles from Ingleborough, the highest hill in England, and not farre from Ingleton, as I am informed by a courteous Gentlewoman, a great lover of these delights, Mistris Thomasin Tunstall, ...and who hath often sent mee up the rootes to London, which have borne faire flowers in my Garden”. Nowadays long since extinct both at Ingleton and _ virtually everywhere else in England, it has been the victim of deplorable over-collection, especially in those early days. Its spectacular appearance and value as a garden plant meant that it was highly coveted. A letter noted by Curtis (1941) which was written in 1781 mentioned a man having brought about forty wild plants to sell at Settle market; plants also were on sale at nearby Ingleton. It appears to have survived at Helks Wood until the late-eighteenth century. Withering writing in 1798 quoted a Mr Thornbeck, an Ingleton surgeon, who told him that it had been “lost in Helks Wood for some years” and Dawson Turner in 1805 recorded that he had searched there for it in vain, “a gardener of Ingleton having eradicated every plant for sale” (see also Lees 1888). It lingered on in a few other parts of Craven into the twentieth century but was eventually reduced to a single, currently surviving, plant. However, re-introductions have recently taken place at some former sites and so eventually it may again become a more familiar sight. PINGUICULA VULGARIS L. Gerard (1597: 645) recorded “Pinguicula sive 66 Sanicula Eboracensis — Butterwort” in “...a 358 M. J. Y. FOLEY field called Crag-Close, and upon Ingleborow fels.....”. The Crag-Close locality is thought to be in the Settle area where Hesketh is known to have operated (see Primula vulgaris x P. veris, below). This will be a Hesketh record. Rocky ground, of which there is a great deal in the area, is sometimes referred to as a “close”. The Ingleborough record, however, is probably Penny’s. The plant still survives in many such localities in the area. POLYGONATUM ODORATUM (MILLER) DRUCE Unlocalised, but this must be a Hesketh record from Clapham since it was from precisely these woods that he also recorded the hybrid Primula (see below), Gerard (1597: 758) stating: “that sort.....with broad leaves groweth in certaine woods in Yorkshire called Clapdale woods, three miles from a village named Settle”. P. odoratum still occurs on nearby scars from Where it was recorded quite recently (P. Abbott, pers. comm., 2008). PRIMULA FARINOSA L. On a return visit to the area in 1581, this was recorded by Penny as “Primula veris flore rubro”, Clusius (1583: 342) stating: “Anno verO 1581, dum Londini essem, de C.V.D. Pennaeo intellexi copiosé admodum nasci in Angliae Septentrionalibus locis pratis humidis & pascuis, atq;”. Although only localised to northern England, this will be from this area of Craven where, even nowadays, it is still frequent in damp areas over limestone. Hesketh also knew it near to his Lancashire home (see below). PRIMULA VULGARIS HUDSON x P. VERIS L. Recorded by Hesketh (Gerard 1597: 637): Primula veris Heskethi — Mr _ Hesketh’s Primrose. “the Primrose found by Mr Hesketh growes in a wood called Clap-dale [near Clapham], three miles from a towne in Yorkshire called Settle”. It appears to be this hybrid and is the first British record. A later comment in the appendix to Johnson’s revision (Gerard 1633) seems to suggest a cultivar: “if Mr Hesket found it there, it was some extraordinary luxurious floure, for now I am well assured no such is there to be seene’”’. The hybrid often occurs where the parents grow close together and is not uncommon in the area and is probably still present at Clapham. Hesketh’s record is also acknowledged by de L’Obel (1655: 118). RUBUS CHAMAEMORUS L. Recorded by both Clusius and Gerard, although that of the former is almost certainly the earlier of the two. Clusius attributed the discovery to Penny who had sent him a description and an excellent drawing, both of which he reproduced in his book. He stated (Clusius 1583: [17=119): > De” ‘Chamaemonros eee iconem & descriptionem nobis communicavit c.v. Thomas Pennaeus supra memoratus, atque Chamaemorum appellare placuit........... [tee aperta loca amat, summosque montes, & magna copia inter Ericas provenit monte Ingleborrow totius Angliae celsissimo, duodecim miliaribus 4 Lancastria. Later, Gerard (1597:1386) under Vaccinia nubis (Cloud-berrie) said it had been found on two high mountains, one “in Yorkshire called Ingleborough, the other in Lancashire called Pendle....”, the latter locality being referred to Hesketh as the finder (see below). R. chamaemorus still occurs on Ingleborough and other high hills within the area. SEDUM ROSEA (L.) SCOP. As with Rubus chamaemorus, this was first found first by Penny who had sent it to Joachim Camerarius prior to 1588. He (Camerarius 1588: 139) recorded it as “Rhodia radix” as found “in Anglia in monte Engleborreno [Ingleborough] dicto, ...... ubi flores & semen producit, ut ad me scripsit Doctiss. & rerum naturalium peritiss. Medicus D. Thomas Pennaeus Anglus”. Gerard (1633: 532) also noted that, Rhodiagiradixn “sroweth very plentifully in the North part of England, especially in a _ place called Ingleborough Fels”, whilst Parkinson (1640: 739-740) recorded what appears to be the same plant (Sedum laciniatus foliis Small Houseleek with divided leaves) “on the Mountaines of Lancashire with us as Mr Hesketh told us”. However, there is no precise reference to Hesketh having found it on Ingleborough. Sedum rosea was reported as rare in Yorkshire by Lees (1888), occurring on high exposed grit or slate rocks. However, it is still known from Ingleborough from where Penny certainly knew it. OTHER POSSIBLE RECORDS Clusius” (1601: 237) “recordedizoliins europaeus L. [as Trollius flos] in mountains in northern England: “ex septentrionalibus Angliae montibus erutum recenter Londini EARLY RECORDS FROM NORTH-WEST ENGLAND 359 anno octuagesimo primo supra millesimu & quingentesimus conspiciebam” but without a finder. Even earlier he (Clusius 1583: 372), using the name ““Trollius flos”, had commented on it having been known in England growing amongst Primula farinosa: “...in Anglia inter primulas flore rubro”. Knowing of the Penny- Clusius connection in relation to P. farinosa and other plants, these references very probably originate from Penny and from this area where it still survives in small populations. Gerard also gave a few additional northern England plant records, but without direct attribution but these almost certainly will have originated from Hesketh. He reported that Hypericum tetrapterum Fries “... Square S. Johns Grasse”... “groweth plentifully in the North part of England, especially in Landsdale and Craven” (Gerard 1633: 542). Again, this is the first record for Britain. He also described Digitalis purpurea L. “Fox-Gloves....in Landes- dale [Lonsdale, north-east of Lancaster], and Craven, in a field called Cragge-close [see Pinguicula vulgaris, above], in the north of England” (Gerard 1633: 791) and_ also Parnassia palustris L. “Grasse of Parn- assus....in Landsdall [Lonsdale] and Craven, in the North parts of England” (Gerard 1633: 840). These plants are all still present in the area in suitable habitats. To these may be added Valeriana officinalis (“‘Mountaine Setwall, or Nardus”’, Gerard 1633:1081) also at Crag-close, and both this and Phyllitis scolopendrium (“Harts tongue’) on Ingleborough (Gerard 1633:1139), both of which may also be relevant here. (B) NORTH-EAST LANCASHIRE (MAINLY V.C. 59) All the records from here are attributable to Hesketh (some possibly jointly with his mother). Other than his discoveries in Craven (given above) and those near his inherited property in west Lancashire, most of the remainder are from a limited area within a few kilometres of his Great Harwood home. Both his birth-place, Martholme Hall, and _ its associated gate-house, still survive. Into the stonework are carved the letters ““T H” and the dates 1561 and 1577 commemorating the renovation work carried out there by his father. Hesketh must have spent much time wandering the local countryside and would have known this area in detail. Most of his records are provided by Gerard who possibly also visited Martholme. LATHRAEA SQUAMARIA L. This remarkably localised record was given by Gerard (1597: 1388) as “Dentaria maior Mathioli — Great Toothwort, or Lungwort [which] groweth likewise neere Harwood in Lancashire, a mile from Whanley [Whalley], in a wood called Talbot banke”. This wood is on the bank of the Calder about three kilometres downstream from Hesketh’s home at Martholme. The plant was seen around 1827 at what is probably Hesketh’s original locality: “George Ward....informs me that.....that he had found this plant about 35 years ago [1.e. c.1827] between Whalley and Harwood, about a mile from the former place and near to the river Calder” (Ashfield 1862). This wood is now very overgrown, a recent search for the plant being unsuccessful but it could still survive there. It still occurs nearby at Mitton (pers. obs. 2006) and is also known elsewhere in the vicinity. LEPIDUM LATIFOLIUM L. This was recorded by Parkinson (1640: 856) for Sawley further to the north but this may be a mis-spelling for Whalley which is closer to Hesketh’s home. In either case it is most likely to originate from him. It is normally a plant of coastal areas but is sometimes naturalised inland and there are modern records from as close as 3 km to the south-east of Martholme (S. Bungard, pers. comm., 2005). MAIANTHEMUM BIFOLIUM (L.) F. W. SCHMIDT Gerard (1597: 330) lists “Monophyllon — One Blade” as occuring “ in Lancashire in Dingley [Dinckley] Wood, six miles from Preston in Aundernesse; and in Harwood [Great Harwood], neere to Blackburne’’. Both of these localities are very close to Hesketh’s home and although he is not specifically named, there can be little doubt these are his records. Nowadays, Dinckley Wood is much reduced and modern searches have been unsuccessful. The plant was probably only doubtfully native there in those days and is now long since gone, as it is from Great Harwood. NARTHECIUM OSSIFRAGUM (L.) HUDSON According to Gerard (1597: 89) “Asphodelus Lancastriae verus — The true Lancashire Asphodil” was found by Hesketh at “Martom’” [Martholme] and that he ~... brought the plants thereof unto me for the increase of my garden’. Hesketh also found the plant at Mawdesley, near his other property. Ne ROREN 360 voie Ysnosogo[suy [Boo] Alo A, ay) UL yng pasifeooyuy) QRQC| Auudd (LEZ F1OO]) snisn[D snavdosna SN JOA], yuasoid [JOS ysnojogo|suy QQC] yoysoH ‘Auudd (QQC]) snieszswiesy Dasod Wnpas yuasoa.d [JOS ysno1oqgaysuy L6S1 YIOysoH ‘AuUdg (ORET WY [60] :L6S1) presoy SNAOWIDULDY SNGNY POYOO]ADAO Jng d1dy) [PHS ATqugold wieydryy ‘poom oepdrya L6S1 YoysoH (LE9 :LOS1) parson SUDA X SIADBINA DINWIA vole YSNOJOgo|[suU] S.1IND90 [DS ay} UL Ng pasi~eoo[uy RRC] Auudd (TOE : L091) snisnya DsSOULD, DINU sivas Aqivau Uo Judso.id [NS wivydryy ‘poom ayepdeya L6S| YloysoH (RCL :L6OS]) parslay LUNJVDAOPO WNIDUOSA]Od Juasaid |[NS sMS Avou pue Yysno1ogo]suy L6S| WoysoH ‘AuUdd (SPO :L6S1) PlesoH SIDS INA DININSUL uasaid [[NS Bole UDARID,, L6S1 — (A[qeqo.id) Waysoy (OP :€€9]) pavslay SLUISN]Dd DISSDUAD yuosaid [[NS Bole UdARID,, LOS) (Ayqeqo.d) ijoysoH] (ZPS :€€9]) presroyH wnsajdvsjay wnIsada yy JUOSOUNT] DY} JJO JUONbd14 vIIv UOJD[SU] L6S1 (Ayqeqo.d) qyoysoy (16L :€€9]) pavsoyH pasndind s1yDj1siq Aanquao YIUIIYSId dv] OY) DUIS JOUNX| UOJI[SU] (POO AA SYJOH 67291 jpeysun |, (QP¢ :6Z9]) UOsUTyeg snjoaa po wnipadudky yuonbay [JOS ysnoloqgasuy 0r9] jpeysun |, (O87 ‘OPO ) UOsUDy.Ieg poiwuaddd DIADA]YIOD 6S8] Isevoy ye [Hun AqJeou UMOUY SBM {}OUTX9 SUO'T UOJI[SUT ‘POO AA SYJOH 679 | jjeysun (QPe :6Z9]) UOsuTyeg piyofisuo] paaysupjpyday yuonba.f ysnojoqgaysuy Moog [8S] AuUdd (IHA]XO [QO]) snisnyiD snppdydosajay snnpavy yuonba.ry AIA [JOS Ysnologasuy] JVaN 679 | jyeysun |, (68E :6Z9]) UOSsUTyAeg sisuajDAd AUNUDPADD BIIV YSNOLOGI[SUT/UIABAD (B) :0) JO1Id snjeys Jung AVeoo7] pop.1locay JOpuly JO.INOS so1ads TIVLSNOL GNV HLANSHH “ANNHd AP GHCYOOdA SLNV Id dO AONANANOOO LNASHad GNV LSVd ‘1 ATaViL 361 EARLY RECORDS FROM NORTH-WEST ENGLAND “AINIUID YIOHUIM) OY} JO d]Ppruu sy) Oo} dn soy) SUTATAINS “OOIOUITD pur AdpeYAA UIIMIOG UOWTWOD [[IYPeH 1e svM vor o}eIPOUUUUT IY] Ul PIOSOI ISP] OUT, “(OX 2] ISYVANIVN' allySVIUVT) LOG] SB APUDEA sv ‘OLUTOYVILIY AeOU OTS asTOoId s YoYsSoH Je A[Qtssod ‘poomiep WwoIH Ww UMOUY SvM IL x. UOSJON Ieou AVMP d0uR}SIP OWIOS UMOUY Ing JOUNX| (pajueyd Ajqeqo.d) jounx| juonbol ‘juasaid |pOs Ayqeqo.d pur syc6| ay) [HUN da19y) UMOUY “IaIOMOTJZ AYS V pe.no} yusoelpe uv WOIJ UMOUY Ing OUTXAE sdeysog x LO6| Ul plodal Ise] “OUNX| JOO [[OMST AM UO Aqievou juasaid yng youNXa ATqeqo.1d d0.19YIWO ye Aqivau popsoda1 APUIIIY *JOUNX| JOO [[OMSTAA JO Agavou s.inddO }OUNXY JUIOY S$,YIOYSOH 0} vou st ADT[PU AA ‘yournxe Ayqeqo.id jng UMOUY JON adyMas]o pur uo, Je Aqivou yng pouLstyuoouy SNCS JUSTIN) (POOMIBH IID) JOP[VD ALY oy) JO syueg L6S1 WoysoH POOMIRH WwIIH L6SI WoysoH] PpOoOMIRH WII L6S1 YloysoH HHH 2[PUsd LOSI IOAS9H (POOMIRH WIIH) dUpOULe|y L6S1 yloysoH POOMIBF IID L6SI YloysoH POOMIL]] IIIH L6SI yoysoy (POOMIR}] WIID) UITOUVV|AY pue Aopyourq L6S1 YJoysoH (POOMARH WwIIH) sUpOyVe| LOSI yoysoy poomiey yeoIg pure Aopyourqd L6S1 YoysoH (,AoTTLY AA :poyfods -stuu sdeyiod yng) Asjmes Or9OT (ATqeqo.d) yaysoH POOMIBF] WILD -AdT[PUAA ‘POOM Yue JOg]e ], LOSI YJoysoH :0) 101d AVVO + pop10say Jopuly ““INODO T HTaVi (96S (CES (1601 (O8ET 2 1601 (COSI (6£9 (SP9 (SCE (68 (OCE :LOS 1) parsay :L6S 1) paesaH :L6S 1) parson :L6S[) paesay :€£9 1) PleIIO :L6S 1) Ple1oH :L6S 1) Ple1oO :L6$1) Plv1oH :L6$ 1) Pre1oH ‘LOS 1) pav.sop (9S8 :O79]) UOsUDy ed (88EI :LOS 1) Paesay d0.INOS snyofissod Uosodosv J WNIPAOIS WNLINA snpavpl Snqny SHAOWIDUDYO SNGNY snppd snunid DSOULID DINU SIDS INA DINIMNSULg pyofiaponb sisvd WNBDALISSO WNIIAYJADN wnyofig wNnWayJUvIDA wnyofuy) wnipiday DIUDUDNDS VADAYIVT IATYSVIUL’T JSBI-YJAON (q) sorsadg 362 M. J. Y. FOLEY These appear to be the first British records. It may still be present in the vicinity of Martholme and certainly survives just a few kilometres to the north-west in the Wiswell Moor area (D Earle, pers. comm., 2005). PARIS QUADRIFOLIA L. “One-Berry, or Herbe True-love” was known “...1n Blackburne at a place called Merton [Martholme] in Lancashire; [and]... in Dingley [Dinckley] wood, six miles from Preston in Aundernesse...” (Gerard (1597: 328). Both localities are at, or very close to, Hesketh’s home and again, although he is not mentioned by name, they must undoubtedly originate from him. The more likely place for the plant to survive will be at Dinckley but there are no modern records from there. Savidge et al. (1963) gave a record for the plant in woods at Moreton Hall, Whalley, a locality only about one kilometre from Martholme. It was more recently recorded near Horrocksford Quarries, Clitheroe in 1964 approximately ten kilometres to the north of Martholme. PINGUICULA VULGARIS L. Gerard (1597: 645) recorded “Pinguicula sine Sanicula Eboracensis -— Butterwort, or Yorkshire Sanicle. ...a field called Crag-Close, and at Crosby, Ravenswaith, in Westmerland, upon Ingleborow fels.....and in Harwood neere to Blackburne”’. The record for Great Harwood will certainly be Hesketh’s. Currently not known precisely from here, it still survives in small numbers on Wiswell Moor, less than 4 kilometres to the north (S_ Bungard, pers. comm.). The Crag-Close locality is in the Settle area where Hesketh had been known to operate (see Primula vulgaris x P. veris) and may be his record. It seems to be a local or vernacular name since rocky ground, of which there is a great deal there, is sometimes referred to as a ‘close’. The} plant ystill survives piney such localities. PRIMULA FARINOSA L. Gerard (1597: 639) calls this “Birds-eine?” and that “...plants grow very plentifully in moist and squally grounds...in Harwood neere to Blackburn...”. This is one of the first British records for P. farinosa, a northern England plant of damp ground over base-rich_ substrates. Again, although not mentioned directly by name, the “Harwood” locality must originate from Hesketh; it is towards the plant’s southern British limit. It was “still to be found .....at Great Harwood...”, possibly at Hesketh’s precise site near Martholme, as recently as 1907 (Anon, 1914). Close to Martholme there were several calcareous flushes (perpetuated by the locality name “Allsprings”). One such flush near to the bank of the Calder about a kilometre downstream of Martholme, - still survives but, unfortunately, in the absence of the plant. P. farinosa was also frequently recorded in the general vicinity, especially around nearby Pendle Hill. The last record appears to have been on Hardhill Common between Whalley and Clitheroe where it survived until the middle of the twentieth century but drainage and agricultural activity has resulted in its loss nowadays. PRUNUS PADUS L. “Cerasus avium nigra & racemosa — Birds Cherry; and blacke Grape Chemysincem er groweth likewise in Martome Park [Martholme], foure miles from Blackeburne, and in Harward [Great Harwood] neere thereunto; in Lancashire almost in every hedge...”. Gerard (1633: 1505). Currently it is known from an adjacent tetrad to Martholme (D. Earl, pers. comm., 2005). RUBUS IDAEUS L. Gerard (1597: 1091) says that “Rubus idaeus” — Raspis, or Hinde-berry” is not wild “except in a fielde by a village in Lancashire called Harwood [Great Harwood] not farre from Blackburne [Blackburn]”. Again, surely a Hesketh record from near Martholme and probably still present there as it is a common plant in the general area. RUBUS CHAMAEMORUS L. The source of the first British record is confused but there is little doubt that it was either from near the summits of Ingleborough or Pendle. Under Vaccinia nubis (Cloud-berrie) Gerard (1597: 1386) stated it was found on two high mountains, one of which is “in Yorkshire called Ingleborough, the other in Lancashire called Rendles= found there by a curious gentleman in the knowledge of plants, called Master Hesketh, often remembred”. However, as remarked earlier regarding Sedum rosea, Thomas Penny who predated Hesketh by thirty years or so, probably first found it on Ingleborough although Hesketh’s record for Pendle will be the first for that hill. At the latter locality it has always been a shy flowerer, but flowering plants were recorded there in 1934 EARLY RECORDS FROM NORTH-WEST ENGLAND 363 (Turner 1934) and where it almost certainly still survives. Its very shy flowering on Pendle is unusual but this was also noted, even into the past (Dobson 1864; 1881), Leigh (1700). The distinctly crude illustration reproduced by Gerard will probably have originated from Hesketh. It appears to show a plant both in flower and in fruit and for that reason it is perhaps more likely to relate to one he saw on Ingleborough rather than Pendle. TEUCRIUM SCORDIUM L. “Scordium — Water Germander.... in a meadow by Harwood in Lancashire” Gerard (1597: 535). Although not named, this must be Hesketh’s record from near his Martholme home. Presumably introduced and not now known there of course, this is a plant of calcareous soils, sometimes found on river banks. It is now extremely rare anywhere in the British Isles. TRAGOPOGON PORRIFOLIUS L. Gerard (1597: 596) states that “Tragopogon purpureum — Purple Goats-beard.....growes not wild in England that I could ever see or heare of, except in Lancashire upon the banks of the river Chalder [Calder], neere to my Lady Heskiths house, two miles from Whawley [Whalley]”. This mainly Mediterranean plant can occur as a casual or is introduced for use as a vegetable. It is occasionally found within the area. Martholme Hall, where Thomas was born and his mother lived, is on the banks of the Calder close to this spot. This unusual direct reference to Lady Hesketh may mean that she had personally sent a specimen to him, or that Thomas had done so mentioning its presence near to his mother’s home, or even that Gerard had seen the plant there if he had visited them at Martholme. The most recent record appears to have been at Reedyford near Nelson (v.c. 59) in 1927 which is about 12 kilometres from Martholme (Savidge ef al. 1963). SUMMARY Many of the plants recorded by the three early botanists still survive within the areas discussed. The Craven area, especially, is undeveloped commercially, although farmland will have come under much more intensive cultivation recently. Since Hesketh’s day, the town of Great Harwood has grown and spread widely, and again, agriculture has intensified but, despite this, many of his plants survive locally. As expected, the major losses have been of the more attractive plants, the rare orchids, which were ravaged through over-collection even as early as the seventeenth century. The past and present occurrence of each of the plants discussed is summarised in Table 1. ACKNOWLEDGMENTS I am most grateful to the library staff of the Royal Botanic Garden, Edinburgh, for allowing me to examine the relevant early literature held there and to the authorities at the Lancashire Records Centre, Preston, for permission to examine documents relating to Thomas Hesketh. I also thank Stephen Bungart, David Earl and Peter Jepson for some of the current records for v.c. 59 and Phyl Abbott for a recent record in v.c. 64 and for her helpful comments on the paper. REFERENCES ANON. (1914). Lancashire Naturalist 6: 80. ASHFIELD, C. J. (1862). The Phytologist: a botanical journal (New Series) 6 352. CAMERARIUS, J. (1588). Hortus medicus et philosophicus. Francofurti ad Moenum. CLUsIUS, C. (1583). Rariorum aliquot Stirpium, per Pannoniam, Austriam, & vicinas quasdam Provincias observatarum Historia. Antverpiae. CLUSIUS, C. (1601). Rariarum plantarum historia. Antverpiae. CURTIS, W. H. (1941). William Curtis 1746-1799. Winchester. DOBSON, W. (1864). Rambles by the Ribble, \st ed. Preston. DOBSON, W. (1881). Rambles by the Ribble, 4th ed. Preston. FARRER, W. & BROWNBILL, J. (1914). (eds). The Victoria History of the County of Lancaster, 8: 232-237. London. FOLEY, M. J. Y. (2006). Thomas Penny, the sixteenth century English botanist: some of his European plant records. Saussurea 36: 87-101. GERARD, J. (1597). The herball or general historie of plantes gathered by John Gerarde of London Master in Chirurgerie. London. 364 Mls ds XG JKOILSNC GERARD, J. (1633). The herbal or generall historie of plants: gathered by John Gerarde of London master in chirurgerie very much enlarged and amended by Thomas Johnson citizen and apothercarye. London. LEES, F. A. (1888). The Flora of West Yorkshire: with a sketch of the climatology and lithology in connection therewith. London. LEIGH, C. (1700). Natural History of Lancashire, Cheshire, and the Peak, in Derbyshire. Oxford. L’OBEL, M. DE (1655). Stirpium illustrationes: Plurimam elaborantes inauditas plantas. Londini. 118. MOUFEFET, T. (1634). Insectorum sive minimorum Animalium Theatrum. Londini. PARKINSON, J. (1629). Paradisi in sole, paradisus terrestris. London. PARKINSON, J. (1640). Theatrum Botanicum = The Theater of Plants: or an Herball of a large extent. London. SAVIDGE, J. P., HEYWoOoD, V. H. & GORDON, V. (1963). (eds.). Travis’s Flora of South Lancashire. Liverpool. TURNER, A. (1934). Cloudberry in flower on Pendle Hill. North Western Naturalist 9: 382-383. (Accepted April 2009) Watsonia 27: 365—368 (2009) 365 The Svalbard plant collection in the Arctic Herbarium at the University of Lancaster (LANC) VIE FORE NG: Faraday Building, Department of Biological Science, University of Lancaster, Lancaster LAl 4YA ABSTRACT A detailed examination (and cataloguing) has been made of material collected from the Svalbard archipelago which is currently held in the University of Lancaster herbarium, U.K. (LANC). This paper briefly outlines the background to its formation and to its collectors and contents. However, full details of the relevant specimens are available on the internet and can be accessed via the link: http://biol.lancs.ac.uk/svalbard/data.pdf INTRODUCTION The Lancaster University herbarium as a whole contains a wide range of preserved European plant specimens as well as an important specialised section loosely known as _ the ‘Arctic Herbarium’. The latter is divided into geographical regions circumscribing the arctic: i.e. Alaska Yukon, Canada, Greenland (the largest section), Iceland, Fennoscandia, Svalbard, and northern Russia (in two sections, west & east). However, it is the Svalbard section of the herbarium upon which this paper is based. The Svalbard archipelago lies well to the north of the Arctic Circle and has several months of permanent daylight each year. It comprises four main islands: Barents@ya, Edgegya, Nordaustlandet and Spitsbergen, of which the latter is the largest and most widely known. Smaller islands include Byjorn@ya, Hopen, Kong Karl’s Land, Kvitgya, Prinz Karl’s_ Forland and the Sjugyane group. Together, they lie between latitudes approx- imately 76° 30’ N and 80° 30’ N, well to the north of northern Norway and within 1000 kilometres of the North Pole. To the west of the archipelago the North Atlantic Current, a remnant of the warm Gulf Stream, favourably influences the climate but the eastern side with its southward-flowing cold Artic Current is very adversely affected. A consequence is that the western coastal regions of Spitsbergen and especially the central parts, have a climate much warmer than would be expected at these high latitudes. Here, and especially in the inner *e-mail: m.foley @lancaster.ac.uk fjord areas, quite a rich flora has developed, such that in addition to monocotyledons such as grasses and sedges, there is a surprising variety of colourful flowering plants. The latter number more than one hundred species and need to make the most of the period of relative warmth and permanent daylight of high summer to complete their flowering and reproductive cycle. Elsewhere, and especially on the much colder north and east sides of the archipelago, the flora can be extremely sparse, much of the region comprising polar desert or even permanent ice-cap. To the south of the archipelago, the two principal coastal currents meet near the isolated island of Bjorngya and here again the climate is harsh, often with periods of persistent fog. In fact the name Svalbard means ‘cold shores’. It is sometimes thought that the discovery of Svalbard dates back to the twelfth century but it was the Dutchman, Willem Barents, who is more often credited with this when he sailed along the Spitsbergen coast in 1596. Thereafter, other nations paid visits and exploited the archi-pelago’s waters by whaling until stocks were drastically depleted and the practice ceased. Later, early in the twentieth century, coal mining became a minor industry in some locations whilst others, especially in the north, became bases for expeditions participating in high arctic exploration either by boat or balloon. Nowadays, apart from the settlement of Longyearbyen on Isfjord (Spitsbergen) with its population of around 2000 and a few other very minor settlements and research stations, Svalbard is completely uninhabited; most of it is designated as a strictly controlled National Park. Recently, the Norwegian government has capitalised on Svalbard’s geographical isolation by constructing a ‘doomsday’ seed bank near Longyearbyen. This is designed to store seed from as many of the world’s crop varieties and wild relatives as possible and is precautionary in preventing their genetic diversity being adversely affected by man-made or natural disasters. 366 M. J. Y. FOLEY EARLY BOTANICAL EXPLORATION ON SVALBARD Prior to the early twentieth century there was little serious botanical exploration carried out on Svalbard. What took place was very much a side-line to other seemingly more serious matters such as arctic exploration and whaling. Probably the earliest to make any contribution to the flora was the Englishman, Constantine John Phipps (1744-1792). In 1773 he sailed north (a young Horatio Nelson was with him as a crew member) in an attempt to reach the North Pole. In doing so he reached the Seven Islands (Sjugyane), which are the most northerly ones in Svalbard, but was turned back there by dense pack ice. One of the islands of this group was later named Phippsgya in his memory. He is thought to be the first European to provide a description of the polar bear and the ivory gull, both of which he saw on the voyage. As for plants, he is credited with the discovery of Phippsia algida (the genus being named after him by Daniel Solander). He also found Ranunculus sulphureus, also subsequently described by Solander, as well as four saxifrages and seven other flowering plants. All these together with other items of natural history are detailed in the Appendix to his book ‘A Voyage towards the North Pole undertaken by His Majesty’s command 1773’ (Phipps 1774). At a much later date, around the turn of the nineteenth century, the Scotsman William Speirs Bruce made several visits to the archipelago. Bruce was essentially a naturalist and paid his first visit when invited to join Prince Albert of Monaco’s exploratory oceano- graphic expedition in 1898; he also paid a further visit a year later. During his time there Bruce collected plants, several preserved specimens of which are retained in_ the Svalbard herbarium (see below). He also discovered the presence of coal and other minerals which, in 1909, led him to form the Scottish Spitsbergen Syndicate with the aim of exploiting these natural resources. Prior to Bruce’s first visit, the Swedish palaeo-botanist, Alfred Nathorst had visited several parts of the archipelago. His main interest, however, was plant fossils and the glacial plant flora of the region. He made several discoveries within his — subject, publishing them in a paper titled “Om vegetat- ionen pa Spetsbergens vastkust’” (Nathorst 1871). As the early twentieth century proceeded, interest in Svalbard’s flora accelerated with botanists such as Dahl, Elton, Hoeg, Lid, Lynge, Mathery-Dupraz, Resvoll-Holmsen, Rudmose Brown, Summerhayes and Tolmatchew making visits there. These can be considered to be the first modern-era botanists to the islands. In the LANC Svalbard herbarium there are in fact specimens collected by Bruce, Dahl, Lid, and Lynge, but all subsequent additions originate from the 1960s onwards. THE SVALBARD HERBARIUM The formation of the Arctic Herbarium began in 1961 and was initiated by Geoffrey Halliday whilst on the staff of Leicester University (1958—-1968).This followed his first expedition to East Greenland to study its flora. On leaving Leicester, Halliday transferred to Lancaster University and became a member of the academic staff there and was permitted to bring the arctic collection with him. The total number of sheets in this collection now numbers several thousand, approximately 665 of which are from Svalbard with about 75% of the 165 known native species being represented by at least one sheet. In con- sequence, this provides a very useful base reference for anyone interested in, or wishing to learn more of, the specialised flora and is also a useful facility to aid identification. Much of the material originates from university expeditions and, whilst parts of the archipelago are reasonably well covered, others are not covered at all. This is largely due to two factors: easier access such as to the Isfjorden region, and to the particular interests of various expeditions which have concentrated on limited areas such as_ Bellsund, van Mijenfjorden and Wijdefjorden. The total number of contributors to the herbarium is only twenty-one and _ several of these have contributed just a very few sheets. However, Derek Spicer (218 sheets), Alasdair Neilson (107) and Sven Manum (70) have been particularly prolific and their combined collections represent almost 60% of the overall total. Other collectors are mentioned below and of these Martin Halliday (no relation to Geoffrey Halliday), Robert Harper and Derek Spicer were Leicester University under- graduates at that time, Alasdair Neilson was a Cambridge graduate then at Glasgow University and K. Vaughton was at Oxford. SVALBARD PLANT COLLECTION 367 Of special historical interest are nineteen sheets collected in 1898 and 1899 by William Speirs Bruce (see above). Most of these are from the remote, and at that time relatively unexplored, islands of Bj@rngya, Barentsgya and Hopen, the two latter situated towards the eastern side of the archipelago. Bruce also made a few collections from Spitsbergen itself, i.e. at Recherchefjorden and Raudfjorden. At the Raudfjorden locality he collected Cerastium arcticum and Luzula confusa, the actual locality being quaintly described on the sheet as ‘opposite the Princess Alice’. This is a reference to his ship (owned by Prince Albert 1 of Monaco) which ran aground on a submerged rock in Raudfjorden, north-west Spitsbergen in 1899 and close to a headland now known as Bruce Point. When the ship was finally floated clear, the expedition terminated as it was necessary to sail south to the mainland for repairs. Bruce’s name is also commemorated in a remote abandoned settlement known as Brucebyen in Billefjorden which lies close to the impressive glacier front of the Nordenskjgldbreen. It still survives today and comprises four huts built by Bruce in the early twentieth century. It is still sometimes used by expeditions. Some of the sheets in the Svalbard herbarium are duplicated in others such as the Botanical Museum, University of Oslo (O). Johannes Lid collected in_ Bellsund, Isfjorden and Wijdefjorden in 1920 and 1924 (represented here by 12 sheets), B Lynge in Bellsund in 1926 (13) and Eilif Dahl in Isfjorden and Dicksonfjorden in 1936 (4). Other than these early collections and those of Bruce and the occasional sheet from other botanists, by far the greatest amount of material derives from the period 1962-1970. This was the result of the combined efforts of several botanists. Listed chronologically these are, F. Hgrl & J. Schweitzer who collected at Adventfjorden and Sassendalen in 1961 (4 sheets); A. H. Neilson mainly at Wijdefjorden and elsewhere in the north and north-east in 1962 (107); S. Manum at Isfjorden (especially Boheman on_ the northern shore) and Bellsund in 1962 & 1964 (70); A. Jenkinson at Reinsdyrflya in 1963 (7); D. P. Spicer mainly in north Spitsbergen (Wijdefjorden, Austfjorden, Ny Alesund) in 1964 & 1965 (218; of which 35% are Poaceae); M. B. W. Halliday at van Mijenfjorden and Isfjorden in 1965 (64); R. S. Harper at van Mijenfjorden and Adventdalen in 1965 (45); S. M. Cobb & C. Martell at Liefdefjorden in 1966 (15); H. Schweitzer on Byjérngya in 1967 (7); K. Vaughton at van Mijenfjorden in 1968 (33) and J. W. Squier in Sassendalen (Isfjorden) in 1970 (32). With collecting nowadays limited, the present author has added a modest number of sheets (22) mainly of the commoner plants but including 6 sheets of Draba; all these were collected in the Isfjorden area in July 2007 & 2008. Other than these, no new material has been added within the last 35 years or so. A series worth commenting upon is Neilson’s nine sheets of Draba collected on the remote, bleak island of Nordaustlandet along with one sheet of the scarce Ranunculus pedatifidus from Edgegya. Neilson devoted much of his time to studying and collecting the plants of these two islands, later publishing an account of their flora (Neilson 1968, 1970). Spicer, who had a special interest in Draba, subsequently produced a dissertation on the Svalbard species (Spicer 1969). As stated above, scientific expeditions to the archipelago during the 1960s have been a major contributor, especially those from the University of Leicester in 1965 (collector M. B. W. Halliday) and from Oxford University in 1968 (K. Vaughton). J. Lid’s small collection originated from the two earlier Norwegian Spitsbergen Expeditions of 1920 and 1924. Of the total number of sheets in the herbarium, 3% are of pteridophytes (Lycop- odium, Equisetum and Cystopteris) and 40% of monocotyledons (65% of these being Poaceae). Lower plants such as lichens are not represented. A summary of all the details of each individual collection is available in an electronic document located at: http://biol.lancs.ac.uk/svalbard/data.pdf The information in that document is taken directly from the sheets and includes locality, date, collector and _ collection number, geographical coordinates (where known) and ecological notes. For non-critical species where the correct identification can be relied upon, these notes provide useful information on the distribution, phenology and ecology of the Svalbard flora without the necessity of having to physically examine each specimen. Where critical species have been determined or confirmed, this is also indicated. When cons- ulting the document it should be borne in mind that a semi-colon [;] is used to separate each individual collection. The nomenclature used is that of the original collector(s) and/or herbarium compiler and, in only a few cases, has this been modified or updated. Locality 368 M. J. Y. FOLEY names are as given on the sheets and the confirmed by: AOC. = A. O. Chater; ED = E. systematic arrangement follows that currently Dahl; AG = A. Game; MF = M. J. Y. Foley; in use in the herbarium. GH = G. Halliday; EH = E. Hultén; KJ = K. Jacobsen; JL = J. Lid; BL = B. Lynge; AM = Abbreviations used in the electronic document A. Melderis; JAN = J. A. Nannfeldt; TP = T. are as follows: N = north; S = south; E=east; Prestg; AJR = A. J. Richards; IR = I. Rgnning; W. = west; c)5 ="cincayidet <= idetermmed/ “RDS RA Deselkt REFERENCES NATHORST, A. (1871). Om vegetationen pa Spetsbergens vastkust. Botaniska Notiser 4: 105-117. NEILSON, A. H. (1968). Vascular plants from the northern part of Nordaustlandet, Svalbard. Norsk polarinstitutt. Skrifter, nr. 143. Troms@. NEILSON, A. H. (1970). Vascular plants of Edgegya, Svalbard. Norsk polarinstitutt. Skrifter, nr. 150. Troms@. PHipps, C. J. (1774). A Voyage towards the North Pole undertaken by His Majesty’s command 1773. Bowyer & Nichols. London. SPICER, D. P. (1969). A contribution to the taxonomy of the genus Draba on Spitsbergen. M. Sc. Thesis (unpublished), University of Leicester. (Accepted June 2009) Watsonia 27: 369-374 (2009) 369 Notes THREE MORE NEW COMBINATIONS IN THE BRITISH FLORA Three further new combinations are required for the forthcoming third edition of New Flora of the British Isles. 1. Poterium L. It has been shown that the earliest epithet at subspecific level for the introduced subspecies of Sanguisorba minor Scop., until recently known as subsp. muricata (Gremli) Briq., 1s balearica, and the combination S. minor subsp. balearica (Bourg. ex Nyman) Munoz Garm. & C Navarro is available. However, molecular data indicate that the genus Poterium L. should now be re-segregated from Sanguisorba L., and the required combination under the latter genus appears to be unpublished: Poterium sanguisorba L. subsp. balearicum (Bourg. ex Nyman) Stace, comb. nov. BASIONYM: Poterium spachianum Coss. subsp. balearicum Bourg. ex Nyman, Consp. FI. Eur. 1: 240 (1878) SYNONYMS: Sanguisorba minor Scop. subsp. balearica (Bourg. ex Nyman) Munoz Garm. & C Navarro, in Castrov., Fl. Iberica 6: 386 (1998) S. minor subsp. muricata (Gremli) Briq., Prodr. FI. Corse 2(1): 209 (1913) S. minor subsp. polygama (Waldst. & Kit.) Cout., Fl. Portugal 296 (1913) Poterium sanguisorba subsp. muricatum (Gremli) Bonnier & Layens, Tabi. Syn. PI. Vasc. France 102 (1894) 2. Malva L. Molecular data have shown that Malva L. and Lavatera L. are artificial genera based on a single character (fusion of epicalyx segments), and that their species form an interlocking mass rather than two separate groups. The amalga- mation of the two genera is preferable to the recognition of several groups of species as separate genera, when, confusingly, Malva moschata L. would fall into Lavatera, and Lavatera cretica L. into Malva. Combinations in Malva exist for all our species of Lavatera, except for the garden hybrid between M. olbia and M. thuringiaca: Malva olbia (L.) Alef. (Lavatera olbia L.) x Malva thuringiaca (L.) Vis. (Lavatera thuringiaca L.) = Malva x clementii (Cheek) Stace, comb. nov. BASIONYM: Lavatera x clementii Cheek, Kew Bull. 55: 1013 (2000) 3. Elytrigia Desv. The variants of E. repens (L.) Desv. ex Nevski and EF. atherica (Link) Kerguélen with awned lemmas are best recognised at the level of forma. The combination EF. repens f. aristata (Schum.) Beetle exists for the former; a new combination is needed for the latter: Elytrigia atherica Kerguélen forma setigera (Dumort.) Stace, comb. nov. BASIONYM: Agropyron littorale Dumort. var. setigerum Dumort., Obs. Gram. Fl. Belg. 97 (1824) SYNONYM: Elymus pycnanthus (Godr.) Melderis var. setigerus (Dumort.) Melderis ACKNOWLEDGMENT I am grateful to Christopher Davis for advice on the nomenclature of Malva and Lavatera. C. A. STACE Cringlee, Claybrooke Road, Ullesthorpe, Leics LEI7 5AB 370 NOTES Watsonia 27 (2009) VALIDATION OF NAMES FOR NEW AVON GORGE SORBUS (ROSACEAE) TAXA I am embarrassed to find that under section 8.2 of the International Code of Botanical Nomenclature (Greuter et al. 2000), the five new Sorbus names I published for the Avon Gorge (Rich et al. 2009) are invalid as more than one gathering was cited for each holotype. The names are republished below’ with corrected citation of their holotypes. Sorbus x avonensis T. Rich, hybr. nov. (Sorbus aria (L.) Crantz x S. porrigentiformis E. F. Warb.). HOLOTYPUS: St Vincent’s Rocks South, Avon Gorge, v.c. 34 West Gloucestershire, England, Sit/5659-/305 19 September 20077 MG 1GaRich & L. Houston (NMW, accession number V. ZOOM MSZ): IsOoTYPI: BM, CGE. Validating description in Watsonia 27: 218. 2009. Sorbus x houstoniae T. Rich, hybr. nov. (Sorbus aria (L.) Crantz x 8S. bristoliensis Wilm.). HOLOTYPUS: rocks Stokeleigh Camp, v.c. 6 North Somerset, England, ST/561.732, 31 October 2004, T. C. G. Rich, A. Robertson & L. Houston (NMW, accession number V.2004.26.276). Validating description in Watsonia 27: 220. 2009. Sorbus leighensis T. Rich, sp. nov. HOLOTYPUS: Quarry 4, Leigh Woods, v.c. 6 North Somerset, England, ST/561.739, 19 September 2007, T. C. G. Rich & L. Houston (NMW accession number V. 2007.1.180). IsOTYPI: : BM, CGE. Validating description in Watsonia 27: 222- 221. 2009. Sorbus xX robertsonii T. Rich, hybr. nov. (Sorbus aria (L.) Crantz x S. eminens E. F. Warb.) HOLOTYPUS: Great Fault, Avon Gorge, Bristol, v.c. 34 West Gloucestershire, England. ST/564.733, 19 September 2007, T. C. G. Rich & L. Houston (NMW, accession number V 2007 175): IsoTYPI: BM, CGE, OXF. Validating description in Watsonia 27: 224— 225. 2009. Sorbus x proctoriana T. Rich, hybr. nov. (Sorbus aucuparia L. X S. scalaris Koehne). HOLOTYPUS: Top edge of Quarry 2, Leigh Woods, v.c. 6 North Somerset, England, ST/557.743, 19 September 2007, T. C. G. Rich & L. Houston (NMW, accession number V2VOTAIENZS)): IsoTYPI: BM, CGE, E, LIV, OXF. Validating description in Watsonia 27: 226. 2009. ACKNOWLEDGMENTS I would like to thank Katherine Challis, Clive Stace and John McNeill for nomenclatural advice. REFERENCES GREUTER, W., MCNEILL, J., BARRIE, F. R., BURDET, H.-M., DEMOULIN, V., FILGUEIRAS, T. S., NICOLSON, D. H., SILVA, P. C., SKOG, J. E., TREHANE, P., TURLAND, N. J. & HAWKSWORTH, D. L. (2000). International Code of Botanical Nomenclature (St Louis Code). Regnum Vegetabile 138. Koeltz Scientific Books, KOnigstein. RIcH, T. C. G., HARRIS, S. A. & Hiscock, S. J (2009). Five new Sorbus (Rosaceae) taxa from the Avon Gorge, England. Watsonia 27: 217-228. TC GARICH Dept. of Biodiversity and Systematic Biology, National Museum of Wales, Cardiff, CF 10 3NP NOTES Watsonia 27 (2009) 371 REINSTATEMENT OF THE NAME DACTYLORHIZA KERRYENSIS (WILMOTT) P. F. HUNT & SUMMERH. FOR THE WESTERN MARSH ORCHID, AND A NEW VARIETAL COMBINATION Recent research has revealed that the taxon formerly known in Britain and Ireland as Dactylorhiza majalis is distinct from Contin- ental populations of that species (Delforge er al. 1991), and further, that its former sub- species do not all belong to a single species (Delforge 2000; Foley & Clarke 2005; Bateman 2006) Until recently, plants in Ireland and in North Uist had been classified under Dactylorhiza majalis subsp. occidentalis and all other populations were referred to subsp. cambrensis (Stace 1997). Recent work has argued that this last is better treated as a variety of D. purpurella (Foley & Clarke 2005; Bateman 2006), and that the North Uist plants are a species of hybrid origin, D. ebudensis (Wiefelspiitz ex R.M. Bateman & Denholm) P. Delforge (Delforge 2000). This means that the Irish plants are referable to an endemic species for Ireland, which has been erroneously called Dactylorhiza occidentalis (Pugsley) P. Delforge (Delforge et. al. 1991; Foley & Clarke 2005; Bateman 2006). Whilst the broader taxonomy has _ been resolved for the present, the nomenclature of the species has become confused. The first description of the taxon was by Pugsley (1935), who described Orchis majalis var. occidentalis. The following year Wilmott (1936) published Orchis kerryensis, a species lacking spots on its leaves. Under the International Code of Botanical Nomenclature, names only have priority at the same rank (McNeil ef al.; Art. 11.2). Pugsley’s 1935 variety is therefore to be regarded as a synonym, as is the combination Dactylorhiza occidentalis published in 1991, and Dactylorhiza kerryensis should be adopted. Given that the combination Dactylorhiza occidentalis dates from 1991, and has not been extensively adopted into British or Irish literature, there is little cause for inconvenience with adopting the correct name. Bateman and Denholm (1983) recognised two varieties of this taxon in Ireland, namely var. occidentalis with spotted leaves, and var. kerryensis with unspotted leaves. Baumann & Kiinkele (1988) made the correct combination, under Dactylorhiza kerryensis, but at the rank of subspecies: Whilst the two varieties are distinctive, the leaf spotting and flower colour is variable, and the two forms are occasionally found in mixed populations. Varietal rank, however, seems justifiable. The full nomen- clature is laid out below. Dactylorhiza kerryensis (Wilmott) P. F. Hunt & Summerh., Watsonia 6: 131 (1965). BASIONYM: Orchis kerryensis Wilmott, Proceedings of the Linnean Society of London 148, 126 (1936). SYNONYMS: Dactylorchis kerryensis (Wilmott) Verm., Studies on Dactylorchids 67 (1947). Orchis_ occidentalis (Pugsley) Wilmott subsp. kerryensis (Wilmott) A. R. Clapham, in) AV ReXElapham, dG; Tut & E. PF: Warburg, Flora of the British Isles, p.1321 G52): Dactylorhiza majalis (Rchb.) P. F. Hunt & Summerh. var. kerryensis (Wilmott) R. M. Bateman & Denholm, Watsonia, 14: 368 (1983). Dactylorhiza comosa (Scop.) P. D. Sell subsp. occidentalis (Pugsley) P. D. Sell forma kerryensis (Wilmott) P. D. Sell, in P. DE ScligceGa Murrell lord “of "Great Britain & Ireland. 5: 365 (1996). Dactylorhiza occidentalis (Pugsley) P. Delforge var. kerryensis (Wilmott) R. M. Bateman & Denholm, Watsonia 27: 247 (2009). HOLOTYPUS: Ireland, Co. Kerry, v.c. H1, near Dingle, 23 June 1934, A. J. Wilmott (BM). Dactylorhiza kerryensis (Wilmott) P. F. Hunt & Summerh. var. occidentalis (Pugsley) Jebb, comb. nov. BASIONYM: Orchis majalis Rchb. _ var. occidentalis Pugsley, Journal of the Linnean Society of London. Botany 49: 586 (1935). SYNONYMS: Orchis occidentalis (Pugsley) Wilmott, Report, Botanical Exchange Club of the British Isles, 11: 458, 551 (1938). Dactylorchis occidentalis (Pugsley) Verm., Studies on Dactylorchids, 67 (1947). Dactylorchis latifolia (L.) Rothm. subsp. occidentalis (Pugsley) So6, Nomina Novis Generis Dactylorhiza 5 (1962). 3i2 NOTES Watsonia 27 (2009) Dactylorhiza majalis (Rchb.) P. F. Hunt & Summerh subsp. occidentalis (Pugsley) P. D. Sell, Acta Facultatis Rerum Naturalium Universitatis Comenianae. Botanica. Bratislava. 14: 19 (1968). Dactylorhiza majalis (Rchb.) P. F. Hunt & Summerh. subsp. occidentalis (Pugsley) E. Nelson, Monographie und Ikonographie der Orchidaceen-Gattungen Dactylorhiza, 90 (1976). Dactylorhiza majalis (Rchb.) P. F. Hunt & Summerh. var. occidentalis (Pugsley) R. M. Bateman & Denholm, Watsonia, 14: 368 @9s3): Dactylorhiza_ kerryensis (Wilmott) P. F. Hunt & Summerh. subsp. occidentalis (Pugsley) H. Baumann & Kiinkele, Mitteil- ungsblatt, Arbeitskreis Heimische Orchideen Baden-Wiirttemberg, 20(3): 629 (1988). Dactylorhiza_ occidentalis (Pugsley) P. Delforge, Naturalistes Belges, 72(3): 100 (iQ). Dactylorhiza comosa (Scop.) P. D. Sell subsp. occidentalis (Pugsley) P. D. Sell forma occidentalis (Pugsley) P. D. Sell, in P. D. Sell & G. Murrell, Flora of Great Britain & Ireland, 5: 351 (1996). HOLOTYPUS: Ireland} (Co: Clarewayreae lo: Lisdoonvarna, 16-20 May 1935, H. W. Pugsley 514 (BM). ACKNOWLEDGMENTS I am grateful to Dr David Middleton and Professor Clive Stace for reading and commenting on an earlier version of this note. REFERENCES BATEMAN, R. M. (2006). How many orchid species are currently native to the British Isles?, in BAILEY, J. P. & ELLIS, R. G., eds. Current Taxonomic Research on the British and European Flora, pp. 89-110. Botanical Society of the British Isles, London. BATEMAN, R.M. & DENHOLM, I. (1983). A reappraisal of the British and Irish dactylorchids, 1. The tetraploid marsh-orchids. Watsonia 14: 347-376. BAUMANN, H. & KUNKELE; S. (1988). Neue Beitraége zur Taxonomie europdéischer und mediterraner Orchideen. Mitteilungsblatt, Arbeitskreis Heimische Orchideen Baden-Wiirttemberg 20 (3): 610-651. DELFORGE P. (2000). Nouvelles contributions taxonomiques et nomenclaturales aux orchidees d’ Europe. Naturalistes Belges 81: 396-398. DELFORGE P., DEVILLERS-TERSHUREN, J. & DEVILLERS, P. (1991). Contributions taxonomiques et nomenclaturales aux Orchidees d’ Europe. Naturalistes Belges 72: 99-101. FOLEY, M. & CLARKE, S. (2005). Orchids of the British Isles. Griffin Press, Maidenhead. MCNEIL, J., et al. eds. (2006). International Code of Botanical Nomenclature (Vienna Code). Regnum Vegetabile 146. PUGSLEY, H. W. (1935). On some marsh orchids. Journal of the Linnean Society of London. Botany 49: 553- 2, STACE, C. (1997). New Flora of the British Isles, ed. 2. Cambridge University Press, Cambridge. WILMOTT, A. J. (1936). New British marsh orchids. Proceedings of the Linnean Society of London 148: 126— 130. M. JEBB National Botanic Gardens, Glasnevin, Dublin 9, Ireland A NEW NOTHOSPECIES IN SYMPHYTUM L. (BORAGINACEAE) INTRODUCTION This note describes and formally names a new nothospecies in Symphytum L., its putative parents being Symphytum asperum Lepech. and S. orientale L. This taxon was first recorded from Intwood, East Norfolk v.c. 27, England, in 1999 by R. M. Leaney. There appears to be no record in the literature of Symphytum asperum xX S. orientale, whether occurring naturally or as a result of experimental or horticultural crossing. Therefore an epithet is assigned in order that this taxon may receive full treatment in the forthcoming third edition of the New Flora of the British Isles (Stace 997): NOTES Watsonia 27 (2009) 373 DISTRIBUTION & ORIGIN This new taxon occurs in two metapopulations at eight localities in East Norfolk, with one metapopulation at three sites between 2 and 20 km south and south-east of Norwich, and the other over five sites between 17 and 25 km north-west of Norwich. The largest colonies at Intwood and Sustead cover around 150 m? with scattered seedlings extending over a further 50 m radius. Plants in all colonies are constant in morphological characters. The two _ meta- populations, represented by plants from the colonies at Intwood and Sustead, have the same chromosome number. This taxon is believed to originate from a horticultural selection, which has subsequently naturalised. DESCRIPTION The description below is based on_ fresh material. For corolla and calyx measurements, 20 flowers were measured by one author (R.M.L.) from the Intwood colony and 20 from the Sustead colony. Symphytum x norvicense R. M. Leaney & C. L. O’Reilly hybrida nova (Boraginaceae) (putative parentage Symphytum asperum Lepech. 2n = 32 XS. orientale L. 2n = 32, 264) Norfolk Comfrey Herba perennis robusta, erecta vel decumbens, 50-150 cm alta, radice palari obliqua. Caules senescentes decumbentes saepe longiores quam 200 cm. Folia mediocriter viridia vel leviter griseiviridia, infra pallidiora, mollia_ vel aliquantum setosa, alis 1-7 mm sed non decurrentibus petiolata. Folia basalia anguste ovata vel late lanceolata, ad apicem leviter cuspidata subacuta vel acuta, ad basin trucata vel subcordata, ubi perhiemantia saepe valde cordata vel ad basin auriculata; folia caulina superiora late ovata, ad apicem acuta vel acuminata, ad basin truncata vel rotundata. Caulis indumentum ex setis longis crassis pilisque brevibus tenuibus uncinatis constans, in caule supero setis brevioribus sparsioribus rigidioribus curvatis subuncinatis, saepe basi bulbosa. Costae in laminae pagina abaxiali indumentum caulis superi indumento simile, sed setas basi bulbosa manifestius praebens. Calycis tubi indumentum ex setis latis curvatis uncinatis basi bulbosa numerosis, pilis brevibus tenuibus uncinatis abundantibus intermixtis, loborumque indumentum ex setis longioribus tenuioribus non uncinatis. Inflorescentia et calyx et corolla sicut typice in Symphyto L. sed differentiis his peculiaribus: Calyx in gemma 6:8-9:0 mm _ longus. Calyx sub anthesi separatus complanatus 7:1-10-4 mm longus, ad 2/5—1/2(—3/5), plerumque paene ad dimidium, dissectus. Calycis lobi magis minusve triangulares (non linearioblongi), ad apicem acuti vel subacuti. Flos gemmis_ roseirubris corollaque variabiliter colorata, primo alba azureo purpureove diluto maculata vel striata, sub anthesi caerulescenti sed adhuc saepiuscule zonas albas saepeque strias longitudinales purpureas et/vel in lobis maculas praebenti. Corolla separata complanata 17:-4—-23:3 mm longa, quam calyx circiter duplo longior. Corollae squamae anguste triangulares, inclusae. Stamina quam corollae squamae breviora, filamentis antheras nonnunquam aequantibus vel plerumque excedentibus. Stylus multum exsertus. Nuculae 3-9—4-1 mm longae, curvatae, mediocriter brunneae, minute tuberculatae obscureque porcatae. Planta fertilis et seminibus propagata. Floret mense Maio usque ad Julium. 2n = 48. Robust, erect or decumbent perennial herb 50- 50s rem> “tally Swaithy an) oblique » taproot. Decumbent senescent stems often exceeding 200 cm in length. Leaves mid green to slightly grey-green, paler beneath, soft or somewhat bristly to the touch, petiolate, with wings 1-— 7 mm but not decurrent. Basal leaves narrowly ovate to widely lanceolate, with slightly cuspi- date subacute to acute apex and truncate to subcordate base, when overwintering often strongly cordate to auriculate-based; upper cauline leaves broadly ovate, with acute to acuminate apex and truncate to rounded base. Indumentum of stem of long, thick bristles and short, fine, uncinate hairs, with bristles on upper stem becoming shorter, sparser, stiffer, curved and subuncinate, often with bulbous base. Indumentum of midrib of abaxial surface of lamina similar to that of upper stem, but with bulbous-based bristles more pronounced. Indumentum of calyx tube of numerous broad, curved, uncinate, bulbous-based bristles inter- mixed with abundant short, fine, uncinate hairs, and of calyx-lobes of longer, finer, non- uncinate bristles. Inflorescence, calyx and corolla as typically found in Symphytum L. but with the key differences comprising: Calyx in bud 6:8-9:0 mm long. Detached, flattened calyx at anthesis 7-1—10-4 mm long, dissected to 2/5—1/2(—3/5), usually almost to half-way. 374 NOTES Watsonia 27 (2009) Calyx-lobes more or less triangular (not linear- oblong), with acute to subacute apices. Flower with pink-red buds and a variably coloured corolla, at first white blotched or streaked with pale blue or purple, becoming sky blue at anthesis but still often with white zones and frequently with purple longitudinal striations and/or corolla-lobe blotches. Detached, flattened corolla 17-4—23-3 mm long and about twice as long as calyx. Corolla scales narrowly triangular, included. Stamens shorter than corolla-scales, with filaments sometimes equalling or usually exceeding anthers. Style strongly exserted. Nutlets 3-9-4-1 mm long, curved, mid-brown, minutely tuberculate and obscurely ridged. Fertile and spreading by seed. Flowers May to July. 2n = 48. Colour photographs of plants from the Sustead colony can be seen on BSBI News 105 cover and front piece. The epithet is derived from Norvicum, the Latinized name for the city of Norwich. HOLOTYPUS: Roadside verge at Intwood, East Norfolk, v.c. 27, TG/1971.0414, 2 June 2006, R. M. Leaney Accession No. 2007.368.5 (NWH). DISCUSSION The Norfolk comfrey is referable to Section Caerulea Buckn. (Bucknell 1913; Sandbrick et al. 1990), but is distinguished from all other taxa in this Section, including Symphytum savvalense Kurtto (Sell & Murrell 2009), by the combination of the indumentum type, the typically under half-dissected relatively large calyx, the shape of the calyx-lobes, corolla colour and size, and the strongly exserted style. The following morphological characters of the Norfolk Comfrey suggest intermediacy between a taxon of Section Caerulea and S. orientale: the soft indumentum with numerous uncinate hairs; the broadly ovate upper cauline leaves with truncate to rounded bases; the large half-dissected calyx; and the white coloration in the developing corolla. The chromosome number of 2n = 48 suggests S. asperum (2n = 32) rather than a cytotype of S. x uplandicum as the other parent taxon. Either parent with 2n = 32 may have produced unreduced gametes, or one parent may have been S. orientale with 2n = 64, as unpublished results have shown 2n = 62, 63 for S. orientale, which may represent a miscount of 2n = 64 (T. W. J. Gadella, pers. comm., 2009). Resynthesis of the Norfolk Comfrey by experimental crossing of putative parent species should resolve this issue, with results expected in 2010, and allow publication of a full account of this plant. ACKNOWLEDGMENTS The authors would like to acknowledge the advice and contributions from Theo Gadella, Martin Godfrey, John Bailey, and Clive Stace. Philip Oswald provided the Latin translation of the description and Eric Clement provided valuable comments on the manuscript. REFERENCES BUCKNELL, C. (1913). A Revision of the genus Symphytum L. Botanical Journal of the Linnean Society 41: 491-556. SELL, P. & MURRELL, G. (2009). Flora of Great Britain and Ireland. Volume 3: Mimosaceae — Lentibulariaceae. Cambridge University Press, Cambridge. SANDBRICK, J. M., VAN BREDERODE, J. & GADELLA, T. W. J. (1990). Phylogenetic relationships in the genus Symphytum L. (Boraginaceae). Proc. Kon. Ned. Akad. v. Wetensch. 93(3): 295-334. STACE, C. A. (1997). New Flora of the British Isles. 2nd edition. Cambridge University Press, Cambridge. C. L. O’ REILLY 3 Railway Cottages, Lambley, Northumberland CA8 7LL e-mail: clare @ptyxis.com R. M. LEANEY 122 Norwich Road, Wroxham, Norfolk NR12 SSA Watsonia 27: 375-387 (2009) 375 Plant Records Records for publication must be submitted to the appropriate Vice-county Recorder (see BSB/ Year Book 2008), and not to the Editors. Following publication of the New Atlas of the British & Irish Flora and the Vice-county Census Catalogue, new criteria have been drawn up for the inclusion of records in Plant Records. (See BSBI News no. 95, January 2004 pp. 10 & 11). These are outlined below: e First records of all taxa (species, subspecies and hybrids) included in the VCCC, designated as native, archaeophyte, neophyte or casual. e First record since 1970 of the taxa above, except in the case of Rubus, Hieracium and Taraxacum. e Records demonstrating the rediscovery of all taxa published as extinct in the VCCC or subsequently. e Newly reported definite extinctions. e Deletions from the VCCC (e.g. through the discovery of errors, the redetermination of specimens etc.) NB — only those errors affecting VCCC entry. e New 10km square records for Rare and Scarce plants, defined as those species in the New Atlas mapped in the British Isles in 100 10km squares or fewer. (See BSBI News no. 95, January 2004 pp 36-43). Records for the subdivisions of vice-counties will not be treated separately; they must therefore be records for the vice-county as a whole. However, records will be accepted for the major islands in v.cc. 100, 102—104, 110 and 113. In the following list, records are arranged in the order given in the List of Vascular Plants of the British Isles and its supplements by D. H. Kent (1992), from which the species’ numbers, taxonomy and nomenclature are taken. The Ordnance Survey national grid reference follows the habitat and locality. With the exception of collectors’ initials, herbarium abbreviations are those used in British and Irish Herbaria by D. H. Kent & D. E. Allen (1984). Records are field records if no other source is stated. For all records ‘det.’ or ‘conf.’ appear after the herbarium if the determination was based on material already in an institutional herbarium, otherwise before the herbarium. The following signs are used: * before the vice-county number: to indicate a new vice-county record. + before the species number: to indicate that the plant is an archaeophyte. + before the species number: to indicate that the plant is a neophyte. © before the species number: to indicate that the plant is a casual. The above 3 signs may also used before the vice-county number to indicate the status of the plant in that vice-county. ®) before the vice-county number: to indicate that this is an additional hectad for a Rare or Scarce plant. ® at end of entry: established taxon not in Vice-County Census Catalogue. Name of authority provided. [ ] enclosing a previously published record: to indicate that that record should be deleted or changed. Records are now published in two separate sections — 1) NATIVES (including archaeophytes) and 2) ALIENS (neophytes and casuals). The following list contains the first set of records up to and including the year 2008. Further records including the year 2008 will be published in the next issue of Watsonia. 376 PLANT RECORDS ARCHAEOPHYTES AND NATIVES Diphasiastrum complanatum 1/4.2. *$1, Berwicks.: Dirrington Little Law, Cattleshiel moor, NT6853, J. Hardy, 1892, det. E. S. Marshall, Hist. Berw. Nat. Club XIV p.17. Found first by Mrs Dodds of Bowden, formerly Miss Sibbald, collected for James Hardy by John Buckham, shepherd to Mr Craw, Rawburn, sent to W. B. Boyd who sent it on to Rev E. S. Marshall; his reply determines it as ‘Lycopodium alpinum var. decipiens, the L. complanatum of British authors’. Equisetum xdycei (E. fluviatile x palustre) 4/1.4x8. *17, Surrey, on metal enriched substrate by a small stream draining a boggy area, Pirbright Ranges, SU928579, F. J. Rumsey, 2008. Equisetum sylvaticum 4/1.7. ®14, E. Sussex: beside Bluebell railway, W of Tremaines, TQ374266, P. J. Acock, 2008. Ophioglossum azoricum 5/1.2. *93, N. Aberdeen: dune slack, Sands of Forvie NNR, NKO12260, D. Welch, 2007. Possibly previous records of O. vulgatum belong to O. azoricum which formerly was a subsp. of vulgatum. ®110, Outer Hebrides: near building, Cnoc na Liana Moire, Uig, Lewis, NB042375, R. J. Pankhurst, P.A. Smith, J. W. McIntosh & E. Pilling, 2008. Pilularia globulifera 9/1.1. | 10, Wight: recently cleared pool, Bouldnor Copse, SZ380902, G. N. Toone, 2002, det. A. C. Jermy, herb. G. N. Toone. Last recorded 1923, from this area. Dryopteris xcomplexa (D. filix-mas x affinis) 17/3.2x3. *$1, Berwicks.: dean, Nabdean Burn, NT931522, M. E. Braithwaite, 2008. Fronds 160cm, winter-green. Dryopteris affinis (Lowe) Fraser-Jenk. subsp. paleaceolobata (T.Moore) Fraser-Jenk. 17/3.3.pal. | *110, Outer Hebrides: road bank next to Loch Sgailler, NBO835, P. A. Smith & J. W. McIntosh, 2008, det. K. Trewren, E. @ Nuphar xspenneriana (N. lutea x pumila) 26/2.1x2. *58, Cheshire: mere fringe, Hatchmere, S$J5472, S. J. Whild, 2001. Status uncertain. Ranunculus bulbosus 28/13.3.— *91, Kincardines.: riverside grassland, Crathes, NO734961, D. Welch, 2003. Omitted in error from VCCC. Betula pubescens subsp. tortuosa 40/1.2b. *63, S. W. Yorks.: waste gound, Broomhill, SE4102, B. K. Byrne, 2008. Suaeda maritima 43/7.2. *81, Berwicks.: Ramsheugh Bay, NT7772, anon, 1940, WFS Berwickshire County Botanical Record, as ‘Cockburnspath’, possibly casual only, recorder probably Mrs E. K. Swinton. Minuartia hybrida 46/4.5. @8, S. Wilts.: disturbed ground, Stokehill North, ST967520, N. Cope, 2005. Common in nearby military training area. Stellaria neglecta 46/5.4.__ *90, Angus: by path through swamp, Restenneth, NO4851, B. G. Hogarth, 2007. Cerastium pumilum 46/7.11. ®8, S. Wilts.: tracksides, Salisbury Plain ATE West, ST945456, S. Pilkington, 2006. Very local in disturbed ground. Herniaria glabra 46/13.1. £64, Mid-W. Yorks.: neglected area of nursery garden, Scarcroft, Leeds, SE347409, J. Scott, 2008, det. N. Jardine. Ist record since 1970. Spergularia marina 46/17.3. *+§8, S. Wilts.: roadside, A303 trunk road, SU24, S. Leach, 2005. Also recorded along major roads elsewhere in VC8 in 2005. Rumex xhybridus (R. longifolius x obtusifolius) 47/8.8x19. *9(, Angus: arable field adjacent to river, near Pirn Brig, NO5768, B. G. Hogarth & P. Gaff, 2008. Tilia platyphyllos 52/1.1. © *£91, Kincardines.: woodland in den, Durris House, NO798969, D. Welch, 2007. Viola xintersita (V. riviniana x canina) 57/1.4x6. *79, Selkirks.: hillside pasture, Tinnis Top, Yarrow, NT382290, R. W. M. Corner, 2008, herb. R.W.M.C. Viola xcontempta (V. tricolor x arvensis) 57/1.12x13. 8, S. Wilts.: arable land, Bulford, SU198408, J. R. Moon, 2006. Ist record since 1945. Area supports many interesting arable species. Viola tricolor subsp. tricolor 57/1.12a. | ©10, Wight: arable, Coppidhall Farm, Havenstreet, SZ564907, G. N. Toone, 2005, det. M. Hardman, herb. G. N. Toone. Ist confirmed record since 1926. +Salix xmollissima (S. triandra x viminalis) 61/2.4x9. *58, Cheshire: mere fringe, Knutsford, SJ755788, D. Woods & M. Woods, 2006, det. D. Wood. PLANT RECORDS B0i Salix cinerea subsp. cinerea 61/2.1la. *17, Surrey: Merstham, Spynes Mere, TQ307524, B. W. Phillps, 2008. *58, Cheshire: marshy field, Mobberley, SJ800817, G. M. Kay, 2007. Arabis petraea 62/15.1. 91, Kincardines.: silt by River Dee, Durris, NO801986, D. Welch, 2007. Ist record since 1970. Draba muralis 62/21.4. *+90, Angus: on rocky slabs in wooded den, Den of Airlie, NO293521, B. G. Hogarth & P. Gaff, 2007. Ist record in a natural habitat. Cochlearia officinalis subsp. scotica 62/23.3b. ®81, Berwicks.: six scattered plants on rocks behind beach, Horsecastle Bay, St Abbs Head, NT917685, M. F. Godfrey, 2007. Last seen here 1884. Cochlearia officinalis x danica 62/23.3x5. — *81, Berwicks.: one plant in turf behind beach, Horsecastle Bay, St Abbs Head, NT917685, M. F. Godfrey, 2007. Cochlearia danica 62/23.5. *90, Angus: plants in natural habitat frequent along rocky coast north of Auchmithie, NO64, B. G. Hogarth & P. Gaff, 2006. Crambe maritima 62/41.1. 81, Berwicks.: one large plant, not flowering, on shingle beach, Yellow Craig, NT921660, C. Dixon, 2008, conf. M. E. Braithwaite. Ist record since 1836. Raphanus raphanistrum subsp. maritimus 62/42.1b. 110, Outer Hebrides: Bhatarsaigh, NL633940, R. J. Pankhurst, 2007. lst record since 1970. Primula veris 69/1.3. *110, Outer Hebrides: on machair, Eoligarry, Barra, NF709065, J. C. M. Alexander, 2005. Samolus valerandi 69/8.1. *17, Surrey: small plants in South Nutfield Churchyard, TQ302487, G. Fookes, 2008. Crassula tillaea 73/1.1. | *£90, Angus: track between arable and policy woodland, Panmure area, NO532381, P. Gaff, 2008. Umbilicus rupestris 73/2.1. *60, W. Lancs.: Silverdale, SD459749, J. Webb, 2006. Rubus saxatilis 75/8.3. 91, Kincardines.: woodland edge, Inchmarlo, NOQ653972, D. Welch, 2007, herb. D. Welch. Ist record since 1970. Rubus albionis 75/8.34. *71, Man: cliff scrub above beach, Port Garwick, SC434814, D. E. Allen, 2008, BM. Rubus errabundus 75/8.51. *91, Kincardines.: roadside by felled plantation, Garvock, NO748965, D. Welch, 2007, conf. A. L. Newton, herb. D. Welch. Rubus lasiodermis 75/8.108. *10, Wight: relic heath scrub, roadside near Shorwell, SZ461822, D. E. Allen, 2008, det. A. L. Newton. Rubus pistoris 75/8.116. *71, Man: hedge-top outside Union Mills, SC3477, D. E. Allen, 2008, BM. [Rubus leucostachys 75/8.156. 10, Wight. Delete record in Watsonia 27:177. On re- inspection in the field this has proved a misdetermination. | Rubus surrejanus 75/8.161. = *+71, Man: hedgebank of garden, Croudle Old Road, Baldrine, SC427809, D. E. Allen, 2008, BM. Accidental introduction. Rubus vestitus 75/8.163. *90, Angus: A90 Tealing, NO435419, D. Welch, 2004, conf. A. L. Newton. *£110, Outer Hebrides: electricity substation, Carishader, NB102320, E. Pilling & R. J. Pankhurst, 2008, conf. G. H. Ballantyne, E. Rubus wirralensis 75/8.174. *11, S. Hants.: roadside hedge, Chark Common, near Gosport, SU576019, D. E. Allen, E. J. Clement & J. A. Norton, 2007-8, conf. A. L. Newton, BM, HCMS. [Rubus wirralensis 75/8.174. 71, Man. Delete. All existing records (including 5 hectads in Atlas of British and Irish Brambles) relate to the subsequently described R. cumbrensis, det D. E. Allen, conf. A. L. Newton)| Rubus raduloides 75/8.199. *71, Man: open roadside verge, Ballabeg to Grenaby road, SC257258, D. E. Allen, 2008, BM. Rubus anglofuscus 75/8.205. *11, S. Hants.: patch, old hedge line bordering recreation ground, near Eastleigh station, SU462195, D. E. Allen, 2005-6, BM, det. A. L. Newton. Rubus scabripes 75/8.286. *10, Wight: trackside by edge of copse, Westover, SZ4186, E. J. Clement & P. D. Stanley, 2008, conf. D. E. Allen, BM. Rubus conjungens 75/8.307. *71, Man: hedgebank, green lane near Grenaby, SC260722, D. E. Allen, 2008, BM. | Potentilla xsuberecta (P. erecta x anglica) 75/9.13x14. *9(), Angus: in dune slack, Barry Links, NO524322, B. G. Hogarth & P. Gaff, 2007. 378 PLANT RECORDS Potentilla xmixta (P. anglica x reptans) 75/9.14x15. *90, Angus: in dune slack, Barry Links, NO524322, B. G. Hogarth & P. Gaff, 2007. Crataegus xmedia (C. laevigata x monogyna) 75/35.7x8. | *£91, Kincardines.: streamside by playing field, Maryculter, NJ858002, D. Welch, 2007. Vicia parviflora 77/14.7. ®@10, Wight: ex-arable land on chalk, Mersley Down, SZ551872, G. N. Toone, 2007. Ulex europaeus x gallii 77/26.1x2. | *58, Cheshire: heathland, Helsby Hill, SJ491753, M. O. Stead, 2006, det. P. M. Benoit. Epilobium xlimosum (E. parviflorum x montanum) 84/1.2x3. *58, Cheshire: riverbank, Brooklands, Sale, SJ816919, G. M. Kay, 2008, det. G. D. Kitchener. Epilobium xmutabile (KE. montanum x roseum) 84/1.3x7. *63, S. W. Yorks.: footpath, Park Drive, Heaton, Bradford, SE144360, M. Wilcox, 2008, conf. G. D. Kitchener. Epilobium xnutantiflorum (E. roseum x ciliatum) 84/1.7x8. *63, S. W. Yorks:.: Fairweather Green, Bradford, SE133334, M. Wilcox, 2008, conf. G. D. Kitchener. Buxus sempervirens 90/1.1. *+91, Kincardines.: conifer plantation on edge of policies, Banchory, NO683967, D. Welch, 2007. Erodium maritimum 103/2.1. 10, Wight: well established in cracks between cobbles in front of old house by coast, Sprinvale, Seaview, SZ618920, P. D. Stanley, 2008, det. E. J. Clement, herb. C. R. Pope. Ist record since 1955. Origin unknown. *+Smyrnium olusatrum 107/10.1. *90, Angus: East Haven, NO594364, B. G. Hogarth, 2005. Also at NO599368, by burn outflows. Crithmum maritimum 107/17.1. *90, Angus: Rumness, NO691456, P. Gaff, 2005. 3 plants still there in 2008. y+Aethusa cynapium subsp. agrestis 107/20.1b. ®10, Wight: arable, Little Stenbury, Whitwell, SZ524796, G. N. Toone, 2003, det. C. R. Pope, herb. G. N. Toone. ®10, Wight: arable, Cheverton Down, Shorwell, SZ4584, G. N. Toone, 2007. *+64, Mid-W. Yorks.: waste ground, Shipley, SEI51380, B. A. Tregale & M. Wilcox, 2008. Heracleum sphondylium x H. mantegazzianum 107/41.1x2. | *90, Angus: on steep slopes of coastal grassland adjacent to arable, N of Rumness, NO688452, B. G. Hogarth & P. Gaff, 2006. Gentianella germanica 108/5.3. ®8, S. Wilts.: chalk grassland, Martin Down NNR, SU057189, D. & M. Nesbitt, 2006, det. A. R. G. Mundell. *Lamium confertum 118/5.5. *© 63, S. W. Yorks.: heathland, Edge Moor, Huddersfield, SE104126, M. J. Lucas, 2004. Lycopus europaeus | 18/22.1. *81, Berwicks.: Bemersyde Moss, NT6133, C. O. Badenoch, 1973. Also recorded here 1982 & 2002. Salvia pratensis 118/25.3. ®@O7, N. Wilts.: grassland, Stratton Woods, SU165893, A. Deakin, 2006, det. S. Pilkington. Single plant, thought to be introduced in wildflower seed in sown grassland. Callitriche truncata 120/1.2. 15, E. Kent: in 2 ditches, Lydden Valley, Hacklinge, TR349564, C. J. Cadbury & A. Parker, 2008. Also at TR351564, TR080643, TRO082642 & TRO87643. Only East Kent records ever outside Dungeness area. Callitriche platycarpa 120/1.4.*79, Selkirks.: inflow stream, Clearburn Loch, NT345156, T. J. Pankhurst, 1998, NMW. See Lansdown (2008) Callitriche Species of Europe, p. 170. Veronica hederifolia subsp. lucorum 124/16.24b. *31, Hunts.: shady, disturbed ground, Alwalton, TL1396, D. Broughton, 2007. [Hebe cupressoides 124/17.cup. 63, S. W. Yorks.: waste ground, Huddersfield, SE1419, M. J. Lucas, 2007. Error - delete]. Euphrasia rostkoviana 124/20.1. *81, Berwicks.: near Gordon, NT6342, G. C. Druce, 1907, Hist. Berw. Nat. Club XXUI p.228, ex Ann. Scot. Nat. Hist. 1907 p.99. Euphrasia nemorosa x micrantha 124/20.7x18. *81, Berwicks.: Penmanshiel Moor, NT8268, L. W. Frost, 1960, det. P. F. Yeo. BSBI party. EKuphrasia marshallii x scottica 124/20.15x19. *110, Outer Hebrides: rocky knoll on moorland, Little Bernera, NB138411, R. J. Pankhurst, 2008, det. A. J. Silverside, E. Euphrasia heslop-harrisonii 124/20.20. ®110, Outer Hebrides: sheep-mown coastal turf, Rubha Liath, Eriskay, NF8009, P. A. Smith, 2007, det. A. J. Silverside, NUW. Orobanche hederae 125/2.8. _*14, E. Sussex: 5 spikes among Hedera helix, on bank of ditch in garden, Burgess Hill, TQ3119, E. Sears, 2008, det. A. G. Knapp. Confidential location. PLANT RECORDS 379 Utricularia australis 128/2.2. *60, W. Lancs.: pool, Storrs Moss, Yealand Redmayne, SD485755, G. Lyons & M. Ausden, 2007, conf. E. F. Greenwood. Population explosion following sediment removal winter 2005/6. Linnaea borealis 131/4.1. 91, Kincardines.: native Scots Pine woodland, Strachan, NO646899, E. Jensen, 2007. Ist record since 1970. Known here by F. Sheridan, former forester, since 1985. + Valerianella dentata 133/1.4. ®10, Wight: approx. 5 plants in set-aside arable field, Warren Farm, Totland, SZ315854, G. N. Toone, 2002, herb. G. N. Toone. Carduus nutans 135/5.4. *+110, Outer Hebrides: Newton, N Uist, NF885773, R. J. Pankhurst, 2006, E. +Centaurea cyanus 135/11.3. 113(S), Channel Islands (Sark): rough grass near school/ village hall, WV464761, S. Synnott, 2008, herb. Société Sercquiaise. Garden escape. Ist record for Sark. Centaurea xmoncktonii (Centaurea nigra x jacea), 135/11.7xjac. 14 Es Sussex: 5 = 10 plants in old car park entrance on north side of Warren Road, Brighton, TQ334056, A. Spiers, 2008. Ist record since 1970. *+Cichorum intybus 135/12.1. 91, Kincardines.: in policies - self-seeded from garden, Banchory House, NJ915023, D. Welch, 2007. Ist record since 1970. Taraxacum subnaevosum 1|35/25.46. *110, Outer Hebrides: roadside verge A888, Grean, Barra, NF682034, R. J. Pankhurst, 1998, det. A. J. Richards & A. J. Dudman, E. Taraxacum bracteatum 135/25.55. *31, Hunts.: damp grassy ride in ancient woodland, Archer’s Wood, TL172812, D. Broughton, 2008, det. A. J. Richards. Taraxacum hesperium 135/25.66. *110, Outer Hebrides: rough grass by ruin, Boisdale, S. Uist, NF741191, R. J. Pankhurst, 2008, det. A. J. Richards, E. [Taraxacum hamatiforme 135/25.87. 31, Hunts.: location as given in Taraxacum database is VC29.] Taraxacum hamatum 135/25.89. *31, Hunts.: ancient woodland ride, Waresley Wood NNR, TL262545, D. Broughton, 2008, det. A. J. Richards. [Taraxacum hamiferum 135/25.90. 31, Hunts.: location as given in Taraxacum database is WiG293| Taraxacum hamiferum 135/25.90. *63,S. W. Yorks.: Kings Road, Bradford, SE167356, B. A. Tregale, 2000, det. A.J. Richards. [Taraxacum lamprophyllum 135/25.92. 31, Hunts.: location as given in Taraxacum database is VC29.] Taraxacum pseudohamatum 135/25.98. *110, Outer Hebrides: causeway verge, Creagorry, Benbecula, NF801506, R. J. Pankhurst, 2008, det. A. J. Richards, E. Taraxacum quadrans 135/25.99. *110, Outer Hebrides: grassy meadow, Griminish, Benbecula, NF/68518, R. J. Pankhurst, 2008, det. A. J. Richards, E. Taraxacum spiculatum 135/25.100. *90), Angus: road from Kirriemuir - Glen Prosen, NO363579, R. Pankhurst, 2003, conf. A. J. Richards. Taraxacum ekmanii 135/25.131. | *110, Outer Hebrides: roadside by river bridge, Milton, S. Uist, NF744262, R. J. Pankhurst, 2008, det. A. J. Richards, E. Gnaphalium luteoalbum 135/33.6. | *@©55, Leics.: in pavement cracks and on roadside, Gaol Street, Oakham, SK859087, A. Lear, 2009, conf. K. Walker. 23 flowering plants and up to 75 seedlings. Arum maculatum x italicum, 147/5.1x2. | 113(S),Channel Islands (Sark): N side of track at foot of hedge, E of Pilcher Monument, WV455755, R. A. Waterman & C. Helyar, 2007, det. R. A. Waterman. R. A. Waterman has decided that all Sark plants (except this one) previously considered hybrids are atypical A. maculatum. Juncus xsurrejanus (J. articulatus x acutiformis) 151/1.13x14. 113(S), Channel Islands (Sark): wet parts of two adjoining fields, S of Eperquerie Common, WV464769, R. M. Veall, 2008, det. M. Wilcox, herb. Société Sercquiaise. Numerous plants previously recorded as one of the parents.1st record for Sark. Juncus xdiffusus (J. inflexus x effusus) 151/1.25x26. *90, Angus: The Lurgies, Montrose Basin, NO675579, B. G. Hogarth, 2008. Juncus xkern-reichgeltii (J. effusus x conglomeratus) 151/1.26x27. | *64, Mid-W. Yorks.: mire at 390m, Ingleborough, SD743759, M. Wilcox, 2008. 380 PLANT RECORDS Luzula xborreri (L. forsteri x pilosa) 151/2.1x2. 10, Wight: ancient woodland, Nunney’s Wood, Ningwood, SZ403895, P. D. Stanley, 2008, det. G. M. Kay, herb. C. R. Pope. Growing with both parents. 1° confirmed record since 1883. Cyperus fuscus 152/11.3. *8, S. Wilts.: ditches and marsh, Breamore Marsh, SU1518, C. Chatters, 2002. Edge of New Forest but Ist record for VC8. Records have been passed to VC11 VCR since 1983. Much monitored population. Carex muricata x divulsa 152/16.8x9. — *10, Wight: roadside verge, Mottistone, SZ4083, P. D. Stanley, 2008, conf. M.S. Porter & M. J. Y. Foley, herb. P. D. Stanley. Ist record for England. Carex maritima 152/16.14. | ®110, Outer Hebrides: Traigh na Clibhe, Lewis, NB083362, A. H. Wilson, R. J. Pankhurst & E. Pilling, 2008. Carex lasiocarpa 152/16.24. 58, Cheshire: fen, Wybunbury Moss NNR, SJ6950, I. Diack, 2008. Last seen here c. 1905. Only current site. Carex pendula 152/16.31. *90, Angus: Backmuir Wood, NO340344, B. G. Hogarth, 2009. Has persisted for several years - probably garden escape. Carex viridula subsp. brachyrrhyncha 152/16.46a. *10, Wight: small base-rich fen, Dolcoppice, Whitwell, SZ5093, J. Norton & D. E. Allen, 2008, det. A. O. Chater, herb. J. Norton. Remarkably, there are no historic records for this taxon. Festuca gigantea 153/12.3. *+110, Outer Hebrides: Carishader, Lewis, NB102320, R. J. Pankhurst & E. Pilling, 2008. Festuca rubra subsp. litoralis 153/12.7c. *©63, S. W. Yorks.: marshy ground near supermarket, Waterloo, Huddersfield, SE178164, M. J. Lucas, 2000. Festuca rubra subsp. arctica 153/12.7e. | 110, Outer Hebrides: gravelly slip on bank of river, Abhainn a’ Bheannain, Lewis, NB050294, P. A. Smith & M. Teneva, 2008, conf. A. Copping, NMW. Ist record since 1970. AC describes this as a very convincing subsp. arctica. ®110, Outer Hebrides: coastal turf, Pabaigh Mor, NB1038, P. A. Smith, 2008, conf. A. Copping, E. xFestulolium loliaceum (Festuca pratensis x Lolium perenne) 153/12x13.1x1. *90, Angus: in arable field adjacent to The Lurgies, Montrose Basin, NO675572, B. G. Hogarth, 2008. xFestulolium holmbergii (Festuca arundinacea x Lolium perenne) 153/12x13.2x1. = *10, Wight: roadside verge, Barton’s Corner, Shalfleet, SZ4189, P. D. Stanley, 2008, det. T. A. Cope. Puccinellia distans 153/16.2. *r8, S. Wilts.: roadside, Deptford A36, SU009384, S. Pilkington, 2006. Also recorded elsewhere in VC8 in 2006. *+Briza minor 153/17.2. ®10, Wight: frequent along field margin, Stenbury Manor Farm, Whitwell, SZ520796, G. N. Toone, 2004, det. C. R. Pope, herb. G. N. Toone. Poa infirma 153/18.1. ®13, W. Sussex: several plants on low bank by cricket field, N of Broadbridge Heath, TQ149322, A. G. Knapp & A. G. Hoare, 2008, det. A. G. Knapp. Poa compressa 153/18.9. *81, Berwicks.: garden wall, Swinton, NT8147, E. K. Swinton, 1952. (Accepted, but could have been P. nemoralis.) Deschampsia setacea 153/32.2. ®110, Outer Hebrides: unnamed lochan, Lewis, NA996218, P. A. Smith & M. Teneva, 2008. Gastridium ventricosum 153/42.1. ®10, Wight: set-aside arable land, Cridmore Farm, Rookley, SZ5080, G. N. Toone, 2002, det. C. R. Pope. Polypogon monspeliensis 153/46.1. ©8, S. Wilts.: arable land, Farley, SU214283, J. Ford, 2006, det. S. Pilkington. Ist record since 1970. Bromus racemosus 153/50.3. _*+81, Berwicks.: Anton’s Hill, NT7843, E. Bell, 1853, det. G. Johnston, Nat. Hist. Eastern Borders, G. Johnston, with four other records by three recorders. Wrongly excluded from VCCC. Bromus hordaceus subsp. thominei 153/50.4c. ®10, Wight: maritime cliffs, Watershoot Bay, Niton, SZ493754, G. N. Toone, 1999, det. L. Spalton, herb. G. N. Toone. Molinia caerulea subsp. arundinacea 153/63.1b. *31, Hunts.: fen birch woodland, Holme Fen NNR, TL209889, D. Broughton, 2008. Sparganium erectum subsp. oocarpum 154/1.1d. *63, S. W. Yorks.: Broomhill, Barnsley, SE4102, M. Wilcox, 2008. [Pontederia cordata 157/1.1. 58, Cheshire: pond, Marston, SJ664700, G. M. Kay, 1989. Delete. Error for Calla palustris] Epipactis phyllanthes 162/3.7. ®64, Mid-W. Yorks.: calcareous grassland, Ellington Banks MoD site, SE280735, Yorks. Nats. Union, 2008. PLANT RECORDS 38] Spiranthes romanzoffiana 162/7.3. 110, Outer Hebrides: Loch na Creige Glais, Benbecula, NF811485, F. Horsman, 2008. 1 flowering plant each side of the loch. Gymnadenia borealis 162/16.Ic. *110, Outer Hebrides: heath, Fuday, NF740083, F. Horsman, 2008, conf. R. M. Bateman. Dactylorhiza fuchsii subsp. hebridensis x traunsteinerioides 162/18.1x7. *110, Outer Hebrides: Luskentyre, S. Harris, NG068998, F. Horsman, 2008. Dactylorhiza fuchsii subsp. hebridensis x ebudensis|62/18.1xebu. *110, Outer Hebrides: one plant Machair Robach, N. Uist, NF862757, F. Horsman, 2008, conf R. M. Bateman. @ This hybrid is new to science. Dactylorhiza maculata x traunsteinerioides 162/18.2x7. | *110, Outer Hebrides: Luskentyre, S. Harris, NG068998, F. Horsman, 2008. Dactylorhiza incarnata subsp coccinea 162/18.3b. *90, Angus: Easthaven - Elliot, NO610376, B. G. Hogarth & P. Gaff, 2005. Reported in 2004 but no details. Dactylorhiza incarnata subsp. coccinea x purpurella var. purpurella 162/18.3bx5. @®110, Outer Hebrides: Uig machair, Lewis, NB103356, F. Horsman, 2007. 2 plants. Dactylorhiza incarnata subsp. coccinea x traunsteinerioides 162/18.3bx7. *110, Outer Hebrides: in series of calcareous flushes, Borve, S. Harris, NGO31947, F. Horsman, 2008. Dactylorhiza incarnata (L.) Soo subsp. coccinea (Pugsley) Soo x D. ebudensis (Wief. ex R.M. Bateman & Denholm) P. Delforge 162/18.3bxebu. *110, Outer Hebrides: Machair Robach, N. Uist, NF861757, F. Horsman, 2008, conf R. M. Bateman. @ Dactylorhiza incarnata subsp pulchella 162/18.3c. | *90, Angus: unimproved damp pasture with scrub, Woodside, NO435541, L. Farrell, 2005. Dactylorhiza traunsteinerioides 162/18.7. © ®110, Outer Hebrides: Machair Robach, N. Uist, NF868759, F. Horsman, 2008. Approx. 50 flowering plants in an isolated Schoenus flush. ®110, Outer Hebrides: Luskentyre, S. Harris, NB069001, F. Horsman, 2008. Orchis purpurea 162/20.5. *14, E. Sussex: one plant in chalk grassland, Beachy Head, Eastbourne, TV569954, J. Curson, 2008, det. D. C. Lang. Status uncertain but thought to be of continental origin. NEOPHYTES AND CASUALS tPteris multifida 8/1.mul. *17, Surrey: several plants on the wall of South Dock, Rotherhithe, TQ365790, N. H. Bertrand, 2008. @ See BSBI News 73. p. 40. +Matteuccia struthiopteris 16/1.1. *91, Kincardines.: streamside, Durris House, NO7996, J. S. Murray, 1962, ABD. Not submitted in 1960s. Seen again 2007 - Ist record since 1970. tPolystichum munitum 17/l.mun. *58, Cheshire: dumped soil heap, Marple, $J945886, P. Owen, 2007, det. F. J. Rumsey, BM. tAbies grandis 20/1.2. *©7, N. Wilts.: in layby, Savernake Forest, SU221672, J. Oliver, 2006. tAbies procera 20/1.3. *90, Angus: mosaic of native woodland and policy woodland, Den of Airlie, NO2950, B. G. Hogarth & P. Gaff, 2007. Picea glauca 20/4.gla. *63, S. W. Yorks.: N. side of Underbank Reservoir, SK2499, J. P. Flanagan, 2007. +Picea pungens Engelm. 20/4.pun. *63,S.W. Yorks.: Wyming Brook, Sheffield, SK265870, Sheffield Wildlife Trust, 2001. @ tThuja plicata 21/3.1. *90, Angus: semi-wild policy in former wooded den, Mains of Auldbar, NO574583, B. G. Hogarth & P. Gaff, 2008. tAraucaria araucana 22/1.1. | *90, Angus: semi-wild policy in former den woodland, Mains of Auldbar, NO573582, B. G. Hogarth & P. Gaff, 2008. tNymphaea marliacea Wildsmith 26/1.mar. *©7, N. Wilts.: pond, Savernake Forest, SU221651, Wiltshire Botanical Society, 2006, det. S. Pilkington. @ Extensive in Bitham Pond. +Helleborus orientalis 28/3.3. *80, Roxburghs.: one clump on wooded bank above river, Hornshole Bridge, Hawick, NT532368, P. R. Green, 2008. +Helleborus argutifolius 28/3.arg. 113(S), Channel Islands (Sark): verge of track near foot of wall, Valette de Bas, WV472760, R. M. Veall, 2008, herb. Société Sercquiaise. Origin unknown. Ist record for Sark. ONigella damascena, 28/5.1. | *31, Hunts.: railway track, Woodston, Nene Valley Railway, TL1898, D. Broughton, 2005. 382 PLANT RECORDS tActaea erythrocarpa Fisch. 28/8.ery. *©81, Berwicks.: one plant in fruit on old road between Old Free Kirk and the bridge, Gavinton, NT7652, M. J. Long, 1965, det. A. G. Long. @ +Anemone apennina 28/9.2._ ©7, N. Wilts.: grassy embankment, Great Bedwyn, SU273647, J. Oliver, 2006. Ist recent record. One site for VC7 given in 1957 flora but not dated. +Berberis darwinii 29/1.9._ *10, Wight: established small bush well away from gardens on railway embankment, Yarbridge, Brading, SZ607867, C. R. Pope, 2008. +Eschscholzia californica 30/6.1. *7, N. Wilts.: wall-road angles, West Overton, SU129679, J. Oliver, 2006. Also recorded at West Overton SU128680 same date. tDicentra formosa 31/1.1. *90, Angus: survey of the Miley with SWT support group for Local Change, Dundee, NO33W, B. G. Hogarth, 2003. Also seen by BGH in Glen Prosen village at NO328658 in 2008. +Dicentra spectabilis Lem. 31/l.spe. © *Mid-W. Yorks.: woodland, Gilstead, SE121396, B. A. Tregale & M. Wilcox, 2008. @