come e Bait: irae = 03! Wee ee ae Bay Breen iit an r PAA pee a Hiviny ' foie we ON OI Ee RO neat ie Ener et eet PO tod Bie er ae» We et ayte Myth tl fopcen y Mg Vh esti bind e am atryavait sf allitn Litpn vis ty eve d'y USAW EN ge P yath a8 hanoen re Hedy mmbin NS HPN Cyy Pye UNA ee ENT NG FP 6, PELE ES Wow lt en Oy ¥ EA vd LORIE Ay he OL ie ae toy ‘ yobs wt Slt es Sv entt'y 45 Gy Nee 8! Loy ieee nv Nf SGA SDE IEIS FON ZY sob DN Redvieel wat p ‘ ey y PAU Ne 1" Coyvewa oy “ O} Ce ee THT ef, 1p (fi fais Ua 7 aM bus. BMS UV On aes ‘S Pree on adel Sty Syste ite CPt ine ssiahodialar nd 00 cof : Weebl g 5p Wee Py rrent tanita ite FE Mee Silt see nt Can TRUCE WHOL ta ag wy, Hance OD ATOMS woe Bg ey ns A adie, Ae prea aed ty Bye (tafe! ee oo nw Wye OLA DY a UA rae Vrogitiy ie a ha, wos ‘ Reena leer 1235, Dieh hala uty diy Leta dl Rrra U thy wy ‘ sea iy eet bias Ey Sitn nee, Panter Hany yl ge WYVE Eere € Neretiee ame 94 4408 MN ws ath Omit nro Per SA et 98 ands alps pee tis WAS DAN oe Died n istonna ng bd sclenahahans 7 sgl alee Gay Hey ee ew 2a Mh Ayia dP ht ) aN ae WAN froth hor Hels oh abr tan bad ee EAN Ate SEN Ue HY LEH A ee Lab Gch i AL SE alt UNV al gs Ke) ‘ Ma Ap tek AWAY AME SE UME OT P wae dt 4 NOP UE eM rey water OM SERN Redeye WD by way ghia Youu TRA aC SON cae sain h Volt Mites PE ES ad eS je be ba pte. y eV ONLL pe 2 Hh et yhac i Pen ex PR are Te ero We Us Lew ir aa) 1.0 ced dean Dp et A NDS IB MNLS for DP hale ater oe ti, wets eH gdm on APY aioe ae Mt ee gied ey aed cont ot 7 Bie AN i. Halk dsbelneedteaeh Ut diya be 8 Ef ON ad Gol BRP ASR NE? Hate aes isin SO Math a Koowiy Ahlen AL bavi ted nN Wier eat aera Lg evn Ste haa d nowy ae Nathaly! ve Up end a anya) SO Veh yas DoPay sta ibaarbact ala Nin ha Serta tsiidset as ndadd alta tdi aes SNS a UMN a hy Mel, eal gd r Ae vets AEN GL Ayivaeee iN sit SP ataal a jain! Nie ir 4 os. 7 os “ad Rpm NN at a Naa he sae Lee wanes wit sew spacer Span i ier rt i s ar Fabtln tant’ ‘ Bo Bedi, a ‘2 arene “ ai alln = yh Se ae te Pamibin (haat petite Ota faeoa yrs! aS adieu jayne am eX babe ent ee” eer exh rire Cr ied oh Ychho ‘alg ae er oat yest Me eer —d . - “WATSONI A wa4ed Journal of the &=st+. Botanical Society of the British Isles =; Is te x es ag N Douglas H. Kent Memorial Issue Volume 23 Part 1 February 2000 Editors: M. Briggs, D. L. Kelly, D. R. McKean, D. A. Pearman, M. N. Sanford, D. A. Simpson Botanical Society of the British Isles Patron: Her Majesty Queen Elizabeth the Queen Mother 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 1999-2000 President, Mrs M. Briggs President elect, Dr G. Halliday Vice-Presidents, Dr R. J. Gornall, Mr M. Walpole, Mr D. J. McCosh Honorary General Secretary, Mr R. G. Ellis Honorary Treasurer, Mr M. E. Braithwaite Editors of Watsonia Papers and Notes, D. L. Kelly, D. R. McKean, M. N. Sanford*, D. A. Simpson Book Reviews, D. A. Pearman Obituaries, M. Briggs *Receiving editor, to whom all MSS should be sent (see inside back cover). © 2000 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 Potentilla anglica Laich. (Trailing Tormentil) was drawn by Rosemary Wise. Douglas H. Kent Memorial Issue of Watsonia This issue of Watsonia is dedicated to the memory of D. H. Kent (1920-1998). It is fitting that the journal of the Botanical Society of the British Isles should commemorate his life and work, for Duggie served the society with dedication for a period of 50 years. Many of the papers in this issue are directly related to Duggie’s botanical interests, including his life-long studies of the flora of Middlesex, his detailed accounts of the spread of Senecio squalidus in Britain and Ireland and the important reference works he prepared. The latter include British herbaria (1957), which was later replaced by British and Irish herbaria by D. H. Kent & D. E. Allen (1984), the B.S.B.I. Handbook covering the Polygonaceae, Docks and knotweeds of the British Isles, written with J. E. Lousley (1981) and the List of the vascular plants of the British Isles (1992). We hope that this wide range of papers will be a fitting tribute to the memory of Duggie Kent. It demonstrates that the subjects he studied are still engaging the attention of British and Irish botanists, who find that his work provides a firm foundation on which they can build. It is also encouraging to see how new lines of evidence such as those provided by molecular techniques are now throwing new light on the subjects he investigated. We are grateful to Professor C. A. Stace for originally suggesting that we produce this memorial issue, and to the authors who have contributed to it. The fact that so many were willing and indeed anxious to contribute papers at relatively short notice says much for the esteem in which Duggie Kent was held by the botanical community. C. D. Preston M. N. Sanford Chairman, B.S.B.I. Publications Committee Receiving Editor, Watsonia ; : = 7 -_ f NS 7 x os 5 y ff 2 Se Watsonia 23: 3—8 (2000) 3 Obituary DOUGLAS HENRY KENT (1920-1998) With the death on 9 September 1998 of Douglas Kent - “Duggie”, or sometimes “Doug”, as he was known to his friends and to British botanists more generally - the B.S.B.I. has lost the last surviving member of the team whose devoted efforts largely sustained the Society at a high level of efficiency throughout its great expansion of the early post-war years, and who were responsible for putting in hand many of the activities that continue to be among its principal features today. Though too self-effacing by nature to enjoy the prominence of the pair with whom he formed such a memorable triumvirate, J. E. Lousley and J. G. Dony, his services to the Society were fully as great and no less varied while also extending over the even lengthier period of 48 years without a break. Duggie was born on 21 March 1920 in south Hammersmith, the only child of Ernest Edmund and Millie Louise Kent, and from the age of six lived for the rest of his life at 75 Adelaide Road in the nearby Borough of Ealing. After attending Bordeston Secondary School in Boston Manor, where he was a contemporary of Charles Hawtrey of ‘Carry On’ fame, he left at fifteen to join the Gas, Light and Coke Company (subsequently, after nationalisation, to constitute the core of North Thames Gas) and remained with that for his entire career, latterly heading a section responsible for mains and pipelines supplies. The fact that his home and work place were both within easy reach of the libraries and collections at Kew and South Kensington was to be an inestimable boon and substantially influenced the pattern of his life. A born collector, Duggie might easily have been lost to botany because of competing youthful passions for philately and train-spotting. Luckily, however, while those persisted (he left a sizeable stamp collection and many notebooks on the G.W.R.), they failed to suppress a developing interest in plants. By the age of nineteen botany had taken so firm a grip that he embarked upon the amassing of records for his home county of Middlesex, drastically altered by the relentless sprawl of London since the days of Trimen and Dyer’s Flora of 1869. To that end he set about exploring the promising habitats that still remained and spent many hours in libraries investigating the work and lives of the many who had preceded him in that task. At no other time in his life were his burrowings to take him so extensively into the manuscript domain, in particular the vast Sloane collection in what was then the British Museum at Bloomsbury. The outbreak of war in 1939 fortunately provided no serious interruptions to botany. Military service was ruled out for him (due to ill-health and/or his ‘reserved’ occupation) and, apart from serving as an air raid warden (no light responsibility during the London blitz), he was free to slip out of the city in summer as often as he could manage. That was not without its wartime risks, as several of his botanical contemporaries also discovered. On one occasion his fieldwork accidentally took him a foot or two inside the perimeter of an army camp; an over-zealous N.C.O. promptly shouted to a sentry to arrest him and he ended up with that day’s haul emptied out of his vasculum on to the guard-room floor, to the picturesque bafflement of his captors. Towards the end of the war his botanical activity took a marked leap forward: he began a herbarium in 1943 and joined both the London Natural History Society and the Botanical Society and Exchange Club (as the B.S.B.I. was then still known) in 1944. Around that time he gained his introduction to Lousley, who was to become his friend and special mentor, and to another future close friend and collaborator, E. B. Bangerter. He also met and was given much encouragement by the veteran H. W. Pugsley, the most respected amateur taxonomist of British flowering plants of his generation. Austere and rather forbidding on first acquaintance, Pugsley was ever on the look- out for youthful promise and invited his protégé to dinners at his Wimbledon home, occasions which Duggie still spoke of with awe half a lifetime later. It was also in that same year, 1944, that marriage to Ivy Minnie Austen brought his one lasting connection to somewhere other than Middlesex. They had met earlier when Ivy, a nurse, had been attending to him during a spell in hospital in London. After marriage, Ivy moved to 75 Adelaide Road and they lived on the upper floor over his parents. Their only child, Michael, was born two 4 OBITUARY D. H. KENT years later. His wife came from Blean, near Canterbury, and from 1943 vice-county 15 began to find reflection in his incipient herbarium. On his visits there from 1945 to 1947 he was also able to give much help to F. Rose, then actively engaged on a projected Flora of Kent, and the two often made forays together both there and in Middlesex. It must be a matter for regret that the considerable expertise he acquired as a field botanist was never deployed more widely. Although holidays were subsequently to take him to several parts of Britain (Devon and Cornwall, Derbyshire, Shropshire, Yorkshire, the Lake District, mid and north Wales, and in 1971 to Scotland), those were more or less strictly family affairs, except that plant lists from many of these destinations found their way into his notebooks. He never once travelled overseas and very rarely attended a field meeting or conference if it involved a night away from home. The London N.H.S. was the body with which he principally identified to begin with. Three years after joining, his first paper, reporting the many records of interest made in Middlesex since 1939, appeared in its journal. The year after that he was elected Chairman of its large Botanical Section, and subsequently he agreed as well to act as the Society’s Honorary Curator; he relinquished this post in 1955 but continued to act as curator until 1961. His most conspicuous involvement, though, was as co-author with Lousley of an updated list of the vascular plants recorded within the L.N.H. S.’s official area (20 miles radius from St Paul’s Cathedral), brought out serially between 1951 and 1957 as supplements to The London Naturalist. Increasing absorption into the B.S.B.I.’s affairs, however, compelled him to resign both his offices in 1955 and thereafter his L.N.H.S. links were largely confined to continuing to use its journal as the outlet for his steady stream of papers on the flora of Middlesex or Inner London. Involvement with the B.S.B.I. followed at first a similar pattern. He was soon reporting in Plant Notes aliens new to the British list that he and others were then turning up in the grounds of a soya processing factory at Harefield, and following those up with papers in quick succession on the Middlesex flora. In 1948 the first list of Local Secretaries and Recorders to appear after the war showed him as having agreed to act in those two capacities for v.c. 21 (he was still its Recorder at the end of his life, the longest to have served a vice-county in either capacity to date), and in March 1950 he was rated an active enough member to be put forward for and elected to Council. Three years later, at the instance of Lousley, who had meanwhile become Honorary General Secretary, he was asked to step in as Honorary Assistant Secretary when W. R. Price had to resign at short notice due to the ill health of his wife. The duties attached to this post were twofold: dealing with applications for membership, and the much more onerous one of editing the original Year Book, then about to be published for the fifth and last time. In his editorial to that final issue, in September 1953, Duggie was able to announce that a twice-yearly Proceedings was taking its place. The editorship of that now fell to him. The new journal, besides publishing papers, book reviews and obituaries, absorbed all the matter that had appeared in the Year Book, such as reports of field and exhibition meetings, and from Watsonia it took over Plant Records, Plant Notes and Abstracts from Literature. It was soon apparent that his natural meticulousness made him an effective editor and, while he interpreted that role for the most part responsively, prepared to publish most of the material that flowed in, he refused to be dictated to by the Publications Committee about the journal’s contents and all along insisted on his editorial prerogative to accept what he saw fit. He likewise resisted pressure to change to a firm of printers with a less antiquated type-face, rightly fearful of the damaging increase in cost that that would involve and personally attuned to the relaxed way in which Buncle’s operated (which included accepting copy in his exceptionally readable handwriting). That twelve years’ spell as editor of, Proceedings had one outstandingly important by-product: it confirmed his enduring commitment to abstracting. He had displayed a penchant for that as early as 1950, in a one-off contribution to The London Naturalist. Two years later he proved willing to take over the burden of the B.S.B.I.’s annual equivalent in Watsonia from A. H. G. Alston’s by then sagging shoulders. As soon as Abstracts was transferred to Proceedings he took the opportunity to make the coverage more complete and its appearances more frequent (twice yearly), convinced that through this service alone the Society would attract considerably more overseas members. Though a number of others, most notably E. B. Bangerter for a period, assisted him with contributions, he realised that the bulk of the work had to be done by a single individual if the coverage was to be other than fragmentary and over-selective. When he eventually handed over Proceedings to E. F. Greenwood, in 1966, he was fortunately content to carry on with abstracting. When, only three years later, the Society found it no longer OBITUARY D. H. KENT 5 possible financially to go on supporting two journals, compelling the absorption of Proceedings by Watsonia - a move hastened, ironically, by the ballooning of the Abstracts section consequent upon the freeing of Duggie from his editorial reins - it was universally agreed that this much- appreciated service that he had built up so assiduously must be preserved come what may. By switching to a cheaper mode of production, dropping the attempt to cover plant ecology and the flora of other western European countries, and publishing no more than bare titles in many more cases, BSBI Abstracts was thereupon turned into a feasible proposition, makings its first appearance in May 1971 and lasting (so far) for 28 years. This must have done much to allay the deep disappointment he felt at the abandonment of the journal on which he had laboured so long and so hard. When it is remembered that for 46 years Abstracts has been the work of essentially just one person, a person moreover unable to read any language other than English, the sheer magnitude of his effort and achievement in that direction is little short of stunning. His unrivalled knowledge of the literature made Duggie the obvious choice to survey progress in British plant taxonomy for two early Flora Europaea conferences. It was Duggie’s desire for as complete a coverage as possible in his Abstracts that led him to begin the weekly trips to the library at Kew, a habit which soon turned him into such a familiar figure in that institution. There he made many friends and began a long association with R. K. Brummitt, with whom he discussed matters of common nomenclatural and abstracting interest; they were co-compilers of Index to European Taxonomic Literature for 1969 and 1970. His work at Kew proved of such mutual benefit that Duggie was appointed to an Honorary Research Fellowship at Kew in 1995, an honour very rarely bestowed on an amateur botanist. He also became equally well known and formed many enduring friendships at The Natural History Museum, in his later years visiting it once a fortnight and Kew once a week. Despite his enormous devotion to BSBI Abstracts, this was very far from being his only activity on behalf of the Society. Until 1965 he combined editing Proceedings with continuing to shoulder a whole range of unglamorous ‘back-room’ chores of the kind so crucial to the efficient functioning of any society of this kind and size. These included handling applications for membership, keeping up the address list and dealing with orders for the Society’s publications and enquiries about advertising in them, while also serving on Council ex officio and on three of the six permanent working committees. And, though committees were never his forte and he was seldom vocal on them, he was always most conscientious in attending their meetings. It should not be overlooked either that his name featured on the Panel of Referees and Specialists for several subjects and genera. As a further service he also compiled a card-index of books desired by members, and he invited enquiries. The very least the Society could do, by way of recognition for this unparalleled performance, was to elect him to Honorary Membership - and that it duly did at the A.G.M. of 1966. Meanwhile, quite apart from a steady flow of papers in journals, he was busily accumulating material for several more major works. As he would have been the first to recognise, he was an almost compulsive compiler: he liked nothing so much as hunting down and extracting innumerable items of information and creating out of the resulting mass a useful volume of reference. He worked fast because he was content to plough shallowly, his one failing being that he was over-inclined to take statements on trust. It was the process of accumulation that he particularly revelled in: arrangement and analysis came very much second and third. When he first took over responsibility for the membership address lists, he resisted as long as he dared the called-for regular repeats of the elaborate geographical versions of those lists produced and published by his two immediate predecessors, in which members were listed under groups of the vice-counties in which they lived. He did eventually produce one other such list, but never another. British Herbaria, “an index to the location of herbaria of British vascular plants with biographical references to their collectors”, was the first of his purely reference works to appear in print. Embarked on in 1949, initially with A. J. Wilmott as intended co-author, it was published by the Society in 1958. It filled a long-felt gap in the literature, and its system of assigning each institution holding one or more collections a distinguishing set of letters, designed so as to complement those used in Index Herbariorum (whose compilers later acknowledged much assistance from Duggie), was promptly adopted for standard use in all the Society’s publications. Together with its considerably larger replacement of 1984, British and Irish Herbaria, which was compiled jointly with D. E. Allen, this was more complicated than Duggie’s other undertakings in 6 OBITUARY D. H. KENT requiring much of the information to be collected by postal questionnaire and by the need to distinguish between a person’s actual herbarium and the presence of merely some sheets of his or her collecting in another institution. Of necessity, too, both volumes were the fruit of a good deal of collaboration, but it is no secret that the bulk of the work was his in each case. At the time of the first survey many museums were disgracefully vague about what they possessed and who had collected their material. Between the two surveys that situation greatly improved and the mass of fresh information that consequently mounted up rendered the second edition inevitable. During the 1950s, while engaged on British Herbaria, Duggie simultaneously gave extensive assistance to the American botanist S. F. Blake in connection with the latter’s second volume, covering Western Europe, of Geographical Guide to Floras of the World, published in 1961. Duggie’s own Index to Botanical Monographs (1967) was similar in character, if much more narrowly focused: “a guide to monographs and taxonomic papers relating to phanerogams and vascular cryptogams found growing wild in the British Isles”. Collected together in this are nearly 1,900 references from the year 1800 onwards. All this time the great mountain of material he had built up on the vascular plants and bryophytes of Middlesex had been left in limbo, with no apparent prospect of ever achieving print. Friends at Kew now made its existence known to The Ray Society and as an avowedly historical Flora the work was at last made publicly available in 1975. Duggie later served on the Council of The Ray Society, from 1984 until his death. By that time a Supplement to the Historical Flora, compiled over twenty years with characteristic meticulous attention to detail, had gone to press. This was expected to be published by The Ray Society in early 2000. A small flood of further books in which he had a greater or lesser hand then followed. Docks and Knotweeds of the British Isles, which he co-authored with Lousley as the third of the B.S.B_I. Handbooks, was the first to come, in 1981. This was originally to be a monograph by Lousley alone, but it was rescued by Duggie after Lousley died in 1976. Lousley, being well-known as an expert in Rumex, had largely written the (for him) more difficult accounts of all the other genera first, which left, somewhat perversely, the account of Rumex for Duggie to compile unaided. A second edition, being undertaken by J. R. Akeroyd, will benefit from copious notes left by Duggie for just such a publication. In later life, especially after retirement in 1980, he had time at last to take on editing work in further directions. In 1976 the Richmond Publishing Company secured him as one of the six founding editors of its periodical Natural History Book Reviews, and in each of the first nine issues - before the publisher changed and the editorship was concentrated in one person only - up to five and once as many as nine pieces appeared over his name. He also clearly had a hand in enlisting just as many from botanical friends. Having been engaged by Ebury Press to vet the nomenclature of W. Keble Martin’s The Concise British Flora in Colour, when that started out in 1965 on its astonishing climb on to the bestseller lists, Duggie was later engaged to revise the text for a revamped new edition in 1982, after the author’s death. He was to remark ruefully that he would have profited greatly if he had accepted the offer of a percentage royalty rather than the lump sum for which he opted! Later, The Ray Society asked him to undertake the much more formidable task of preparing for the press the monograph by E. S. Edees and A. Newton, Brambles of the British Isles, a task which involved much painstaking checking of dates and references as well as two exceptionally arduous proof readings before the book’s appearance in 1988. As a fitting culmination, the List of Vascular Plants of the British Isles - immediately known to everyone as the “Kent List” - was produced for the Society and published by it in 1992 after seven years in preparation. By then it was over three decades since J. E. Dandy’s analogous volume had appeared and numerous nomenclatural and taxonomic changes had accumulated, many new hybrids had been detected and a marked increase in the number of established aliens had occurred. The new list consequently extended to many more taxa than its predecessor. In one sense it was the stepchild of BSBI Abstracts, for through his work for that Duggie had kept abreast of the trickle of new names entering the literature and made a regular point of seeking authoritative advice on whether or not there was a strong enough case for accepting those for B.S.B.I. purposes. Compilation of the List proceeded in parallel to the preparation of New Flora of the British Isles (1991) by C. A. Stace, with whom consultations on an agreed common nomenclature and classification occupied hundreds of pages of correspondence and many days of convivial meetings over those seven years and thereafter up to his death. The “Kent Lis?’ is merely the tip of a very OBITUARY D. H. KENT 7 Duggie Kent (1920-1998) with a copy of W. Keble Martin’s The concise British Flora in colour (Photo. c. 1965, courtesy of North Thames Gas). sizeable iceberg: characteristically, Duggie had compiled, as well as the places and dates of publication of all the taxa listed, many more synonyms than the strictly limited number included in the finished work. The product of an immense amount of meticulous library research, all this information is safely stored on the B.S.B.I. Database at the University of Leicester, where it is kept updated and is the object of many enquiries from researchers. In these later years he also received further honours. In 1977 he was the recipient of the H. H. Bloomer Medal, which The Linnean Society of London awards each year to an amateur naturalist of particular distinction. His amateur status was in fact so well disguised by the prolific nature and consistently high standards of his works that distinguished foreign botanists often found it hard to believe. In 1979, after reputedly two unsuccessful attempts to persuade him to accept nomination as President, the B.S.B.I. was able to overcome his diffidence to the extent of his agreeing to let his name go forward for election as one of the Vice-Presidents. And finally in 1993, to the great pleasure of his many friends, he was awarded the Order of the British Empire and personally invested with that by Her Majesty The Queen. This lengthy account of his botanical career must not be allowed to leave the impression that that was the limit of his interests. Indeed, if he himself was to be believed, botany was not even his principal one: it came a mere third, he liked to insist, to cats and Wagner, “in that order’. With the burly frame and ruddy countenance that gave him the appearance of Farmer Giles the prime place he accorded to cats could not but seem incongruous, but he was undoubtedly heartbroken in his last year by the demise of his “dear friend and companion Smudger’, following not long after the death of his beloved Ivy. Watching cricket (especially Middlesex) was another, lifelong passion. He took a connoisseur’s delight, moreover, in forming collections of certain garden plants. He was a dedicated member of the British Fuchsia and International Asclepiad Societies and of the Geraniaceae Group, contributing to their journals, and he compiled a very extensive card-index of all the Fuchsia cultivars he managed to trace. 8 OBITUARY D. H. KENT Some of the time and energy he was able to devote to this remarkable medley of interests was saved by not engaging in some activities that he might have been expected to pursue. He showed little or no interest in conservation, for example, and, disliking public speaking, rarely gave a lecture. His herbarium consisted of only about 2500 sheets, mostly from Middlesex in the 1940s and 1950s, and was scarcely added to in his last 35 years. Living close to his place of employment also meant he lost minimal leisure to commuting; being a non-driver, his travelling was by public transport or by being driven by his wife. A good idea of the way Duggie operated can be gained by a study of his botanical notes, which were carefully listed by him in “Manuscript 24” and indexed in “Manuscript 25”. There were four categories: notebooks, manuscripts, folders, and maps of Middlesex. His Notebooks, started in 1941 and ending in 1983, numbered 113 and contained a wide range of data. These included field notes (principally from Middlesex but also from his family holidays); notes on Middlesex taken from the literature, other botanists’ manuscripts and herbaria; notes that were to become the nucleus of some of his papers (e.g. Senecio squalidus notes started in 1950, British herbaria in 1953 and Veronica filiformis in 1954); data supplementary to various papers (e.g. to Handlist of Plants of the London Area); references to deceased botanists; and the raw data collected for BSBI Abstracts. Interestingly, up to 1945 he mainly used English names for the plants he listed. One notebook included the full plant list that he made on his 1971 holiday to Scotland, starting at West Ealing and finishing in West Ross and Wick. The 24 Manuscripts included many relating to his Historical Flora of Middlesex, including (no. 1) “Flora of Middlesex” which was intriguingly “destroyed 6/7/1950”; a catalogue of his botanical library, including the cost of the books he bought and their origin if second-hand; a list of British natural history journals; and a cumulative list of the species he found within one mile of 75 Adelaide Road. Somewhat surprisingly, there were no lists of his own herbarium or publications, but here is the abundant evidence of an instinctive and methodical collector and collator. One keen interest of Duggie that never came to fruition in his lifetime was his commitment to the system of vice-county recording. One will not find mention of grid squares in his Middlesex flora (or its Supplement), and we doubt that he ever spent much time ‘square-bashing’. He had intended producing an updated ‘Comital Flora’ well before his paper proposing one at the 1961 B. S.B.I. conference: notebooks 42—45 (1951) listed common species for which v.c. records were required. As an example of lengths to which he would go to achieve his goals, and as a comment on days before ordinary people had much spare cash and before the days of photocopiers, in the 1940s he copied out by hand the entire first and second Supplements to Watson’s Topographical Botany (1905 & 1929-30), amounting to 217 printed pages. Duggie later became one of the most active members of the Vice-county Census Catalogue Committee that was set up in 1988, and did the lion’s share of the initial data-gathering. His efforts should see the light of day very early in the new millennium. Duggie’s final years were sadly blighted by ill health and by the death in 1988 of his only child and in March 1997 by that of his wife. As it turned out, he outlived her by only eighteen months. The B.S.B.I., The Natural History Museum and Kew were all officially represented at the funeral. His herbarium has been acquired by The Natural History Museum, while his botanical books and bound manuscripts have been purchased by Summerfield Books of Brough, Cumbria. It is often said that no-one is irreplaceable, but also that exceptions prove the rule. Duggie was such an exception, one of the last of the great amateurs to devote most of their lives to the study of wild British plants and their literature, and there is sadly no-one in sight to rival his knowledge, expertise, long memory and dedication. A man of very definite views, he was well aware of his own limitations, and rather scornful of those who did not know theirs. Many have lost a loyal friend and colleague, the whole B.S.B.I. an unfailing servant; at least we can be thankful that so many of his contributions remain for all to admire and use. We shall not see his likes again. This account would not be complete without extending our deepest sympathy to his two granddaughters, Shelley and Nicola, to whom we are indebted for some of the key dates and family information. Others who have made valuable contributions are Mary Briggs, Dick Brummitt, Rodney Burton, Arthur Chater, Joan and Peter Hall, Keith Hyatt, Ron Payne and Francis Rose. D.E. ALLEN & C.A. STACE Watsonia 23: 9-14 (2000) 9 Botanical publications of Douglas Henry Kent The following list includes all the publications that we have been able to trace, with the exception of the many book reviews in Natural History Book Reviews from 1976 onwards. Also excluded are numerous administrative and editorial notes that have appeared in the BSBI’s publications. When presenting an equivalent list of J. E. Lousley’s publications, Kent (1977) said that they “span more than half a century”. In fact they spanned 51 years, exactly the same as did his own works at his death, although in the latter case there were a few more to follow. 1947 Notes on the flora of Middlesex. London naturalist 26: 56-65. 1948 Middlesex plant records for 1947. London naturalist 27: 38-42. Plant note - Panicum dichotomiflorum Michx. The Botanical Society and Exchange Club of the British Isles. Report for 1946-1947 13: 278. 1949 Middlesex plant records for 1948. London naturalist 28: 31-35. Additions and corrections to the Comital Flora for Middlesex, I. Watsonia 1: 92-101. John Blackstone, apothecary and botanist (1712-1753). Watsonia 1: 141-148. Excursions, 1947. September 27th. Rubbish tip, Hanwell, Middlesex. Botanical Society of the British Isles. Year Book 1949: 45-46. Plant note - Ipomoea lacunosa L. Watsonia 1: 120. Plant note - Verbascum phlomoides L. var. albiflorum (Rouy) Wilmott. Watsonia 1: 120. 1950 Tothill Fields, Westminster: a lost botanical area. London naturalist 29: 3-6. Abstracts from literature relating to botany. London naturalist 29: 20-22. Notes on the flora of Kensington Gardens and Hyde Park. Watsonia 1: 296-300. Herniaria glabra L. in Middlesex. London naturalist 29: 6-7. BOS The flora of bombed sites in Canterbury. Watsonia 2: 12-17. Additions and corrections to the Comital Flora for Middlesex, II. Watsonia 2: 78-79. Plant note - Lysichitum americanum Hult. & St. John. Watsonia 2: 109. Book review - Wild flowers of chalk and limestone (J. E. Lousley). London naturalist 30: 40. 1951-1957 (with J. E. Lousley) Handlist of the plants of the London Area. London naturalist 30-36, Supplements. 1952 Book review - The study of the distribution of British plants (ed. J. E. Lousley). London naturalist 31: 94. 1953 (editor) Botanical Society of the British Isles. Year Book 1953. (compiler) Abstracts from literature. Watsonia 2: 308-334. Book review - Drawings of British plants, part VI (S. Ross-Craig). Watsonia 2: 425-426. 1954 Plant note - Psoralea americana L. Proceedings of the Botanical Society of the British Isles 1: 37. Plant note - Cerinthe minor var. hispida Turrill. Proceedings of the Botanical Society of the British Isles 1: 39. Plant note - Cypripedium calceolus L. Proceedings of the Botanical Society of the British Isles 1: 39-40. Plant note - Cotoneaster horizontalis Decne. Proceedings of the Botanical Society of the British Isles 1: 156. Plant note - Bromus carinatus Hook. & Arn. Proceedings of the Botanical Society of the British Isles 1: 159- 160. Book review - The natural history of the City (R. S. R. Fitter & J. E. Lousley). Proceedings of the Botanical Society of the British Isles 1: 104. 1954-1965 (editor) Proceedings of the Botanical Society of the British Isles 1(1) — 6(2). (22 parts) 10 BIBLIOGRAPHY 1954-1969 (compiler) Abstracts from literature. Proceedings of the Botanical Society of the British Isles 1(1) — 7(4). (28 parts) 1955 The Scottish records of Senecio squalidus. Proceedings of the Botanical Society of the British Isles 1: 312- 6 9 te Plant note - Genistella sagittalis (L.) Gams. Proceedings of the Botanical Society of the British Isles 1: 320. Plant note - Elodea callitrichoides (Rich.) Caspary. Proceedings of the Botanical Society of the British Isles 1: 321-322. Plant note - Egeria densa Planch. Proceedings of the Botanical Society of the British Isles 1: 322. Plant note - Lagarosiphon major (Ridley) C. E. Moss. Proceedings of the Botanical Society of the British Isles 1: 322-323. 1956 Senecio squalidus L. in the British Isles, 1. Early records (to 1877). Proceedings of the Botanical Society of the British Isles 2: 115-118. Plant note - Juncus tenuis var. anthelatus Wieg. Proceedings of the Botanical Society of the British Isles 2: 23. Plant note - Pulsatilla vulgaris Mill. Proceedings of the Botanical Society of the British Isles 2: 134. Plant note - Polygonum campanulatum Hook. f. Proceedings of the Botanical Society of the British Isles 2: eye 1957 (with the assistance of E. B. Bangerter & J. E. Lousley) British Herbaria. Botanical Society of the British Isles, London. (with E. B. Bangerter) Veronica filiformis Sm. in the British Isles. Proceedings of the Botanical Society of the British Isles 2: 197-217. Senecio squalidus L. in the British Isles, 3. East Anglia. Transactions of the Norfolk and Norwich Naturalists’ Society 18(5): 30-31. An interesting 19th century hortus siccus. Proceedings of the Botanical Society of the British Isles 2: 365-366. Plant note - Pulsatilla vulgaris Mill. Proceedings of the Botanical Society of the British Isles 2: 236. Plant note - Ranunculus ficaria var. aurantiacus Turrill. Proceedings of the Botanical Society of the British Isles 2: 236. Plant note - Silene vulgaris subsp. macrocarpa Turrill. Proceedings of the Botanical Society of the British Isles 2: 236-237. Plant note - Hypericum canadense L. Proceedings of the Botanical Society of the British Isles 2: 238. Book review - Flora van Nederland (S. J. van Ooststroom). Proceedings of the Botanical Society of the British Isles 2: 424-425. 1958 Plant note - Juncus ensifolius Wikstr. Proceedings of the Botanical Society of the British Isles 3: 49. Plant note - Eleusine africana O’ Byrne. Proceedings of the Botanical Society of the British Isles 3: 50. Cyperus fuscus in Middlesex. London naturalist 37: 180-181. Book review - Handlist of the Shropshire flora (L. C. Lloyd & E. M. Rutter). Proceedings of the Botanical Society of the British Isles 3: 128. 1959 The poisonous thorn-apple. Country life (London) 126: 489. Plant note - Corrigiola litoralis L. Proceedings of the Botanical Society of the British Isles 3: 283-284. Plant note - Geranium retrorsum L’ Hérit. ex. DC. Proceedings of the Botanical Society of the British Isles 3: 284. Book review - Welsh flowering plants, 2nd ed. (H. A. Hyde & A. E. Wade). Proceedings of the Botanical Society of the British Isles 3: 250-251. Book review - British trees and shrubs (R. D. Meikle). Proceedings of the Botanical Society of the British Isles 3: 306. 1960 Senecio squalidus L. in the British Isles, 2. The spread from Oxford (1879-1939). Proceedings of the Botanical Society of the British Isles 3: 375-379. A contribution to the flora of central London. London naturalist 39: 41-62. Short note - Bird-seed aliens at Ealing, Middlesex. Proceedings of the Botanical Society of the British Isles 3: 442. . Plant note - Berberis manipurana Ahrendt. Proceedings of the Botanical Society of the British Isles 4: 41. BIBLIOGRAPHY 11 Plant note - Lathyrus palustris var pilosus (Cham.) Ledeb. Proceedings of the Botanical Society of the British Isles 4: 41. Plant note - Prunus mahaleb. L. Proceedings of the Botanical Society of the British Isles 4: 41-42. 1961 A contribution to the flora of central London: addenda. London naturalist 40: 28. The flora of Middlesex walls. London naturalist 40: 29-42. 1962 (with E. B. Bangerter) Further notes on Veronica filiformis. Proceedings of the Botanical Society of the British Isles 4: 384-397. Additions and corrections to the Flora of Central London. London naturalist 41: 16-22. Plant note - Acaena ovalifolia Ruiz & Pav. Proceedings of the Botanical Society of the British Isles 4: 415. 1963 Progress in the study of the British flora since World War 2. Webbia 18: 129-150. Progress in the study of the Irish flora since World War 2. Webbia 18: 303-317. A new Comital Flora? In Wanstall, P. J., ed. Local Floras, pp. 70-80. Botanical Society of the British Isles, London. Senecio squalidus L. in the British Isles, 7. Wales. Nature in Wales 8: 175-178. Plant note - Crataegus succulenta Schrad. Proceedings of the Botanical Society of the British Isies 5: 25-26. Plant note - Holcus lanatus x mollis. Proceedings of the Botanical Society of the British Isles 5: 27. Plant note - Pteris cretica L. Proceedings of the Botanical Society of the British Isles 5: 121. Field meetings, 1961. October 7th. An alien hunt in north London. Proceedings of the Botanical Society of the British Isles 5: 99. Book review - Welsh timber trees, native and introduced, 3rd ed. (H. A. Hyde). Proceedings of the Botanical Society of the British Isles 5: 104. Book review - Welsh ferns, 4th ed. (H. A. Hyde & A. E. Wade). Proceedings of the Botanical Society of the British Isles 5: 192. Book review - Atlas of the British flora (ed. F. H. Perring & S. M. Walters). Watsonia 5: 396-397. 1964 Senecio squalidus L. in the British Isles, 4. Southern England (1940->). Proceedings of the Botanical Society of the British Isles 5: 210-213. Senecio squalidus L. in the British Isles, 5. The Midlands (1940->). Proceedings of the Botanical Society of the British Isles 5: 214-216. Senecio squalidus L. in the British Isles, 6. Northern England (1940->). Proceedings of the Botanical Society of the British Isles 5: 217-219. Senecio squalidus L. in the British Isles, 9. Ireiand. [rish naturalists’ journal 14: 203-204. Notes on the flora of churchyard walls in Middlesex. London naturalist 43: 13-16. Plant note - Senecio. Proceedings of the Botanical Society of the British Isles 5: 232. Plant note - Elodea ernstiae St John. Proceedings of the Botanical Society of the British Isles 5: 232. Plant note - Scleranthus perennis subsp. prostratus P. D. Sell. Proceedings of the Botanical Society of the British Isles 5: 338. Plant note - Salvia glutinosa L. Proceedings of the Botanical Society of the British Isles 5: 343. Plant note - Telekia speciosa (Schreb.) Baumg. Proceedings of the Botanical Society of the British Isles 5: 343-344. Plant note - Galanthus elwesii Hook. Proceedings of the Botanical Society of the British Isles 5: 344. Plant note - Curtonus paniculatus (Klatt) N. E. Br. Proceedings of the Botanical Society of the British Isles 3: 344. Book review - A Flora of Cambridgeshire (ed. F. H. Perring et al.). Proceedings of the Botanical Society of the British Isles 5: 415-416. ; Book review - North Atlantic biota and their history (ed. A. Love & D. Love). Watsonia 6: 82-83. 1965 (nomenclatural editor) The concise British Flora in colour, by W. K. Martin. Ebury Press & Michael Joseph, London. (with E. B. Bangerter) Additional notes on Veronica filiformis. Proceedings of the Botanical Society of the British Isles 6: 113-118. Further additions to the Flora of Central London. London naturalist 44: 18-28. Plant note - Minuartia recurva (All.) Schinz & Thell. Proceedings of the Botanical Society of the British Isles 6: 48-49. Plant note - Brunnera macrophylla (Adam) I. M. Johnston. Proceedings of the Botanical Society of the British Isles 6: 50. 12 BIBLIOGRAPHY Plant note - Senecio squalidus L. [in Northern Ireland]. Proceedings of the Botanical Society of the British Isles 6: 51-52. Plant note - Dactylis glomerata subsp. hispanica (Roth) Nym. Proceedings of the Botanical Society of the British Isles 6: 52. Plant note - Asplenium trichomanes L. subsp. quadrivalens D. E. Meyer emend. Lovis. Proceedings of the Botanical Society of the British Isles 6: 121-122. Book review - Joseph Dalton Hooker (W. B. Turrill). Proceedings of the Botanical Society of the British Isles 6: 197. Book review - Introduction to botany (H. J. M. Bowen). Proceedings of the Botanical Society of the British Isles 6: 199-200. 1966 Senecio squalidus L. in the British Isles, 8. The recent spread in Scotland. Glasgow naturalist 18: 407-408. Book review - Glossary of the British flora (H. Gilbert-Carter). Proceedings of the Botanical Society of the British Isles 6: 299-300. Book review - William Turner (facsimiles by The Ray Society). Proceedings of the Botanical Society of the British Isles 6: 300. 1967 Index to Botanical Monographs. Academic Press, London & New York. Additions and corrections to the Flora of Central London (3). London naturalist 46: 26-35. Notes on some Middlesex grasses. London naturalist 46: 36-39. Buddleja, the Butterfly Bush. Wildlife and the countryside 266: 9-10. Book review - Huntia, vol. 2. Proceedings of the Botanical Society of the British Isles 6: 416. Book review - Companion to flowers (D. McClintock). Proceedings of the Botanical Society of the British Isles 7: 66. 1969 (nomenclatural editor) The concise British Flora in colour, 2nd ed., by W. K. Martin. Ebury Press & Michael Joseph, London. Fritillaria meleagris L. in north-west Middlesex. Journal of the Ruislip and District Natural History Society 18: 15-22. Book review - The Penguin dictionary of British natural history (R. Fitter). Proceedings of the Botanical Society of the British Isles 7: 619. Book review - Excursion Flora of the British Isles, 2nd ed. (A. R. Clapham, T. G. Tutin & E. F. Warburg). Watsonia 7: 107. . Book review - B-P-H: Botanico-Periodicum-Huntianum (ed. G. H. M. Lawrence et al.). Proceedings of the Botanical Society of the British Isles 7: 182. 1970 Short note - Spartina glabra Muhl. - in the British isles. Watsonia 8: 163-164. 71 (with M. Kovanda & R. K. Brummitt) Index to European taxonomic literature for 1969. Regnum Vegetabile 80. 1971-1998 (compiler) BSBI Abstracts, 1-28. Botanical Society of the British Isles, London. 1972 Notes on the Scottish flora: observations in West Ross. Transactions of the Botanical Society of Edinburgh 41: 547. Book review - The flowering plants and ferns of the Isle of Man: an annotated list (D. E. Allen). Watsonia 9: 169. 1974 Progress in the study of the British flora, 1961-1971. Memorias da Sociedade Broteriana 24: 353-375. (nomenclatural editor) The concise British Flora in colour, 3rd ed., by W. K. Martin. Ebury Press & Michael Joseph, London. 1975 The Historical Flora of Middlesex. An account of the wild plants found in the Watsonian vice-county 21 from 1548 to the present time. The Ray Society, London. BIBLIOGRAPHY 13 1977 (with R. K. Brummitt) Index to European taxonomic literature for 1970. Bentham-Moxon Trustees , Kew. J. Edward Lousley’s published works. Watsonia 11: 315-325. Rumex X lousleyi hybr. nov. (R. cristatus DC. x R. obtusifolius L.). Watsonia 11: 313-314. 1981 (with J. E. Lousley) Docks and knotweeds of the British Isles. Botanical Society of the British Isles, London. Notes on Hoya in cultivation, 1. Asklepios 23: 24-28. Floras great and small. Natural history book reviews 5: 3-15. British and Irish herbaria [appeal for information]. BSB/ news 27: 13. Viola riviniana as a garden weed. BSBI news 28: 19. 1982 The native and introduced succulents of the British Isles [1 & 2]. Xerophyte 5: 65-72 & 97-99. (editor) The New Concise British Flora, by W. K. Martin. Ebury Press & Michael Joseph, London. (with M. E. Kennedy) Obituary - Thomas Gerald Collett (1911-1982). London naturalist 62: 87. British and Irish herbaria [appeal for information]. BSB/ news 30: 23. 1983 The native and introduced succulents of the British Isles [3]. Xerophyte 6: 20-24. Book review - A seventeenth century Flora of Cumbria (E. J. Whittaker). Watsonia 14: 290. 1984 (with D. E. Allen) British and Irish Herbaria. Botanical Society of the British Isles, London. Two more decorative non-succulent genera. Asklepios 31: 89. 1985 Book review - A concise Flora of Dorset (R. Good, ed. H. J. M. Bowen). Watsonia 15: 413. 1987 (with R. K. Brummitt, P. S. Lusby & R. C. Palmer) The history and nomenclature of Thomas Edmondston’s endemic Shetland Cerastium. Watsonia 16: 291-297. Obituary - John Patrick Micklethwait Brenan (1917-1985). Watsonia 16: 211-214. Obituary - Barbara Welch (c.1904—1986). Watsonia 16: 461-463. Profile - Clive Anthony Stace. BSBI news 47: 6-7. Book review - Index of collectors in the Welsh National Herbarium (S. G. Harrison). Watsonia 16: 345-346. 1988 Rubus procerus ‘Himalayan Giant’. Kew Magazine 5: 32-36. (editor) Brambles of the British Isles, by E. S. Edees & A. Newton. The Ray Society, London. Book review - Flora of the British Isles, 3rd ed. (A. R. Clapham, T. G. Tutin & D. M. Moore). Watsonia 17: 107-108. Book review - Biosystematics in the Nordic flora (ed. B. & L. Jonsell). Watsonia 17: 204. 1989 Book review - The natural history of the Chew Valley (K. Allen et al.). Watsonia 17: 372. Book review - The fauna and flora of Exmoor National Park. A provisional checklist (Exmoor Natural History Society). Watsonia 17: 372. Book review - Natural science collections in Scotland (botany, geology, zoology) (ed. H. E. Stace ef al.). Watsonia 17: 373. 1990 Short note - The Shasta Daisy. Watsonia 18: 89. 199] In defence of The Hoyan. Asklepios 53: 3-6. (with E. M. Nic Lughadha) ““Momenclature” - a new taxonomic concept. BSBI news 57: 26. The year of Yellow Corydalis [Pseudofumaria lutea (L.) Borkh.]. BSBI news 59: 15. New Zealand Bittercress [Cardamine corymbosa Hook. f.]. BSBI news 59: 49. Book review - Flora of New Zealand, vol. 4 (C.J. Webb et al.). Watsonia 18: 328-329. 1992 List of vascular plants of the British Isles. Botanical Society of the British Isles, London. The type of Hoya imperialis Lind]. Asklepios 57: 9. Pseudofumaria lutea. BSBI news 60: 20-21. 14 BIBLIOGRAPHY 1993 List of vascular plants of the British Isles. Errata. Botanical Society of the British Isles, London. 1994 List of vascular piants of the British Isles. Errata, List 2. Botanical Society of the British Isles, London. Monsonia brevirostrata in England. Geraniaceae Group news 14: 9-10. (with D. J. Goyder) Micholitzia obcordata (Asclepiadaceae - Marsdenieae) reinstated. Asklepios 62: 13-19. 1995 List of vascular plants of the British Isles. Errata, List 3. Botanical Society of the British Isles, London. Book review - Dictionary of British and Irish botanists and horticulturists (R. Desmond with C. Elwood). Watsonia 20: 307. 1996 List of vascular plants of the British Isles. Errata, List 4. Botanical Society of the British Isles, London. (with P. Macpherson, J. H. Dickson, R. G. Ellis & C. A. Stace) Plant status nomenclature. BSBI news 72: 13- 16. 1997 List of vascular plants of the British Isles. Supplement I. Botanical Society of the British Isles, London. (“1996”) Note - The correct authority for Lesser Chickweed, Stellaria pallida (Caryophyllaceae). Watsonia 21: 364. 1998 List of vascular plants of the British Isles. Errata, List 5. Botanical Society of the British Isles, London. 1999 (with R. Wisskirchen) 1436-1437: Proposals to conserve two names in Persicaria (Polygonaceae). Taxon 48: 829-830. 2000 Supplement to the Historical Flora of Middlesex. Ray Society, London, in press. D. E. ALLEN & C. A. STACE Te Se Watsonia 23: 15-21 (2000) Ls A selective supplement to British and Irish Herbaria D. E. ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ ABSTRACT Since the publication in 1984 of British and Irish herbaria by D. H. Kent and D. E. Allen, a much-enlarged version of Kent’s British herbaria of 1957, a considerable number of additions and amendments have come to notice. Those rated the more important or of more especial historical interest are listed. INTRODUCTION After British and Irish herbaria was published (Kent & Allen 1984) both authors independently began entering in their respective interleaved copies the additions and amendments that came to their notice. D. H. Kent went so far as to produce a manuscript list of those he himself had accumulated down to July 1984, adding a further page at a later date; but the disappointing sale achieved by the volume led him to assume that the publishing of a further edition was not a realistic proposition for the foreseeable future. As he preferred anything he compiled to be as complete as he could make it, he resisted the suggestion of publishing a limited, selective supplement only, content for his annotated copy to pass into other hands in due course and trusting that its updating would be continued thereafter. It nevertheless remains the case that with the passing of the years the information in a reference work such as British and Irish herbaria cannot help but become less useful and reliable, as collections are relocated, new ones discovered and known ones subjected to further research. Since the 1960s, largely at the instigation of the Biology Curators Group, many institutions have been publishing lists of their holdings of a valuably comprehensive and incisive kind, and hopefully that fashion has some way to run yet; indeed, only shortly after British and Irish herbaria was published three further, major examples of that genre (Harrison 1985; Stace et al. 1987; Webb 1988) appeared. To rectify this gradual obsolescence even if only in part would seem to be worthwhile. The following list of just those changes judged to be the more important or the more historically interesting has therefore been put together to that end from the two interleaved copies. It is dedicated to the memory of D.H.K., who devoted so many years to the long-overdue task of establishing the precise extent of the herbarium resources underpinning the study of the taxonomy and distribution of the vascular plants of the British Isles, a task with which his name will always be pre-eminently associated. CRITERIA FOR INCLUSION Priority for inclusion in this supplement has been accorded to: 1 larger collections of a non-specialised character (on the assumption that the whereabouts of collections of critical groups will mostly be known to specialists in those and will be mostly in the half-dozen or so principal herbaria); 2 collections which have hitherto been wholly or partly anonymous (for those are deprived of much of their value, at least for historical purposes, as long as the person(s) responsible for forming them remain(s) unknown); 3 older collections made by those known to have been particularly active as field botanists, if not necessarily of national prominence; 4 collections of interest by virtue of their association with figures of historical, but not necessarily botanical, significance - such as W. H. Fox-Talbot, the pioneer of photography. 16 DoE AEE At the same time the opportunity has been taken to correct errors of a more than merely trivial character. Regretfully, though, with very few exceptions, the many instances where names or life dates have been ascertained with greater precision have had to be excluded for space reasons. ABBREVIATIONS USED For the initial volume, Kent (1957) devised a set of symbols and abbreviations which were repeated for British and Irish herbaria but which are undeniably daunting to a novice reader. These were adopted partly to minimise space, but partly also to distinguish between a person’s actual herbarium and specimens of his or her collecting (in some cases considerably more numerous) existing outside that context. The distinction is by no means a simple one to make. A herbarium may have been formed by, or at least belonged to, someone else before the person with whose name it has come to be identified: many herbaria were purchased or inherited and then added to by the new owner, often by extensive exchanges with other collectors. Sometimes only a fragment survives and whether that once constituted a larger whole regarded by that person as ‘my herbarium’ can be a matter of fine judgement. In a few cases people formed more than one herbarium, at different periods of their lives or on moving to a different area or after a first had been destroyed. The distinction between ‘herbarium’ and ‘other specimens’ may be of little or no moment for scientific purposes but it matters historically and biographically - and the value of so many herbaria today is essentially as historical documents. For the sake of consistency, Kent’s use of * to denote a herbarium identified with a named individual has been retained in the List of Collectors below. Other symbols and abbreviations used in the successive volumes, such as *‘D’ (donated) and ‘Pd’ (purchased), have been avoided. A new one (+) has, however, been introduced in order to indicate names of collectors additional to those in the analogous list in British and Irish herbaria. INTER-INSTITUTION TRANSFERS As schools, smaller museums and now even some universities have no further use for herbaria in their possession or cannot adequately curate them, collections are increasingly becoming concentrated in a limited number of larger institutions, more particularly those where research or teaching in taxonomy is still active. The development of county museums services with centralised facilities for the conservation of biological material, has been furthering that trend, which is surely one to be welcomed. It is, however, the greatest cause of change, and is best noticed first and separately: transferred transferred from to from to AMBL LANC (1988) GGO GLAM (1991) AST part to BM, part to WARMS (1988) KCK NWH BATHG BTH KRG LSG BRIST BRISTM except for ferns (to NMW) LIMFC DBN (1983) CHEL LIV (1994) except for 18th century NCCE E (1993) material (to BM) CHRG LIV (1985) NSS LIV (1986) CRK DBN (1989); over half had to be OHM OLDM (1995) destroyed as irrecoverably insect-damaged CYN SLBI TMAL SGEL ESNHS — GLAM (1982) WPP LSR EXR LIV (1993) except for the Keble Martin herbarium (to RAMM) SUPPLEMENT TO BRITISH AND IRISH HERBARIA 17 Among relocations which have taken place within a single museums service that of THO from Thurso to Wick (Caithness District Council, Library Annexe, Sinclair Terrace) and of BDK from Baldock to HITCHIN (Museum Resource Centre - new code letters HITN - Bury Mead Road) are worthy of mention. The herbaria previously held in various towns (but not cities) in Hampshire have also been brought together at Winchester (Hampshire County Museums Service - new code letters HCMS - Chilcomb House, Chilcomb Lane). LIST OF COLLECTORS: ADDITIONS AND CORRECTIONS 7+ ADDISON, William, fl. 1836-52, surgeon of Malvern. *Seen by E. Lees (Naturalist, 1837, 1: 210): untraced. + ANDERSONIAN NATURALISTS *GLAM (1880s—1903). + ARBROATH HORTICULTURAL AND NATURAL HISTORY ASSOCIATION *DUE 520 (v.c. 90; 1877-90). + BACKHOUSE, Edward, 1808-1879. SUN under 650 (N. England) - not *. BALFOUR, John Hutton, 1808-1884. The 8500 at PTH, though long attributed to him, is the British Isles portion of a world-wide Edinburgh University teaching collection (1764~1863), with numerous contributors, donated in 1880 to the Literary and Antiquarian Society of Perth. The MSE specimens date from 1855—90 and were donated by J. C. Howden pre-1897. TCD also has 100—200. BARCLAY, William, 1846-1923. The collection at PTH extends to many hundreds of specimens and is presumed to be his herbarium. + BATEMAN, Rev. John, 1665?—1774. *Untraced, but at least part was in *Edward JACOB. This v.c. 15 collector is not to be confused with the London physician John BATEMAN, fl. 1663-— 1728, whose * is in PHA. *BEILBY, M. A., later Mrs AVERY, fl. 1830s, of Moseley, Birmingham. This is the identity of the “M.A.B.” listed on p. 193. The 400 specimens in BIRM are presumed to be the * mentioned by E. W. Benson in Analyst, 1837, 6: 294. BONIFACE, Ronald Archie, 1913-1985. *NMW. BORRER, William, 1781—1862. Many specimens are or were in *H. Collins (now in *F. H. Arnold in HSM). BRIGGS, Thomas Richard Archer, 1836-1894. Hardly any Rubus in K, contrary to statement: c. 100 each of Rubus in BM and CGE instead. BROMFIELD, Willim Arnold, 1801—1851. TCD 375 (v.cc. 10-12). BROOME, Christopher Edmund, 1812-1886. * not in E but in BTH (especially 1830s). BROWNING, F. R., fl. 1939-1970, of Pembury, Kent. * discovered in 1994 in a stable-cum- pigsty in Swaledale, Yorks.; mostly S. E. England and 1931-1955 (P. J. Cook, B.S.B.I. News, 1994, no. 65: 30). BUCKNELL, E. This is Edward BUCKELL (no ‘N’), a physician of Romsey, Hants. BULL, Henry Graves, 1818?—1885. *HFD (incorporating * of his father, Henry BULL, 1816—post 1859). BURNETT, Percy, 1903-1972. * (12 drawers) in possession of Dr R. L. Gulliver, Carraig Mhor, Imeravale, Port Ellen, Isle of Islay, PA42 7AL. ‘C’ and ‘E.F.’ This should be ‘E.F.C.” The collection in GL relates to v.cc. 85, 87-89. CHILDS, Kathleen Amelia, 1880-1952. Though the sheets themselves are unlocalized (and undated), the localities are given in an accompanying catalogue. All the specimens are from the Alton area, v.c. 12, and most date from the 1920s. CLARK, William Andrew, 1911-1983. *BM (v.c. 110). + CLEAR, H. William, fl. 1880-1926, Leicester pharmacist. * untraced (“one of the largest. . . in the Midlands” (Chemist & Druggist, 17 June 1926: 138). COOKE, Randle Blair, 1880-1973. * in *William Andrew CLARK, q.v. COTTON, Thomas Atkinson, 1857-1925 and Mrs Charlotte COTTON, née SPENCE, d. 1925. *BMH. COULTER, Thomas, 1793-1843. TCD 185 (160 Ireland, mainly Dublin area, 1820-1822; 25 Britain, 1842). 18 D. E. ALLEN DICK, Robert, 1811—1806 (sic). For *‘500’ read ‘50’ (an album of ferns). DICKINSON, D., fl. 1875-1902, pharmacist of Hoole, Chester. * (1 vol. out of original 3: 323, mostly Ripon area, v.c. 64; 1878-1902) in possession of J. M. H. Shaw, 4 Albert St., Stapleford, Nottingham, NG9 8DB (B.S.B.I. News, 1995, no. 69: 34-35). DIXON, Mrs (*formerly at CCR) and Mrs H. DIXON, fl. 1832-1850, are the same person. She was Anne DIXON, née AUSTEN (a cousin of Jane Austen, the novelist), 1799-1864, wife of the Rev. Henry Dixon, vicar of Ferring, Sussex. She was an intimate from childhood of Mrs Anna ATKINS, née CHILDREN, 1797-1871, with whom she is known to have collected and whose * (in BM) may thus include specimens of hers. DORRINGTON, John, 1814?-post 1861. Delete entry. DOWNES, Rev. John, Downes, 1810-1880. *NTN 800 (10 vols., 220 v.c. 29; 1829-1832). Listed in Kent (1957:33) as that of an unknown collector and in Kent & Allen (1984) as that of John DORRINGTON, this * has been identified as almost certainly that of the ‘J. DOWNES’ of the two entries on p. 131 (Crompton 1997). + DRUMMOND, James Lawson, 1783-1853. GL 250 (mainly Belfast area). + DUNLOP, Miss, of Biggar, Lanarkshire. E 300 (mostly v.c. 77; 1890s). EDEES, Eric Smoothy, 1902-1993. *(part: Rubus) NMW 7500. *RILSTONE (Rubus). EHRET, Georg Dionysius, 1708—1770. *Formerly CHEL now LIV 26 (unlocalized; c. 1730). ELDRIDGE, Ellen, fl. 1859-1868. *DZS 300 (Salisbury area). + EMSLEY, F. *TCD 215 (mostly Yorks.). + FORBES, Nanna A. B. *INV 300—400 (1 vol., v.cc. 105 & 106; pre-1914). FORSTER, Edward, 1765-1849. Another “very large” * said to be his was auctioned by Stevens in Dec. 1861 (Phytologist, 1861, 6 (N.S.): 16). FOX, Brian William (d. 1999). *BON 700+ (Bolton area; adventive spp. 100+; 1948-78). + FRANCIS, W. QMC (Kent, Essex, Aberdeenshire; 1830s). GIBSON, Jabez Marriage. Correct dates are 1822-1877. GIBSON, Samuel, 1790°?-1849. Part of * (700 Pteridophyta) still in SALFM. GORDON, Rev. George, 1801-1893. *ELN 101 (pre-1894) believed to be his at least in part. + GORDON, George Hamilton, Earl of Aberdeen, 1784-1860. *ABD (5 vols.; v.cc. 92 & 93; 1817). + GORRIE, David. *1394, donated to Alloa Society of Natural Science and Archaeology in 1875; Society’s museum was cleared out in the 1940s and * presumed lost. +GOUGH, Rev. William Edgar, 1864—c. 1930. * formerly at Adcote School, near Shrewsbury, now CGE (mostly collected by Rev. Thomas STEPHENSON, 1855-1948, includes many B.E. C. sheets). + GRAHAM, George John, 1803-1878, of Enfield, Middx. *formed with his friend John Stuart Mill, later destroyed (J. Britten, Journal of Botany, 1905, 43: 318). + GRENFELL, Adrian Leonard, 1939-1991. *BRISTM 2000 (Bristol area; aliens). GRIMSTON, Lady Mary Augusta Frederica, Countess of Radnor, d. 1879. The * attributed to her in BM is most unlikely to be hers, as most specimens date from 1841 - the year in which her first child was born (J. G. Dony, in litt., 1985). HALL, Patrick Martin, 1894-1941. Hampshire component of * only on loan to BM pending transfer to HCMS. HEMSTED, Rev. John. Dates should be 1747?-1824. Delete ‘1711-1801’, which apparently refers in part to * Joseph ANDREWS, which was once in his possession and was subsequently donated by Hemsted descendants to BM. BM holds only a set of unlocalized Mentha specimens collected by him (attached to the 10 vois. of *“ANDREWS). Specimens of his from v.cc. 26 and 29 are also in * James SOWERBY in BM (Oswald 1991). + HENDERSON, Alistair A. R., 1937-1988. *GLAM 1300+ (Scotland; 1948—), HILL, Sir John, 1714-1775. * acquired by the de Jussieu family, donated in recent years to P. HOPE, John, 1725-1786. Despite the disappearance of his * its MS catalogue (1768—1780s) survives in E. HOWITT, Richard Crewdson Leaver, 1911-1984 and Mrs Brenda HOWITT, née CHALK, d. 1981. *CGE (v.c. 56; Salix, Potamogeton). HUTCHINSON, Rev. Thomas, 1815-1903, nephew of Wordsworth. * awarded ‘Best Herbarium’ prize of Herefordshire Horticultural Society, 1854; not traced. SUPPLEMENT TO BRITISH AND IRISH HERBARIA 19 I ANSON. This should be JANSON. JAMES, Robert, LORD PETRE, 1713-1742. * (16 vols., incl. 2 of British plants) sold 1886, later acquired by Sutro Library, San Francisco, U.S.A. Most volumes annotated by Sir Jonn Hill (McLean 1984). KENNETH, Archibald Graham, 1915-1989. *E (v.c. 101). KIRBY, Rev. H., fl. 1830-1870, of Mayfield, Sussex. * (large folios; mostly c. 1840) donated in 1904 to Eastbourne Technical Institute (Journal of Botany, 1905, 43: 72). Presumably the same person as the Rev. F. KIRBY, whose * at EBE was destroyed by enemy action in 1940. + LATHAM, Brian. ABDAM 800 (v.cc. 82—84, 92, 93; 1954-1957). LEE, Sarah. Error for LEE, Susan, fl. 1850-1871. The specimens are some mere scraps from the wall of a church, contained in a packet attached to the 1841 collection misattributed to Lady Mary GRIMSTON, q.v. (J. G. Dony, in litt., 1985). Delete entry. LEIGHTON, Rev. William Allport, 1805-1889. SHY 270 (v.c. 40; pre-1835). + LIBBEY, Richard Pearse, 1911-1987. LTR 8,000 (v.cc. 27 & 28). MacGILLIVRAY, William, 1796-1852. The collection in ABD extends to a few thousand specimens and thus appears to rank as *. MADDEN, Edward, 1805-1856. Brought up in Ireland, in India 1830-1849 and a known collector with an untraced *, possibly the former of an anonymous two-volume collection of the 1820s acquired by SKT from a Stockport bookshop and subsequently transferred to MANCH. The specimens are mainly from v.c. 55B (but the handwriting was unknown to the late K. G. Messenger) and H11. MAHER, D., fl. 1847. Believed to be a misreading of ‘Dr. MATEER’, i.e. the Dr William Mateer, fl. 1831-1850, listed on p. 201, a contributor to the exchanges of the Botanical Society of London in 1847-1850. MATHEWS, William, 1828-1901. K has main part of * exclusive of v.cc. 36 & 37 specimens (which are the 5,000 in WOS). MOSELEY, Harriet, fl. 1836-1867. * mostly 1830s, but one specimen 1799. The 9 vols. include material from many other collectors. MUIRHEAD, Clara Winsome, 1915-1985. * (main) PLYP. MURPHY, Rose J. *BM (v.cc. 1 & 2). + PEIRSON, Archdeacon, d. 1805/6, of York. * (1,745 British specimens) raffled in York (Sir Thomas Frankland in litt. to J. E. Smith, 4 Feb. 1806, Smith correspondence, Linnean Society); not traced. + PENNANT, Thomas, 1726-1798. * (mostly unlocalized) donated to CGE by Earl of Denbigh in 1912 (Sherborn 1940) but cannot now be traced. mEOLEITT, A., fl. 1920. LIV 500 (v.c. 58). + PRATT, Anne, afterwards Mrs Pearless, 1806-1893. * (“extensive”’; 1820s—1830s; v.cc. 15 & 16 (Woodward 1896)); not traced. PURCHAS, Rev. William Henry, 1823-1903. Many sheets (of many genera) in BMH, none of them from exchange clubs: possibly a separate *. RAYNER, John Frederick, 1854-1947. Only a few sheets survive of the once-extensive * in BMH. The rest are believed to have been discarded in the 1950s. REID, William, d. 1852. This Montrose doctor apparently collected in v.c. 90 exclusively. The 80 sheets in KRDY (v.cc. 83 & 85) were collected in 1851-1858 by a ‘Mr. Reid’ of Edinburgh, clearly a different person. REYNELL, Elizabeth T., fl. 1864, of Westmeath (not ‘Meath’). *TCD 166. ROBERTS, John, of Bangor. Correct dates: 1792—1849. Apparently had 2 *, one formed from _ c. 1820, a second seen by J. E. Griffith (Jones 1998: 62). Both untraced. ROBINSON, Gilbert Wooding, d. 1942. *formerly CHEL, now LIV; incorporates one or more older collections, many specimens being from Botanical Society of London members and dating back to the 1830s (see Edmondson 1994). ROBSON, Stephen, 1741-1779. * not in SUN, as stated, but still in possession of a descendant, E. Robson, Market Weighton, Yorks. (Davis 1981: 72; Horsman & Davis 1990: 77). ROGERS, Rev. William Moyle, 1835-1920. * (part, except Rubus) - presumably that formerly in THE - now in LANC. 20 D.E. ALLEN SALMON, John Drew, 1802-1859. The * part in NWH was bequeathed, at which time it totalled c. 2,900 (mostly Thetford area; 1830s) according to the museum’s records. Either ‘1400’ is a miscount or much has been lost. The * part auctioned by Stevens in 1861 comprised his main British collection minus “the Surrey folios” and totalled c. 1000 sheets (Phytologist, 1861, 6 (N.S.): 350). It must be some of these latter which NMW (via *SHOOLBRED) and WARMS (via WAR) possess. SANDYS, Rev. George William, 18127-1848. * (Stroud area) seen by Riddelsdell (Riddelsdell et al. 1948: cxxi) at a private house somewhere in Gloucestershire. “H.C. WATSON presumably contains merely duplicates from Botanical Society of London exchanges. + SELBY, Prideaux John, 1788-1867. A ‘dry plant book’, item 707 in the MS catalogue of his library, may have been his * (Jackson 1992: 32); if so, not traced. SHAW, Rev. Charles Edward, 1910-1994. *’dispersed among friends on moving house” (Hartley 1994: 28); but 63 in OLDM (“Alien flora of Oldham’, 1940-1960). + SIMONYYI, Erno, fl. 1853-1868, of Hungary. BP 300 (1853-1859; England and Scotland). SMITH, Sir James Edward, 1759-1828. Also a * formerly LIVB, now LIV (c. 10,000 specimens - fuller details in Edmondson 1985). SMITH, William, d. pre-1836. * (Aberdeenshire?) mentioned by Murray (1836: 10, 129), his cousin. SPALDING GENTLEMEN’S SOCIETY. The * in SGGS listed on p. 67 is a 19th century one. Gough & Nichols (1812: 9, f 2) record that the Society possessed a “fine hortus siccus”’ c. 1745. STOVIN, (not ‘Mrs’) Margaret, 1756-1846. Catalogue of * in MDH published by Simmons (1993). SWANN, Eric Lister, 1904-1989. *NWH (v.c. 28; Carex, Gramineae). TALBOT, William Henry Fox, 1800-1877. * (“‘large”; partly formed by his mother, Lady Elisabeth Theresa FOX-STRANGWAYS, afterwards TALBOT, afterwards FEILING, 1773- 1846; includes sheets from a purchased collection) in Talbot Museum, Lacock, Wilts. THOMAS, Anna. More precise dates: fl. 1849-1862. Of Bristol (*H. C. WATSON). Must thus be the Mrs Thomas, of Bristol, listed on p. 257. + THOMSON, fl. 1797. “Mr Thomson continues to collect every plant near Penzance” (Davies Giddy in litt. to William Withering, 19 Nov. 1797 (Mann 1986: 121); * untraced. THRELKELD, Caleb, 1676-1728. *TCD 22 (only 2 localized). An anonymous collection attributable to him on internal evidence (Doogue & Parnell 1992). TRAILL, James William Helenus, 1851-1919. Contrary to the total of ‘304’ stated, Stace et al. (1987) say “many thousands” of his British flowering plants are in ABD. TRIST, Philip John Owen, 1908—1996. *CGE 3,275 (mainly British Gramineae). VIGURS, Chambré Corker, 1867—1940. “THURSTON (in K) incorporates a second * of his. WALLACE, Edward Charles, 1909-1986. *RNG 22,855 inclusive of foreign (v.cc.. 13-17, 37, 72, 80-88, 99, 100). WEBB, David Allardice, 1912—1994. TCD 2,165 (Ireland, 1950-5). WESTRUP, Alick William, 1910—1994. * formerly HIWNT, now HCMS. WHITE, D. Presumed to be the Montrose surgeon of that name whose records are cited by the Aberdeen physician Alexander Murray (1836). Specimens of White’s are in *E. H. ROBERTSON, a fellow Angus collector; Robertson was not born until 1867, so White’s may be among the specimens in that dating back to 1789 (see p. 231). WHITEHEAD, John, 1833-1896. OLDM 100 (v.c. 59). WILLIAMSON, William Crawford, 1816-1895. * formed in youth (E. Yorks.) (Williamson 1896: 33); not traced. + WILSON, Thomas Watkins, fl. 1820s — 1830s, physician of Shrewsbury. SHY 331 (200 v.c. 40, 131 Scotland). WITHERS, Robert, 1814?-1855. Donated to TTN in 1852 “small collection of the rarer plants of the Bath area” (Proc. Somerset Arch. & Nat. Hist. Soc. 1852, 3: 12). Survives in TTN unidentified? WRIGHT, Charles H., fl. 1832-1844, and his daughter. TCD 245 (v.c. 70; mostly 1839-1843). + YOUDALE, W.‘H., fl. 1890-1908. *LIV 314 (v.c. 70; 1891-1908) (Hofmann et al. 1990). SUPPLEMENT TO BRITISH AND IRISH HERBARIA 21 ACKNOWLEDGMENTS Considerably more authors and correspondents have been responsible for providing information drawn on for this supplement than it has been feasible to acknowledge individually in the text. I would like to take this opportunity of thanking them for their contributions and of assuring them that everything else they have published or reported about particular collections in addition to what appears above has been carefully noted and entered up for future reference. I would also like to thank Dr J. R. Edmondson for reading through the paper in manuscript and commenting helpfully. REFERENCES CROMPTON, G. (1997). The *J.D.” herbarium at Northampton Museum. Watsonia 21: 362-364. DAVIS, P. (1981). Stephen Robson’s hortus siccus. Naturalist 106: 67—73. DOooGugE, D. A. & PARNELL, J. (1992). Fragments of an eighteenth century herbarium, possibly that of Caleb Threlkeld, in Trinity College, Dublin (TCD). Glasra 1: 99-109. EDMONDSON, J. R. (1985). The other Smith herbarium. Linnean 1(4): 17-19. EDMONDSON, J. R. (1994). The G. W. Robinson herbarium: newly acquired by Liverpool Museum. B.S.B./. news: 65: 29. GOUGH, R. & NICHOLS, J. (1812). Some account of the Gentlemen’s Society at Spalding, in NICHOLS, J., ed. Literary anecdotes of the eighteenth century 6: 1-162. Privately published, London. HARRISON, S. G. (1985). Index of collectors in the Welsh National Herbarium. National Museum of Wales, Cardiff. HARTLEY, M. M. (1994). West Yorkshire herbaria, in LAVIN, J. C. & WILMORE, G. T. D., eds. The West Yorkshire plant atlas, pp. 26—28. City of Bradford Metropolitan Council, Bradford. HOFMANN, H., EDMONDSON, J. R. & HALLIDAY, G. (1990). The Cumbrian herbarium of W. H. Youdale. Watsonia 18: 204-206. HORSMAN, F. & DAVIS, P. (1990). Edward Robson’s list of the rarer plants of County Durham. Archives of natural history 17: 67-77. JACKSON, C. E. (1992). Prideaux John Selby, a gentleman naturalist. Spredden Press, Stocksfield. JONES, D. (1998). The doctor and the guide: two Snowdonian plant hunters. Transactions of the Caernarvonshire Historical Society, 1998: 55-75. KENT, D. H. (1957). British herbaria. Botanical Society of the British Isles, London. KENT, D. H. & ALLEN, D. E. (1984). British and Irish herbaria. Botanical Society of the British Isles, London. MCLEAN, E. P. (1984). A preliminary report on the 18th century herbarium of Robert James, Eighth Baron Petre. Bartonia 50: 36-39. MANN, R. D. (1986). William Withering and the foxglove. MTP Press, Lancaster. MuRRAY, A. (1836). The northern flora; or a description of the wild plants belonging to the north and east of Scotland. Part 1. Edinburgh. OSWALD, P. H. (1991). The Revd John Hemsted (1747?-1824). Nature in Cambridgeshire 33: 26-29. RIDDELSDELL, H. J., HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. Buncle, Arbroath. SHERBORN, C. D. (1940). Where is the ? collection. Cambridge University Press, Cambridge. SIMMONS, M. (1993). A catalogue of the herbarium of the British flora collected by Margaret Stovin (1756— 1846). Dorman Museum, Middlesborough. STACE, H. E., PETTITT, C. W. A. & WATERSTON, C. D. (1987). Natural science collections in Scotland. National Museums of Scotland, Edinburgh. WEBB, D. A. (1988). The Irish and British plants in the herbarium of Trinity College, Dublin. 1. The vascular plants. Jrish naturalists’ journal 22: 482-485. WILLIAMSON, W. C., ed. WILLIAMSON, A. C. (1896). Reminiscences of a Yorkshire naturalist. G. Redway, London. WOODWARD, B. B. Anne Pratt, in LEE, S., ed. Dictionary of national biography 46: 284—285. Smith, Elder & Co., London. (Accepted April 1999) fimt at i Watsonia 23: 23-38 (2000) 23 The herbarium of William Skrimshire (1766-1829) of Wisbech G. CROMPTON 103 Commercial End, Swaffham Bulbeck, Cambridge, CB5 OND and E. C. NELSON* c/o Wisbech & Fenland Museum, Museum Square, Wisbech PE13 1ES, Cambridgeshire ABSTRACT William Skrimshire (1766-1829), a surgeon of Wisbech, Cambridgeshire, assembled a substantial herbarium before 1818 when he sold it to Viscount Milton. Subsequently Skrimshire went on collecting plants in the vicinity of Wisbech and adjacent counties. He was in contact with other botanists, most notably the Revd Richard Relhan, author of Flora Cantabrigiensis, and obtained specimens from them. Other members of the Skrimshire family, especially his brother Dr Fenwick Skrimshire, also contributed to his herbarium. While a manuscript catalogue of William Skrimshire’s extensive herbarium survives, providing valuable records of the flora of Cambridgeshire, only about 200 specimens from his herbarium can now be traced; these are preserved in Wisbech and Fenland Museum (WBCH) and some constitute hitherto unpublished first records of Cambridgeshire and Northamptonshire species. The extant specimens show that William Skrimshire was an inquisitive botanist and observant gardener. They also point to changes in the environment of Wisbech and its hinterland during the past two centuries. Some significant specimens from the herbarium of Skrimshire’s one- time pupil and fellow surgeon, John Rose Weatherhead (c. 1787-1849), are also noted. KEYWORDS: Cambridgeshire, Norfolk, Northamptonshire, history, John Rose Weatherhead. INTRODUCTION The Wisbech Museum (now Wisbech and Fenland Museum) was founded in 1835, and today still occupies the same building, in the town centre, that was erected in 1846 (Arthur 1985) to house the collections. Among the earliest donations were numerous natural history specimens, including several collections of pressed and dried plants which had been made mainly in the vicinity of the town by local amateur naturalists; the donors were Richard Dykes Alexander and Miss Alexander, J. R. Weatherhead, Mrs Algernon Peckover and Charles Metcalfe. In 1862, Mrs W. G. Townley and Mrs J. R. Weatherhead donated two substantial herbaria; the Weatherhead hortus siccus comprising seven volumes contained few specimens collected later than 1835. While the subject of our paper, the Wisbech surgeon William Skrimshire, had died six years before the Museum was established, specimens that he had gathered in the counties of Norfolk, Cambridgeshire, Northamptonshire and Lincolnshire (v. cc. 27, 28, 29, 32, 53 respectively) during the 1790s and the first three decades of the 1800s, came to Wisbech Museum because they were contained in those separate herbaria. The Wisbech Museum herbarium (WCBH: Kent & Allen 1984) has not been added to for many decades, so its principal interest is as an historic collection focused on the flora of Wisbech town and hinterland during the first half of the 19th century. For other reasons, many of the extant specimens are by far the earliest surviving specimens known to have been collected in Cambridgeshire (v. c. 29) and Northamptonshire (v. c. 32), and a smaller number are first records for West Norfolk (v. c. 28; these were noted by Petch & Swann (1968: 15)). With regard to *Address for correspondence: E-mail: tippitwitchet @ zetnet.co.uk 24 G. CROMPTON AND E. C. NELSON Cambridgeshire (which is the special interest of G.C.), while the foundations of the herbarium in the Department of Plant Sciences at the University of Cambridge (CGE) were made by Professor John Henslow from 1818 onwards, Henslow’s first specimens from the Wisbech area are dated 1831. The Wisbech Museum collections therefore are of considerable interest because they predate most of the material at Cambridge and so provide some voucher specimens for the first record of species from Cambridgeshire. Information about the herbarium in Wisbech Museum was incorporated in D. H. Kent & D. E. Allen’s British and Irish herbaria (1984: 71); they employed the acronym WBCH. The Museum Documentation Association’s acronym for the Wisbech and Fenland Museum is WISFM, and all the herbarium specimens, which are presently being catalogued by E.C.N., now bear accession numbers prefaced by that acronym (e.g. WISFM: 1862.17.2.86). A fully annotated list of the specimens discussed in this paper can be obtained from The Curator, Wisbech and Fenland Museum; this list includes transcriptions of annotations on the specimens and the original nomenclature. In the following account all botanical names have been updated according to Stace (1991). When the botanical names inscribed on a herbarium specimen differ from present-day nomenclature the names in manuscript are quoted in brackets, e.g. Helianthemum nummularium (as “Cistus Helianthemum’’). WILLIAM SKRIMSHIRE’S SPECIMENS AND MANUSCRIPTS We are able to record and assess the extent of William Skrimshire’s botanical interests and expertise by examining his reliquiae, which comprise three separate items preserved in Wisbech and Fenland Museum, namely: | His own manuscript catalogue of his herbarium; this was first described by Perring (1956; see also Bridson et al. 1980 (entry 435.1); Crompton 1994, 1996; Petch & Swann 1968). 2 An assortment of approximately 200 specimens, mainly representing higher plants, of miscellaneous provenance. 3. An hortus siccus of bryophytes (see Bridson et al. 1980) (this is not discussed here). It should be noted that Petch & Swann’s (1968) reference to “a large collection” of William Skrimshire’s specimens in Wisbech Museum is inaccurate; Skrimshire’s material represents a very small fraction of the Museum’s herbarium (E. C. Nelson, unpublished data). There are also some specimens incorrectly ascribed to William Skrimshire in Norwich Museum (NWH: see Kent & Allen 1984: 60, 244). The identity of the collector of this series of 300 specimens, all collected in 1801, is not known; despite an annotation in the Museum’s register (23.91), William Skrimshire was not the person who assembled it, although a few specimens came from his garden (see below). THE SKRIMSHIRE FAMILY William Skrimshire (1766-1829) was the oldest son of William Skrimshire (1739-1814), and he had three brothers, Thomas, Fenwick, and George, and a sister, Elizabeth. Both Williams, father and son, were surgeons in Wisbech, and until his father’s death William junior probably lived with, and practised with, his father in Ship Lane. Later he moved to the outskirts of the town, to Bushy Place — while the property’s exact location has not been discovered, Bushy Place was almost certainly situated on the eastern bank of the River Nene, less than a mile north-north-east of the centre of Wisbech near Horseshoe Corner. There William had a garden, including a “peppermint plantation” where he evidently grew peppermint (Mentha xX piperita) in some quantity presumably for distillation for medicinal use. William Skrimshire never married; he died in Wisbech on 22 July 1829. Fenwick Skrimshire, born at Wisbech in 1775, took his medical degree at Edinburgh, and dedicated his thesis, De assimilatione (F. Skrimshire 1798), to his father and to his brother William. During 1797-1798 Fenwick was President of the Natural History Society of Edinburgh WILLIAM SKRIMSHIRE (1766-1829) 25 (this society was established in 1782, and survived until 1812 when it amalgamated with the Royal Physical Society). During this period, Fenwick sent his brother many plant specimens from Scotland, including Chrysosplenium alternifolium (“From Dickson’s Garden, Edinburgh ...”), Gnaphalium sylvaticum (“Given to my Brother Fenwick by his friend Mr. Brown. 1796.”), a wild pansy from Ben Lawers that had been named “Viola montana” by John Mackay (1772-1802), and Campanula rapunculoides (“A very good discovery this, and as such Dr. Smith has noticed it in his Mem. A doubt arises with me whether it be not a reyectamentum Horti, as it is cultivated in Gardens. You will do well to be quite certain about this, as otherwise it cannot well be admitted as a native of our Island. ... Discovered by my Brother Fenwick Skrimshire, M. D. From Blair, Scotland. 1796.”). Dr Fenwick Skrimshire settled in Peterborough where he was physician to the Fitzwilliam family (Fenland notes & queries 4: 211, 212 with copy from Stamford Mercury 5 November 1819), and by 1838 he was “Physician to the Peterborough Public Dispensary and Infirmary”. He attended the poet John Clare (1793-1864), who was a native of Helpstone, a small village between Peterborough and Stamford whence came a number of the specimens collected by William Skrimshire (see below). In 1841 Fenwick Skrimshire filled in the application for Clare’s admission to Northampton Lunatic Asylum, and on 29 December 1841 he certified Clare “following years addicted to poetical prosings.” Fenwick Skrimshire wrote a number of books (see References) including A series of essays introductory to the study of natural history, the eleventh essay in this was entitled “The useful application of botanical knowledge” (F. Skrimshire 1805: vol. 2, pp. 125-181). He died in 1855. Thomas, whose date of birth has not been discovered, was admitted a pensioner at Clare College, Cambridge, on 22 December 1791, and matriculated at Michaelmas 1793. Subsequently he “migrated” to Magdalene College under Holmes Charity, and graduated in 1798. He was later ordained and held a number of livings in Norfolk between 1800 and his death in 1836. While not himself a contributor to his brother’s herbarium, Thomas’ son George, who was apprenticed for five years (from 2 July 1817 to 31 March 1823; see Crompton 1994) to his uncle William Skrimshire, did; a specimen of Aceras anthropophorum “From Helpstone, near Milton. Northamptonshire. Given to me by my Nephew G. Skrimshire. 1821” is a first record for that county (v. c. 32; G. Gent pers. comm.; see below). The fourth brother, George, a subscriber to one of Fenwick’s books, was a farmer at Kettlestone, Norfolk. Their only sister, Elizabeth, married Charles Metcalfe, a solicitor, and their daughter “my niece Eliz. Metcalfe” sent specimens to William Skrimshire shortly before her early death in 1824 aged 22. Elizabeth Metcalfe’s specimens include Epilobium parviflorum and E. tetragonum from Emneth, a village near Wisbech, an unlocalized specimen of Helianthemum nummularium (as “Cistus Helianthemum’’), and Minuartia verna (as “Arenaria verna”’) and Parnassia palustris both from Derbyshire (v. c. 57). As noted above, Charles Metcalfe was one of the early benefactors of Wisbech Museum, presenting his brother-in-law’s collection of mosses. William, Thomas and Fenwick Skrimshire were keen entomologists. William and Thomas were elected Fellows of the Entomological Society, and both published papers in the first volume of the society's Transactions (see References). While Thomas Skrimshire “purpose[d] doing myself the honour of transmitting the Society a letter annually upon this subject [rearing insects]”, we have traced only one published letter from him (T. Skrimshire 1812). The large copper butterfly, a native of the “marshes of Cambridgeshire”, was discovered by Adrian Haworth “and my very dear friends, W. Skrimshire and F. Skrimshire, M. D.” (Haworth 1803). Tutt (1905) commented that the Skrimshires probably saw the butterfly near Ely in 1797 or 1798 and afterwards went with Haworth to capture it, perhaps in Bardolph Fen, a short distance east of Wisbech in West Norfolk. WILLIAM SKRIMSHIRE AS NATURALIST Dr Fenwick Skrimshire (1802) characterised his older brother as “‘a very accurate and attentive observer of nature”, and, judging by William’s published articles (see References), as well as notices of plants and animals he collected, his interests were broad-ranging from plants and insects to the weather and phosphorescence. He reported that the roasted seeds of Iris pseudacorus could be used as a substitute for coffee (W. Skrimshire 1809); Grigson (1955: 419) noted this, but merely credited the discovery to a “Fenland doctor’. More than a century later, during World War II, well- 26 G. CROMPTON AND E. C. NELSON roasted seeds from wild iris were used as a coffee substitute in the Channel Islands (see Vickery 1995: 409). William Skrimshire assembled a substantial herbarium (hortus siccus). His manuscript catalogue of this now-lost herbarium is extant in Wisbech and Fenland Museum. The manuscript, purchased in 1829 by J. R. Weatherhead (see below) within two months of William Skrimshire’s death, bears this annotation on the fly-leaf: Catalogue of Plants contained in Mr Skrimshire’s Hortus Siccus sold some years since to Lord Milton — and also of Plants contained in a Hortus Siccus bought by me and collected by Mr Skrimshire after the sale of the former. The mark Y indicates those contained in this collection. Sept. 12th 1829. Comparison of the manuscript catalogue with specimens in Weatherhead’s herbarium, presented to the Wisbech Museum by his widow, shows that very few of the ticked specimens have survived. Viscount Milton (Charles William Wentworth Fitzwilliam, 1786-1857), son and heir of the second Earl Fitzwilliam whose physician was Dr Fenwick Skrimshire, had paid seventy pounds for William Skrimshire’s collection of minerals and his hortus siccus in February 1818 (“Feb. 2 paid Dr [Fenwick] Skrimshire for his brothers minerals & Hortus Siccus (by draft on Snow) [£] 70.00.” Northamptonshire Record Office Misc. Vol. 236: p. 75 (Private accounts (payments) of Fitzwilliam. Jan. 1812—May 1822. Fitzwilliam (Milton)). Lord Milton was a patron of natural history, “one of the oldest and most zealous friends of Natural Science in this country” (see Desmond 1994). He was one of the few subscribers to Flora Graeca (1806-1840) to receive the entire work (see Lack 1999: 213). No trace of Milton’s herbarium and mineral collections has been found, and we must assume they no longer exist. However it is clear from the dates in William Skrimshire’s manuscript catalogue and on herbarium specimens in WBCH that as soon as he had sold his herbarium to Lord Milton, he immediately started collecting and pressing plants again. Like any modern botanist William Skrimshire often gathered, and evidently retained, a number of duplicate specimens, and thus there are some pre-1818 specimens in WBCH, derived from the now-amalgamated herbaria of Skrimshire’s relatives and friends. The earliest date on an extant specimen, among his brother-in-law Charles Metcalfe’s donation, is 1791; that year William Skrimshire gathered Origanum vulgare at Hunstanton, West Norfolk. The series of specimens shows that over the next 38 years Skrimshire not only collected in and around Wisbech but also wandered through Cambridgeshire and adjacent counties in search of interesting plants. Skrimshire’s diligence in observing and collecting plants means that he can be credited with the first records of several species for Cambridgeshire (v. c. 29) and Northamptonshire (v. c. 32) (see Table 1). Although his contribution to Cambridgeshire first records was published in Perring ef al. (1964), those data were based only on Skrimshire’s manuscript catalogue and Relhan’s Flora Cantabrigiensis (2nd edition 1802). Further research by G.C., particularly into the historical records and the Wisbech Museum herbarium, allows us to publish a revised first records list for the county (Table 1). The Norfolk records noted by Petch & Swann (1966: 15) all derive from Skrimshire’s catalogue; they are Cladium mariscus (as Schoenus mariscus) and Utricularia vulgaris, both from Bardolph Fen, and Persicaria bistorta (formerly Polygonum bistorta) from Framingham (see Petch & Swann 1966: 165). The Northamptonshire data have not previously been published; its flora was much less well documented in the early nineteenth century than that of, for example, Cambridgeshire, so Skrimshire’s specimens add significantly to the tally of first records. One set of specimens provides a vivid record of one of Skrimshire excursions, which took place just three months after the sale of the hortus siccus to Lord Milton. Skrimshire travelled from Wisbech via Peterborough to Milton and Helpstone, the home-village of John Clare. On 25 May 1818, William Skrimshire foraged in Thistlemoor Wood near Milton, Northamptonshire, finding Ajuga reptans, Anemone nemorosa, Fragaria vesca (as “Potentilla sterilis”), Ranunculus auricomus and Viola canina. On the following day, perhaps having stayed overnight with Fenwick in Peterborough, William found Pentaglottis sempervirens (as “Anchusa sempervirens”) in the cathedral precincts; “there were 3 or 4 Roots, growing in the Minster Yard at Peterborough on the E. side of the Building. May. 26. 1818”. As he often did, Skrimshire also commented about his knowledge of the species elsewhere: “It is not found in the vicinity of Wisbech.” On 27 May, he returned to the Milton area and visited Helpstone, gathering Helianthemum nummularium (as “Cistus Helianthemum’’) and Hippocrepis comosa. In some old gravel pits at Helpstone, he found WILLIAM SKRIMSHIRE (1766-1829) 2 TABLE 1. FIRST COUNTY RECORDS FOR CAMBRIDGESHIRE AND NORTHAMPTONSHIRE Cambridgeshire (v.c. 29). § Adonis annua: (as “Adonis autumnalis”’); near Bowling Green, Wisbech, c. 1820. * Anagallis arvensis subsp. arvensis f. azurea: (as” Anagallis coerulea. Abbot ... A. arvensis var. 2”); Bushy Place, Wisbech, 26 October 1822. § Atriplex littoralis: marshes below Wisbech, 1795. § Atriplex pedunculata: near The Black Tunnel below Bushy Place, Wisbech, 1826. Beta vulgaris subsp. maritima: Relhan (1802). Cardamine hirsuta: between Wisbech and Tydd Gote, 1794. Cochlearia anglica: salt marshes below Wisbech, 1796. Cochlearia danica: salt marshes below Wisbech, 1796. Corydalis lutea: (as ““Fumaria lutea’) Mr Peckover’s garden wall, Wisbech 1818. Datura stramonium: Long Drove, Wisbech, 1818. Doronicum pardalianches: Bushy Place, Wisbech, 1818. Limonium bellidifolium: below Wisbech, 14 September 1795 (W. Skrimshire to J. Sowerby) [NB perhaps a Lincolnshire record; see discussion on p. 10.] § Mentha spicata: (as “Mentha viridis”); Bushy Place garden, Wisbech, 1813. * Misopates orontium: Bushy Place and Crab Marsh, Wisbech, July 1819. § Orobanche ramosa: (Wisbech, before 20 November 1795. T. J. Woodward to J. E. Smith); hemp fields about Wisbech, c. 1820. Rubia tinctoria: (as “Rubia peregrina” in ms catalogue); Relhan (1802, as ‘Wild Madder’). Ruppia cirrhosa: (as “Ruppia maritima’); salt-water ditch, east of Wisbech Tydd Gote road, 1795. Silene uniflora: (as “Silene maritima’); see Perring (1956). Sisymbrium irio: near Barton Lane, Wisbech 1797. Triglochin maritima: Relhan (1802). CHR MU KK %& UM CR LPR % M&M Northamptonshire (v. c. 32). * Aceras anthropophorum: Helpstone, 1821; leg. George Skrimshire. Ajuga reptans: Thistlemoor Wood, near Milton, 25 May 1818. * Anemone nemorosa: Thistlemoor Wood at Milton, 25 May 1818. Bupleurum rotundifolium: (as “Euphorbia hiberna’’); neighbourhood of Peterborough, 1821; /eg. George Skrimshire. Epipactis helleborine: (as “Serapias latifolia ... Epipactis latifolia’); Milton Wood near Peterborough, 1816. Euphorbia exigua: Peterborough, between that City and Thorpe, 1821. Jeg. George Skrimshire. Fragaria vesca: (as “‘Potentilla sterilis’”); Thistkemoor Wood at Milton, near Peterborough, 25 May 1818. Hippocrepis comosa: Helpston, 27 May 1818. Pentaglottis sempervirens: (as “Anchusa Sempervirens’’); Peterborough Minster-yard, 26 May 1818. * Ranunculus auricomus: Thistlemoor Wood near Milton, 25 May 1818. Viola canina: Thistlemoor Wood, at Milton, near Peterborough, 25 May 1818. YC ey waieicay cua} Unless otherwise stated, these records are attributed to William Skrimshire. * = voucher specimen exists in WBCH; § = record in William Skrimshire manuscript catalogue. a splendid array of orchids including Orchis ustulata, as well as Pulsatilla vulgaris (as “Anemone Pulsatilla”). Although dated June 1818, a specimen of Ophrys apifera, which he called “Ophrys aranifera. Spider ophrys”, was also picked from the gravel pits. Viola canina grew on “a small Heath or Cover — on the left hand between Milton, and Helpstone”’. Perhaps it was on one of his journeys to see his brother at Peterborough that Skrimshire stopped at Cross Guns, a public house on the bank of the Nene at Wisbech St Mary’s parish boundary; the site of this pub is now occupied by the Cross Guns Pumping Station (TF/347.015). The nearby marsh yielded, for example, Schoenoplectus lacustris (as “Scirpus lacustris”) and Potamogeton perfoliatus (there are no voucher specimens for these in WBCH). Black Boy Low was another small pub, a short distance east of Wisbech almost on the Cambridgeshire-Norfolk border, illustrated by Algernon Peckover in 1841 (reproduced in Crompton 1994). Skrimshire collected 28 G. CROMPTON AND E. C. NELSON Geranium pratense “By the side of a ditch adjoining the Toll Bar between Wisbech and the Black Boy Low at Walsoken”. Further into West Norfolk, William Skrimshire visited Dersingham, and there collected plants typical of heath and bog, Erica cinerea and Drosera intermedia (as “Drosera longifolia”) and D. rotundifolia. The voucher specimens are not dated. As for the round-leaved sundew, Skrimshire annotated it with a negative comment: “I have not found it nearer Wisbh. than Bardolph Fen.” The specimens of Pentaglottis sempervirens and Drosera rotundifolia jast mentioned illustrate a remarkable aspect of William Skrimshire’s herbarium. Frequently he added comments that inform us about plants not found in the vicinity of Wisbech. In this regard, Skrimshire is most unusual because the overwhelming majority of herbarium specimens are only labelled with the actual locality where each specimen was obtained. Another good example, displaying his intimate knowledge of the local flora, is a specimen of Silene latifolia labelled: ... Lychnis dioica [?] Smith L. vespertina Sibthorpe var. f. White Campion, or Cuckow flower. In a hedge at Bushy-Place 1818. Not uncommon in hedges and by Road’s sides about Wisbech, but I have never observed the red flowered one, within 20 miles of this place; and I am persuaded that it is a distinct species from this Plant. W. S. 1818. William Skrimshire was not just a gatherer of specimens. He was observant and inquisitive, and one of the plants that intrigued him was Sagina maritima which grew abundantly in the salt marshes on the banks of the River Nene (see Fig. 2) below Wisbech, not far from Bushy Place. Two series of specimens, each carefully pressed and mounted, are annotated at length (see Fig. 1): ... Sagina procumbens. var. maritima. These Specimens gathered the beginning of May, on the Salt Marsh near Bushy Place, grew upon the bare soil from which the Grass-sward had been cut the preceding year; and as the root leaves grew in a perfectly star-like form and the branches spread upon the bare ground as radii from the centre of the Plant the whole had a striking appearance. I labelled it at the time as above; but from frequent and attentive observation on the growth of this plant I have no doubt of its being the same with Dr. Smith’s Sagina maritima. See the series of specimens on this sheet [f. 2] No. 1 Gathered the later end of May } 1818. No. 2 Gathered the middle of June } 1818. From the Salt Marsh near Bushy Place. f. 3 These three specimens gathered on the Salt Marsh near Bushy-Place June 30th. 1818. These six specimens were gathered by the River side below Wisbech. Near Bushy Place. Sr. J. E. Smith has made / this Plant / a distinct Species, under the specific name of maritima. Dickson published this plant under the name of Sagina apetala, in his Hort. Sicc. Brit. Fasc. HI. But I think it differs from S. apetala sufficiently to entitle it to be ranked as a distinct Species. W.S. 1818. James Dickson (1738-1822), one of the founders of the Royal Horticultural Society and of the Linnean Society of London, “a most wonderful fellow” according to the Revd Richard Relhan, issued his Hortus siccus Britannicus: being a collection of dried British plants, named on the authority of the Linnaean herbarium ... in 19 fascicles between 1793 and 1802 (Henrey 1975: ILL, 34). Although Skrimshire evidently had an extensive library of contemporary botanical works including William Hudson’s Flora Anglica, Richard Relhan’s Flora Cantabrigiensis, and James Edward Smith’s Flora Britannica, he did not possess the relevant fascicle of Dickson’s Hortus siccus. However John Nainby of King’s Lynn had a copy and generously removed the specimen labelled Sagina apetala and gave it to Skrimshire so that he could compare it with those from his immediate stamping grounds. And Skrimshire duly and accurately labelled it: ... Sagina apetala. S. maritima. Comp. Fl. Br. * This specimen was taken from Dickson’s Hort. Sicc. Brit. Fasc. III. and given to me by John Nainby of Lynn. This is exactly similar to what I find by the river side at Bushy Place, & which Dr. Smith has made a distinct species. 1818. Plants that excited Skrimshire’s curiosity were often gathered at different seasons, so that his herbarium specimens sometimes comprise more than one gathering. For example, he found an odd-looking Lamium purpureum at Bushy Place and was uncertain about its exact identity: Skrimshire labelled it “... Lamium incisum. Smith L. dissectum. With. L. purpureum [?] Huds.”. He gathered specimens in March and April 1818, and mounted them together. Another sheet with WILLIAM SKRIMSHIRE (1766-1829) 29 LLL OLE Ri pee petecumberd, “AS RiApellmmle . Mes Ppeeimend gall a Keg hoff: Zé heganeng / use SAY Lay, i The ere OTL da VF /2 ay led BY tibe2 7 ae: tke tt Mifeo Thi é g 2 Lhe y = FLOWS t4O L207; By, Z ai . ere Je j Hom whieh Wei Gited 20 fee CEH: Bie JF iy L472 ¢ Z asth: Vin tia ike“ 4g ¥ KEY i LH a ‘fie ee Ley L. EAS # i : 7, SEM) Le OES aE The Crun shes uf? head ae (lin bbdt pe GIONS td 7} Ye j ? Keb‘. IOS. ee: ore G ft bleh, - FFA tt ft re flitd a dt: ping df = Jathtint e CUD LE tf AV TG lint. Lf MOGUL, Cath hb J Pe fltewl Pts (ahlen he ‘fel } y Ae ; ; / g SA ELUM Oy’ & Me Show ¢ Vhio be t/ (VALU NO ¢ es i bt bow Ie tot f a4 Pg Z : See MH ‘LS 7 ‘Sr a7 ee yine mate: Lynd wl és [Ae Sex Pied Sheth FIGURE 2. Alexander Peckover’s watercolour showing the River Nene in 1825 (“Nene from near the Black Sluice in 1825”). The location was a few miles north of Wisbech town, and the tower and distinctive spire of the parish church can be seen right of centre on the horizon. Bushy Place, William Skrimshire’s property, was between the Black Sluice and the town, and so would lie towards the right-hand margin of this view. (Reproduced by courtesy of Dr Peter Cave, and Wisbech and Fenland Museum; ©Wisbech and Fenland Museum. ) four specimens of Mentha x piperita displays the stages in the growth of peppermint: . M. piperita sylvestris. Sole. No. 1. A young Shoot, in Spring. No. 2. The Stem in Summer before it flowers. No. 3. One of the Stolones from the main Plant in Autumn. No. 4. The Plant in full flower. 1813. He frequently noticed colour variants. A good example is Geranium molle: “These specimens, one with Pink and the other with pale coloured Blossoms were gathered by the Road side at Bushy- Place, May 15. 1819. Common about Wisbech. 1819.” The previous year on 29 June 1818 he had obtained it “With pale coloured Blossoms”, probably from the same roadside bank. WILLIAM SKRIMSHIRE’S GARDENS William Skrimshire was evidently a keen gardener, and was just as interested in growing native plants as exotics. From annotations on specimens and in his manuscript catalogue, we know that he had a garden in Wisbech town, perhaps at Ship Lane where his father had lived, as early as 1801. It was from this garden that several of the anonymous specimens, now in the Castle Museum, Norwich (NWH), came: Alchemilla vulgaris “May 13™ 1801 ... Archemilla [sic] - Vulgaris ... gathered in Mr. W. Skrimshire’s Garden Wisbech ...”, Geranium lucidum and Cymbalaria muralis (as “Antirrhinum Cymbalaria’), as well as Ajuga reptans (although the latter specimen is not dated). John Weatherhead collected Allium ursinum “in Mr Skrimshire’s garden Wisbech [in] May 1807” (WBCH). One interesting annotation in his manuscript catalogue prea that he sowed seeds of Arabis WILLIAM SKRIMSHIRE (1766-1829) 31 turrita “on my Garden Wall at Wisbech where it continued to grow for several years”; this may have been as early as 1796 when the Revd John Hemsted (see below) gave Skrimshire a herbarium specimen of the species. By 1813 Skrimshire had also developed a new garden and peppermint plantation at Bushy Place. Again using his annotations and notes, we can build up a picture of Bushy Place. It was probably surrounded by walls: Skrimshire gathered Erodium moschatum “on an old Wall at Bushy Place, Wisbech; May 21st. 1818. Very rare.” As well as the garden proper, there was the “peppermint plantation’, presumably a field in which he grew a crop of Mentha x piperita. There was an orchard, and “nut trees” (perhaps hazels) under which Doronicum pardalianches grew. The peppermint plantation was adjacent to Long Drove. Skrimshire found Epilobium hirsutum “In Long Drove, in the hedge of my Peppermint Plantation. Augt. Ist. 1818. It grows in a hedge at Tip-cock-Trees about a mile below Bushy-Place. Not very common at Wisbech”. The still-house in which he would have distilled the essential oil from the peppermint was also by Long Drove: Vicia hirsuta (as “Ervum hirsutum’) grew in “a Cornfield opposite my Still-house in Long Drove, Crab-Marsh. July 1. 1822”. (A house name “Mint House” is marked on the 1887 six-inch Ordnance Survey map at TF/457.109; it may have been built on the former site of Skrimshire’s still-house.) He was fascinated by the weeds that grew in his garden and surrounding fields and often gathered specimens for his herbarium. For example the “Anagallis coerulea. Abbot ... A. arvensis var. 2. ... female pimpernel” (Anagallis arvensis subsp. arvensis) which “was wont to grow as a weed annually in my Old Garden at Wisbech”, was also “among the Peppermint in the Orchard at Bushy-Place; Wisbech” in October 1822. The voucher specimen represents A. arvensis L. subsp. arvensis f. azurea Hyl. (det. G. Crompton), and is the first record of this forma from Cambridgeshire. While he deliberately cultivated Mentha x piperita, Skrimshire found that M. arvensis Was a “troublesome weed”, and he also gathered M. spicata in his garden at Bushy Place as early as 1813 according to a note in his manuscript catalogue. Herniaria glabra was collected by J. R. Weatherhead, who noted that “This came up as a weed in Mr Skrimshire’s garden”. The manuscript catalogue contains several other weeds. By 1828 great willowherb had become a weed at Bushy Place. Euphorbia exigua “used to come up annually as a weed in my garden ...” (the voucher specimen came from his nephew, George Skrimshire, and was collected in Peterborough); Tussilago farfara grew “On the Ploughed grounds by the river side, below [Wisbech]” and also was “in my Garden as a Weed at Bushy-Place. 1822”. Other casuals were Campanula patula “among the Peppermint in the Orchard, Bushy-Place, July 22, 1820”, and Chenopodium urbicum. Carex spicata grew in his “Grass-field in Tinker’s Drove. Wisbh.” — presumably the pasture for his horse (or horses) that kicked him (see below)! Like present-day gardeners, Skrimshire battled with such weeds as Convolvulus arvensis, and his annotated specimens vividly illustrate his frustration; at Bushy Park in June 1818 this was “one of the most troublesome weeds that I have to contend with”. Solanum nigrum was condemned in stronger terms: “A common weed in Gardens, about Wisbech. It is one of the pests in my Peppermint Plantation at Bushy Place”. Less troublesome, perhaps, but no less worthy of pressing and mounting for his herbarium, were Lamium amplexicaule “Gathered at Bushy-Place, Septr. 21st. 1818. A Common Weed both in Spring & Autumn, in Gardens & Cornfields in Crab Marsh, Wisbech.”; Lapsana communis “From Bushy-Place, July 2nd. 1819 Common weed in Gardens and under hedges. 1819.”; and Misopates orontium (as “Antirrhinum Orontium’’) “From my Garden at Wisbech, where it constantly comes up as a Weed. 1. July. 1819. I have also observed it as a weed in my Garden at Bushy-Place, and in Cornfields in Crab-Marsh, Wisbech. 1819.” THE RIVER NENE BELOW WISBECH The grounds of Bushy Place merged with the River Nene (Fig. 2). Although the Nene is still tidal at Wisbech, Skrimshire’s herbarium specimens, and his manuscript catalogue, reveal the greater extent of saline habitats, including salt marshes and flats, in the vicinity of Wisbech two centuries ago, and thus also the extent to which maritime species penetrated inland in the early 19th century. There were substantial areas of salt-marsh not far from Bushy Place, on both west and east banks of the Nene, where he gathered Salicornia fragilis (as “Salicornia herbacea”), Puccinellia maritima (as “Poa maritima, Glyceria maritima”), Centaurium pulchellum (conf. G.C. & E.C.N. a2 G. CROMPTON AND E. C. NELSON 1999), and several scurvy-grasses (including Cochlearia danica; cont. T. C. G. Rich 1999). Other species from saline (or maritime) habitats that Skrimshire found near Wisbech, but which are now lost to the local flora, were Limonium bellidifolium (as “Statice reticulata”) and Atriplex pedunculata. Limonium bellidifolium was reported as being “very common on the salt marshes below Wisbech” by Skrimshire in a letter to James Sowerby dated 14 September 1795 (Boulger & Britten 1918: 259-261). Sowerby later illustrated this species in English botany (1796: tab. 328) using material collected “below Wisbech” and supplied by the Revd John Hemsted. A specimen (WCBH; Fig. 3) gathered by Skrimshire, “Growing with the S[tatice] Limonium on the marsh between St Peter’s Point and the Washway Aug. 14. 1821”, perhaps represents a Lincolnshire (v. c. 53) record due to the convoluted vice-comital boundaries. The same must be said of Hemsted’s gathering. In 1818 Skrimshire saw Silene uniflora (formerly S. maritima) “on the River-Bank on Eastfield” and entered the locality in his manuscript catalogue; as Perring (1956) noted, this was a new record for Cambridgeshire. The banks of the Nene near Bushy Place were also inhabited by Arenaria serpyllifolia, Cerastium glomerata (as “Cerast. vulgatum. Smith Cer. viscosum With. ... Cerastium viscosum”; det G.C. & E.C.N. 1999), Mentha arvensis, Papaver rhoeas (as “Papaver Rhaeus var?”) and Trifolium campestre (as “Trifolium procumbens”). Other “river bank” plants, noted in Skrimshire’s manuscript catalogue, include Briza media, Valerianella locusta (as “Valeriana locusta”) and Bupleurum tenuissimum. Carlina vulgaris grew “upon Mount Pleasant with the Carduus nutans and upon Banks about Wisbech, otherwise not common here.” The local hedgerows yielded a bramble, Rubus cf. corylifolius (as “Rubus [fruticosus deleted] corylifolius. Smith. R. fruticosus var. 2. major. With[ering].”), and a “red purple variety” of Viola odorata. WILLIAM SKRIMSHIRE’S ASSOCIATES William Skrimshire did not botanize in isolation. He maintained contacts with many botanists both amateur and professional. An entry in Skrimshire’s manuscript catalogue provides an important record of one of his contacts. Blackstonia perfoliata (as “Chlora perfoliata”) was “gathered ... at Cherry Hinton, near Cambridge; in company with my friend Relhan. 1792” (there is no corresponding herbarium specimen in WBCH). The Revd Richard Relhan had published his Flora Cantabrigiensis in 1785 and it contained some records from Wisbech, but none ascribed to Skrimshire. We do not know when he became a friend of Skrimshire, but there are five herbarium specimens in WBCH annotated by Skrimshire as having come from Relhan. These derive from the J. R. Weatherhead hortus siccus donated to the Museum in 1862 by Weatherhead’s widow, and are Gnaphalium luteoalbum “From Shelford, Cambridgeshire”; Malva neglecta (as “ Malva verticillata’) “Discovered in Cambridgeshire a few years since and sent to me by my worthy friend The Revd. R Relhan. 1818.”; Clinopodium grandiflorum (as “Melittis grandiflora, Melissa grandiflora’) “from Mr. Relhan’s Garden, at Cambridge. 1795.”; Pyrus communis “From Hinton near Cambridge...”; and Potamogeton compressus (without a locality; conf. C. D. Preston). There is one other Relhan specimen in Norwich (NWH); Trifolium subterraneum was annotated first by Skrimshire — “Given to me by Mr. Relhan. I have not found it about Wisbh. 1795” — and subsequently by the anonymous owner, “Mr. Skrimshire gave me this Specimen ...”. Four further Relhan specimens in WBCH do not appear to have been in Skrimshire’s possession, rather they were somehow acquired by J. R. Weatherhead; these are Carex cf. viridula (as “Carex Oederi?’”’), Eleocharis acicularis (as “Scirpus pauciflorus?”; conf. C. D. Preston, G.C. and E.C.N. 1998), Jasione montana and Liparis loeselii (as “Malaxis Loeselii, Ophrys Loeselii’). As well as indicating a close link between these two men, the specimens reveal that Relhan may have had a garden. As we have already noted, Skrimshire was an enthusiastic and observant gardener, and perhaps he and Relhan exchanged living plants as well as herbarium specimens. In the second edition of Relhan’s Flora Cantabrigiensis (1802) there are more than 30 new county records ascribed to Skrimshire, including many for mosses, lichens and algae. Relhan (1802) paid this tribute to his friend, acknowledging William Skrimshire’s contribution to this edition, and stating that Skrimshire was “happily devoted to natural science and most skilled in his surgeon’s art.” WILLIAM SKRIMSHIRE (1766-1829) 33 j) : re | “67; - MCplipuirtl Fr tyne, SPE - 4 Hatice reLoutaale . Maller, hoa Liver tL? Or S epeet Jatt lati ted Ceivw Wiibee/ Gryiwindg swith the My pil ri 72 Whiz ; pp fips | a res ees ; Lelwetn e/- d Lets Pyar a7 : 4 5 The Warsyird tt. « ees FIGURE 3. This specimen of Limonium bellidifolium may have been collected in Lincolnshire (v. c. 53) (WBCH): Skrimshire first reported the species from the Wisbech area in September 1795. 34 G. CROMPTON AND E. C. NELSON Likewise, Skrimshire received many specimens, particularly bryophytes from the Newmarket area, from the Rev. John Hemsted (c. 1747-1824), one-time Curate of St Mary’s, Newmarket (Oswald 1991). Skrimshire’s acquaintance with Hemsted is confirmed by Hemsted’s correspondence with James Sowerby (Mss in Sowerby Correspondence Vol. 26. The Natural History Museum, London). On 13 September 1796, Hemsted informed Sowerby that: ... [have received an Invitation from my Friend Skrimshire to come and spend a few Days with him which I propose to accept & to go down to Wisbeach next Monday, stay with him to the latter End of the Week & then go across Country to Bedford ... after which I shall be out no more & shall be happy to see you & Mr Skrimshire desires me to say that if you can give us the Meeting at Wisbeach he shall be exceedingly glad of your Company... [Vol. 26: 171]. Just over a week later, on 22 September 1796 Hemsted admitted he “was disappointed of my Excursion to Wisbeach. Our Friend Skrimshire having got a Hurt in his Leg by a Kick of a Horse ...” (Vol. 26: 172). A later visit was more productive, as Hemsted told Sowerby: “I had forgot to mention Conferva Capillaris found last Week in Salt-Water Ditches near Wisbeach.” (Newmarket, 26 August 1797: Vol. 26: 174). Apart from the numerous bryophytes, Hemsted gave Skrimshire the following specimens: Arabis turrita “From Cambridge ... 1796.”; Groenlandia densa (as “Potamogeton densum”: conf. C. D. Preston) “gathered near Newmarket by Mr Hempsted ...”; and Mentha x smithiana (as “Mentha gentilis. M. rubra.a. Huds.”) “Sent to me by Mr. Hemsted, from Newmarket. I have not met with it growing about Wisbech. 1795.” William Skrimshire had a wide circle of botanical friends. They included a fellow surgeon John Pitchford (c. 1737-1803), a Roman Catholic who practised in Norwich (Bull 1999: 36; Petch & Swann 1968), one of the first British botanists to study Linnaeus’ works (Walker 1988). The two men first met late in 1795, as Thomas Woodward (1745—1820) related in a letter to James Edward Smith dated 20 November 1795; although Skrimshire is not mentioned by name the reference is clear: M' Pitchford has been at Wisbech to sell his estates, he met with a brother surgeon there who has found very good things in that country. Orobanche ramosa in plenty, as I always supposed likely from the soil on which I found it here, Dactylis stricta [= Spartina maritima], Riccia glauca, as he supposed but which I [? took] at first sight (no great proof of extraordinary sagacity) to be a much better thing — Riccia natans, also Ulva incrassata. ... [Ms in Archives, Linnean Society, London; Smith Correspondence 18: 210] This requires further comment. Orobanche ramosa is recorded in Skrimshire’s manuscript catalogue from Wisbech. Spartina maritima has never been recorded for Cambridgeshire and is now believed extinct in Norfolk (see Beckett et al. 1999): as this record is not localized, and there is no entry in Skrimshire’s catalogue, it is impossible to assign it to a vice-county. It is not improbable that S. maritima did grow in the lower reaches of the Nene in the 1790s. The comment by Woodward on Riccia natans (now Ricciocarpos natans), a liverwort, is confirmed by a specimen in Skrimshire’s moss herbarium (WBCH) labelled “found in a ditch at Outwell between the town and the Toll-Gate 1795”. There is also a specimen labelled Riccia glauca (= Riccia sorocarpa) that was “gathered by the side of some silt-pits, in a field opposite the Black-Boy-Low, Walsoken” (v.c. 28) but it is dated 1796, after Pitchford’s meeting with Skrimshire. In return, presumably, for specimens from Wisbech, Pitchford gave Skrimshire specimens of Arabidopsis thaliana (as “Arabis thaliana 8”) from ‘ta Wall at Norwich ... 1799.”; Monotropa hypopitys “gathered near Norwich ... 1797.°; Herminium monorchis and Orobanche elatior, both unlocalized. Pitchford was as interested in the mints as Skrimshire (see Dawson 1934: 74-75: Bull 1999). He was a keen field-botanist, credited with the discovery of Holosteum umbellatum (Bull 1999) and the earliest Norfolk record of Liparis loeselii (Bull 1999), and he made this comment about the contemporary problems of a field-botanist in a letter to Dawson Turner dated 15 July 1797 (Ms in Dawson Turner papers, Trinity College, Cambridge): ... but for Dianthus prolifer [= Petrorhagia prolifera; see Beckett et al. 1999: 82] — alas! What shall I say, but that it is lost & lost I fear for ever to this Country, the unfeeling farmers, those mortal enemies of botanists, having ploughed up the ground to the very edge so that I had WILLIAM SKRIMSHIRE (1766-1829) 35 some difficulty forcing my way along among the barley & weeds to the very end of the hedge row, as I was determined not to lose an inch of the precious ground where the dear creature used to grow. [Nelson 1997a, 1997b] Although Skrimshire did not live far from the open sea — and as we have noted the River Nene is tidal — in 1821 he acquired a series of specimens of coastal plants from F. Coates of Brancaster, including Salsola kali, Calystegia soldanella (as “Convolvulus Soldanella’”) and Honckenya peploides (as “Arenaria peploides”’). Mr Lilly Wigg (1749-1829), who was variously a shoemaker, schoolmaster and a clerk in the Yarmouth bank owned by Dawson Turner, was another of Skrimshire’s contacts. Wigg is recorded as having instructed Dawson Turner in algae (Petch & Swann 1968; Desmond 1994). One specimen from Wigg was Hieracium umbellatum: “This specimen was sent to me by Mr. Wigg, from Yarmouth. — 1795. I have not found it near Wisbech.” According to Petch & Swann (1968: 219) Wigg’s hawkweed, reported from Norfolk by Smith (1807: tab. 1771) was “gathered on the beautiful wooded hills at the back of Thorpe, Norwich”. The botanist and entomologist, Adrian Hardy Haworth, who was friendly with William and Fenwick Skrimshire, also donated herbarium specimens to William’s hortus siccus. Haworth was a succulent enthusiast and is commemorated in the southern African genus Haworthia (Aloaceae). There are several Haworth specimens in WBCH, including Scilla verna and S. autumnalis, Potentilla rupestris and Salix lapponum; whereas the squills could be native (the specimens are not localized), the rock cinquefoil and downy willow specimens were from plants cultivated in Haworth’s garden. Individuals represented by singletons are Mr Hull — perhaps the Manchester physician, John Hull (1761-1843). William Skrimshire was “indebted for this wild Specimen [of Polycarpon tetraphyllum| to Mr Hull, who gave it to Fenwick. 1797”. The Rev. John Burrall presented Skrimshire with a specimen of Dawson Turner’s discovery Tolypella glomerata (Desv.) Leonh. (as “Chara nidifica’’), from “splashes of Salt water at Cley on the Norfolk Coast’. There is a specimen of Silene uniflora (as “S. amoena”; conf. S. M. Walters 1999) “Given to me by Mr. Don, out of the Physic Garden. I have not found it about Wisbeach. 1795”. Almost certainly this came from the Botanic Garden in Cambridge, from its curator James Donn (1758-1813) (Walters 1981: Fig. 38, p. 45). Other evidence for contacts with fellow botanists is obtained from correspondence and publications. Not surprisingly, William Skrimshire knew the remarkable duo of Norfolk-based botanists, Dawson Turner and Dr (later Sir) James Edward Smith, and the naturalist and publisher James Sowerby. There is a letter from Skrimshire to Turner, dated 10 July 1797, among Turner’s voluminous correspondence (Trinity College, Cambridge). There is also a letter to Sowerby (Ms. in The Natural History Museum, London) about specimens Skrimshire sent for illustrating in Sowerby and Smith’s English botany (see e.g. Limonium bellidifolium above). Among these illustrations was Stratiotes aloides published in 1797. Although Sowerby drew the illustration (Sowerby & Smith 1797: tab. 379) from a living plant sent by Skrimshire, the accompanying text does not state where he collected it. However in the manuscript catalogue of Skrimshire herbarium, there is an undated entry for Stratiotes aloides from Bardolph Fen, West Norfolk, and this may well have been its source. JOHN ROSE WEATHERHEAD (C. 1787-1849) John Rose Weatherhead was Skrimshire’s apprentice from 1803 for five years. By 1817 he was practising as a surgeon in partnership with Skrimshire, and lived in The Crescent, Wisbech. He became a keen botanist and corresponded with William Jackson Hooker, Professor of Botany in the University of Glasgow (and from 1841, Director of the Royal Botanic Gardens, Kew) on both botanical and entomological subjects. William Skrimshire credited his pupil with “probably” discovering a new species of beetle which had been washed ashore in a flood at Wisbech in 1812: “This pretty little Insect, which I expect is new to Britain, was first discovered this summer at Wisbech, by my Pupil, Mr. J. R. Weatherhead, an ingenious and industrious Entomologist” (W. Skrimshire 1812: 318). Weatherhead’s own herbarium forms a substantial proportion of the collections in WBCH (Kent & Allen, 1964: p. 269 as Weatherhead, I. R.), but is not the principal subject of this paper. 36 G. CROMPTON AND E. C. NELSON However, a small number of specimens from him were evidently incorporated into William Skrimshire’s herbarium, and sheets annotated in Skrimshire’s distinctive writing (see Figs | and 3) include one of the hybrid Stachys x ambigua J. E. Sm. (then called “‘ambiguous woundwort’), the progeny of S. sylvatica and S. palustris. This had been discovered by William Borrer and W. J. Hooker in Orkney and on the mainland of northern Scotland in 1808. Weatherhead, independently, found it near Edinburgh in August 1809 and sent a specimen to Smith who mentioned this fact in the protologue (Sowerby & Smith 1809). The specimen that belonged to Skrimshire records other details of Weatherhead’s discovery, the annotation on the verso by Weatherhead himself being the more detailed: [recto; W. Skrimshire] Found by Mr. J. R. Weatherhead in Augt. 1809 at the foot of one of the Pentland Hills, in a moist boggy place about a Mile and a half from Hubbies How. He presented me with this specimen. 1810. [verso; J. R. Weatherhead] Stachys ambigua. “I found this plant in Augt. 1809 at the foot of one of the Pentland Hills in a moist & boggy place about a mile and a half from [Edinburgh — del.] Hubbies How, a specimen was sent to Dr Smith who considered it as new; it had however been found during the preceding summer in the orknies & some parts of the North of Scotland. Dr Smith has named it ambigua. It was growing inter-mixed with S. sylvatica, but no specimen of S. palustris could be found near it.” J. R. Weatherhead. The specimen from Skrimshire’s herbarium, and a second one, both contained in Weatherhead’ s hortus siccus (WBCH), must be isotypes, although their precise status remains to be determined. In September 1829, following William Skrimshire’s death, Weatherhead bought Skrimshire’s herbarium and with it the manuscript catalogue. As already mentioned, Weatherhead then marked with a tick all the specimens he had acquired. These specimens were presumably amalgamated with Weatherhead’s own extensive herbarium, and many may even have been remounted and re- labelled because most of the specimens marked by a tick are not present in the two sets of Weatherhead specimens now in WBCH. Weatherhead himself presented four volumes of herbarium specimens to the Wisbech Museum in 1835; there do not appear to have been any of William Skrimshire’s specimens in this collection. Thirteen years after Weatherhead’s own death in 1849, his widow presented seven further volumes of herbarium specimens. It was this collection tnat contained most of the Skrimshire specimens now in WBCH, including the first confirmed and the only record for Cambridgeshire of the rare Ruppia cirrhosa (as “Ruppia maritima’; det. C. D. Preston 1993) collected by Skrimshire: ... in a Salt water Ditch at Walton, near the River about 3 miles below Wisbech [v. c. 28]. I have also gathered it in a ditch which admits the tides, in a lane on the Right hand side of the Road from Wisbech to Tidd between the 3rd and 4th. milestones [v. c. 29]. It grows plentifully in the ditches at Long Sutton Wash-way, below Wisbech. 1795 [v. c. 53]. CONCLUSION William Skrimshire’s manuscript catalogue and the few of his specimens that are available for study have provided several previous authors with material for their respective Floras. The Cambridgeshire records extracted by F. H. Perring from these reliquiae are listed on the Cambridge Natural History Society's card index held in the herbarium, Department of Plant Sciences, Cambridge. The Norfolk specimens are listed in the late E. L. Swann’s card index, now with Mrs G. Beckett. With his varied interests in natural history, and his undoubted knowledge, William Skrimshire must have been a focus for the amateur naturalists living in the Wisbech neighbourhood in the late 18th and early 19th century. His manuscript catalogue and herbarium specimens represent an archive of considerable value, especially as a foundation for historical studies of the flora of Wisbech and the north-eastern sector of the old county of Cambridgeshire and the Isle of Ely, and the adjacent parts of West Norfolk. WILLIAM SKRIMSHIRE (1766-1829) 37 ACKNOWLEDGMENTS We are grateful to successive Curators of Wisbech and Fenland Museum, Mr D. Devenish and Dr R. Hanley, for allowing us to examine the herbarium collections, and we would like to express our particular thanks to Jill Rayment and Robert Bell, not only for all their assistance but also for finding so many sources of information and enlightenment about the Skrimshire family history. We are also indebted to Dr Michael Lewis of Oundle for information on Dr Fenwick Skrimshire’s achievements as a physician at Peterborough. G.C. also acknowledges the co-operation of Dr Peter Gillam who kindly showed her relics of his ancestor Dr George Skrimshire. Dr C. D. Preston (especially Potamogeton, Ruppia), Dr S. M. Walters (Silene) and Dr T. C. G. Rich (Cochlearia) checked the identities of some of Skrimshire’s specimens. Mrs Gill Gent kindly confirmed Skrimshire’s first records for Northamptonshire. We are grateful to the Wisbech Society for allowing us to publish excerpts from G.C.’s papers published in its annual reports (1994 and 1996). REFERENCES ARTHUR, J. L. (1985). Al among country museums. Friends of The Wisbech and Fenland Museum, Wisbech. BECKETT, G., BULL, A. & STEVENSON, R. (1999). A flora of Norfolk. Gillian Beckett, Stanhoe, Norfolk. BOULGER, G. S. & BRITTEN J. (1918). Joseph Andrews and his herbarium. Journal of botany 56: 257-261. BRIDSON, G. D. R., PHILLIPS, V. C. & HARVEY, A. P. (1980). Natural history manuscript resources in the British Isles. Mansell, London. BULL, A. (1999). The Norfolk botanists, in BECKETT, G., BULL, A. & STEVENSON, R. A flora of Norfolk. 39- 41. Gillian Beckett, Stanhoe, Norfolk. CROMPTON, G. (1994). William Skrimshire 1766-1829. The Wisbech Society annual report 55: 17-20. CROMPTON, G. (1996). Postscript to William Skrimshire. The Wisbech Society annual report 57: 6. DAWSON, W. R. (1934). Catalogue of the manuscripts in the library of The Linnean Society of London, part I. The Smith papers (The correspondence and miscellaneous papers of Sir James Edward Smith, M. D., F. R. S., First President of the Society). Linnean Society, London. DESMOND, R. G. C. (1994). Dictionary of British and Trish botanists and horticulturists. Taylor & Francis & The Natural History Museum, London. GRIGSON, G. (1955). The Englishman’s flora. Phoenix House, London. HAWORTH, A. (1803). Lepidoptera Britannica. London. HENREY, B. (1975). British botanical and horticultural literature before 1800. Oxford University Press, Oxford. KENT, D. H. & ALLEN, D. E. (1984). British and Irish herbaria. Botanical Society of the British Isles, London LACK, H. W. (1999). The Flora Graeca story. Sibthorp, Bauer and Hawkins in the Levant. Oxford University Press, Oxford. NELSON, E. C. (1997a). Botanists and the farming community. BSBI/ news 74: 31. NELSON, E. C. (1997b). Botanists and the farming community. BSBI news 75: 31. OSWALD, P. (1991). The Revd John Hemsted (1747?-1824). Nature in Cambridgeshire 33: 26-29. PERRING, F. H. (1956). An MS. in Wisbech Museum. Proceedings of the Botanical Society of the British Isles 2133: PERRING, F. H., SELL, P. D., & WALTERS, S. M. (1964). A Flora of Cambridgeshire. Cambridge University Press, Cambridge. PETCH, C. P. & SWANN, E. L. (1968). Flora of Norfolk. Jarrold & Sons, Norwich. RELHAN, R. (1785). Flora Cantabrigiensis, Pp. xxii, 490. Cambridge. RELHAN, R. (1802). Flora Cantabrigiensis, 2nd ed. Pp. xvi, 568.Cambridge. SKRIMSHIRE, F. (1798). Dissertatio ... de assimilatione ... . Edinburgh. SKRIMSHIRE, F. (1800). Information to parents, respecting the vaccine inoculation; or the inoculation for the cow-pox. Wisbech. SKRIMSHIRE, F. (1802). A series of popular chymical essays. London (2nd edition. London 1804). SKRIMSHIRE, F. (1805). A series of essays introductory to the study of natural history. 2 vols. London. SKRIMSHIRE, F. (1834a). Letters on the consumption of malt, addressed respectively to the farmer, labourer and labourer’s friend. J. Ridgeway & Son, London. SKRIMSHIRE, F. (1834b). Every poor man his own brewer; a letter addressed to the labourer of every class. J. Ridgeway & Son, London. SKRIMSHIRE, F. (1838). The village pastor’s surgical and medical guide; in letters from an old physician to a young clergyman his son, upon his entering upon the duties of a parish priest. London. 38 G. CROMPTON AND E. C. NELSON SKRIMSHIRE, T. (1812). On rearing insects. Transactions of the Entomological Society 1: 70-72. SKRIMSHIRE, W. jun. (1806). On the absorption of electric light by different bodies. Nicholson’s journal of natural philosophy 15: 281-283. SKRIMSHIRE, W. jun. (1807a). On the absorption of electric light by different bodies, and some of their habitudes with respect to electricity. Nicholson’s journal of natural philosophy 16: 101-107. SKRIMSHIRE, W. jun. (1807b). On the habitudes of saline bodies with regard to electricity. Nicholson’s journal of natural philosophy 17: 12-18. SKRIMSHIRE, W. jun. (1808a). On the phosphorescence of bodies, from the action of the electric explosion. Nicholson’s journal of natural philosophy 19: 153-156. SKRIMSHIRE, W. jun. (1808b). On the quantity of fecula in different varieties of the potato. Nicholson’s journal of natural philosophy 21: 71-76. SKRIMSHIRE, W. jun. (1808c). On the fecula of potatoes, and some other British vegetables. Nicholson’s journal of natural philosophy 21: 182-191. SKRIMSHIRE, W. jun. (1809). Account of a British vegetable product [Jris pseudacorus], that may be substituted for coffee. Nicholson’s journal of natural philosophy 22: 70-73. SKRIMSHIRE, W., jun. (1812). Remarks on some rare insects found during a flood at Wisbech. Transactions of the Entomological Society 1: 315-322. SKRIMSHIRE, W. (1816). Observations on the late excessive cold weather. Philosophical magazine 47: 182- 185. STACE, C. A. (1991). New flora of the British Isles. Cambridge University Press, Cambridge. TuTT, J. W. (1905). A natural history of the British butterflies. Vol.1, pp. 420-421. Elliot Stock, London. VICKERY, R. (1995). A dictionary of plant lore. Oxford University Press, Oxford. WALKER, M. (1998). Sir James Edward Smith M. D., F. R. S., P. L. S. 1759-1828 First President of the Linnean Society of London. Linnean Society, London. WALTERS, S. M. (1981). The shaping of Cambridge botany. Cambridge University Press, Cambridge. (Accepted June 1999) Watsonia 23: 39-46 (2000) 39 John Blackstone (1713-1753); a London apothecary and botanist of his time E. M. and C. BOWLT 7 Croft Gardens, Ruislip, Middlesex, HA4 SEY ABSTRACT John Blackstone was born, educated, apprenticed and practised as an apothecary in London during the first half of the 18th century. He made early contact with several distinguished persons, notably Sir Hans Sloane, to further his botanical interests. His passion for plants resulted in two small Floras at a time when few were published. One dealt with the plants around Harefield, Middlesex, in which he recorded 464 currently acceptable flowering plant species. Comparison with the plants recorded in that area during 1965-76 indicates that 26% of plants had become extinct in the intervening 240 years (18% if the additional surrounding 2 km is included) in contrast to an estimated 1% for Britain as a whole in the last 300 years. He is commemorated in having a plant and genus (Blackstonia perfoliata) named after him. KEYWORDS: early Floras, extinctions. INTRODUCTION “The period from Ray’s death in 1704 until the introduction of the Linnean system into Britain about 1760 was a lean one for the study of British plants” (Gilmour & Walters 1973). Interest in the subject was kept alive by a few, some only minor figures in the history of botany, who in their time made small but significant contributions to its study. Such a person was John Blackstone, who developed a tremendous enthusiasm for plants and whose two small Floras, Fasciculus Plantarum circa Harefield sponte nascentium (1737) and Specimen Botanicum quo Plantarum plurium rariorum Angliae indigenarum loci natales (1746), were among the few published during the period. Harefield, Middlesex, is remarkable in having had one of the earliest censuses of a human population in Britain in 1699 (Cuthbertson 1992) and one of the earliest catalogues of plants (John Blackstone 1737). Blackstone was a predecessor in the study of plants in Middlesex in whom Douglas Kent took a special interest. Kent’s (1949) paper stimulated our investigations into Blackstone’s life and work. Inevitably there is much common ground between the papers of Kent and ourselves, but corrections and new material are included here. In addition we have analysed Blackstone’s (1737) plant records for Harefield deriving statistically meaningful numbers for extinctions during the following years, and illustrating the continuing relevance of the work of early botanists such as Blackstone. EARLY LIFE John Blackstone was born in 1713. His birth on 22 September and baptism on the 25th, are recorded in the registers’ of the parish of St Leonard Eastcheap on the east side of the City of London. The church of St Leonard had been destroyed in the Great Fire of 1666 and not rebuilt and the parish joined with its neighbour St Benet Gracechurch, where he must actually have been christened. Blackstone was not born in humble circumstances. His family background, on both sides, whilst not providing great wealth, gave him education and opportunity to enter a profession which seems eventually to have provided sufficient income for him to pursue his passion for botany. It also seems to have given him influential contacts, which he early exploited to further his interest in plants. 40 E. M: AND C. BOWLTE APOTHECARIES IN THE FAMILY His father was Edward Blackstone, citizen and ironmonger’ of London, but his grandfather, another John Blackstone, was a successful apothecary living within the precinct of Pater Noster Row in St Faith’s parish in 1690°. Indeed the Gentleman’s Magazine’ described him as an eminent apothecary of Newgate Street, a friend and associate of Sir Hans Sloane and a liberal benefactor of St Bartholomew’s Hospital. He was Warden of the Apothecaries’ Company 1712-13. He had three sons as well as Edward: William, who became Keeper of Wychwood Forest, Oxfordshire; Charles, a silk mercer, citizen and bowyer of London; and Richard of whom nothing is known. The elder John Blackstone apparently retired from business in 1713 “to enjoy the fruits of a competant fortune’ and was living as a gentleman in Wandsworth, Surrey, at the time of his death in 1715. He left £1200 to Edward to set up in business as an ironmonger. Just what Edward had been doing until then (he was already married with a child) is unclear. His name appears neither in the list of freemen of the Ironmongers’ Company® nor in the lists of apprentices. However, an Edward Blackstone, son of John Blackstone, pharmacist, was apprenticed to Ambrose Crowley, a draper, in 1703, freed in 1711 and a member of the Livery of the Drapers’ Company 1719-27’. From these dates this could be Edward, the father of John Blackstone. If so, in preferring ironmongery to drapery, he would have been able to set up in business easily enough, if he had capital, at a period when the companies were not exercising so strict a control over the traders in the city as they had in earlier years. William, a brother of John Blackstone the elder, was also an apothecary in London and his name appeared among those who subscribed towards the revival and improvement of the Chelsea Physic Garden about 1700. This side of the family had property at Chipping Walden, Essex*. A William Blackstone (apprenticed as an apothecary to a John Blackstone in 1696’) may be Edward’s elder brother, who later became keeper of Wychwood Forest. THE HAREFIELD CONNECTION Edward Blackstone married Sarah, the fifth child of Francis Ashby Esq. (1671-1743) of Breakspears, Harefield, in Middlesex, one of the lesser landed gentry with a family coat of arms. Harefield played a large part in the life of John Blackstone. He obviously got to know it well and had a great affection for the area. His parents were married at Harefield church on 28 February 1712'°. After John’s birth in September 1713 two brothers followed: Edward, born and baptized (perhaps he was delicate) on the 29 December 1717. and Francis on 27 March 1723''. All the children’s baptisms were at St Leonard’s Eastcheap. However, when the father died, in 1730, he was buried at Harefield, and John’s mother and younger brothers moved to Breakspears to live with her parents. John, by then apprenticed'” to Thomas Bearcroft, apothecary, was a frequent visitor to that new home. It is possible that the family had moved to Harefield before Edward Blackstone’s death, as the record of John’s apprenticeship’’, dating from Sth August 1729, refers to him as the son of Edward Blackstone, “late of London, ironmonger’’. EDUCATION Whether John Blackstone spent most of his childhood in the city or stayed with relatives elsewhere for long periods, perhaps with his grandparents at Harefield or with his paternal uncle, the Keeper of Wychwood Forest in Oxfordshire, one can only guess. His early education is also unknown, but it was sufficiently good for him to enter Merchant Taylors’ School, then situated in Suffolk Lane not far from London Bridge, in 1727, at the age of 14'*. He must already have been reasonably proficient in Latin, for when he left only two years later’’ he knew sufficient, not only for his subsequent profession as an apothecary, but also to write his subsequent Floras in that language. A contemporary at Merchant Taylors’ School was William Watson who also became an apothecary, and ultimately a physician, with a strong interest in plants. Watson similarly entered the school in 1727, but remained a year longer. If the two were not friends at school, they certainly were afterwards, for Watson supplied Blackstone with herbarium material and records for his Specimen Botanicum. He was a man of wide scientific interests, conducting electrical experiments JOHN BLACKSTONE | and contributing to the Philosophical Transactions of the Royal Society, becoming an F.R.S. in 1741, and eventually being knighted in 1786, the year before he died. APPRENTICESHIP On leaving Merchant Taylors’ John was “apprenticed and bound to Mr. Thomas Bearcroft for eight years from this day (5 August 1729) in the sum of £100”'°. That Bearcroft was an apothecary was probably significant in introducing him to the serious study of plants, herbs having long played a major part in an apothecary’s pharmacopoeia. It seems that not many grew or gathered herbs by this period (as Gerard had done in his gardens at Holborn, for instance), but they were certainly expected to be able to identify species. HERBARIZINGS Simpling expeditions for apothecaries are recorded from 1620. Led by such eminent persons as Thomas Johnson, the parties met early in the morning at St Paul’s and visited rural areas like Hampstead Heath or Greenwich and occasionally, as in 1634, went as far afield as Bath, Bristol and the Isle of Wight. (Hunting 1998). After the founding of the Chelsea Botanic Garden in 1675, the Society of Apothecaries, “to excite a taste for botany among apprentices” according to Pulteney (Pulteney 1790), encouraged private herbarizings beginning on the second Tuesday in April, and held monthly on the same day until September, in the countryside in the immediate neighbourhood of London, to which Apothecaries could send their apprentices. These seem to have been in the nature of botanical field meetings with a leader from the Society until 1724 when a Botanical Demonstrator took over the job (Hunting 1998). Isaac Rand, an apothecary and botanist, filled that post 1724—43'’ There had been complaints about disorderly behaviour among the apprentices at the herbarizings and henceforth they had to enter their own and their master’s names on a list kept by the leader and bring a letter from their masters permitting their attendance. There was to be no question of the apprentices treating the herbarizings as holidays. All had to wait to the end to attend the showing of the plants and from thence return directly home without loss of time'®. There is no actual record of John Blackstone having attended, but it seems highly likely that he did so. William Watson is known to have done so because “during his apprenticeship he gained the honorary premium (a bound copy of Ray’s Synopsis) given annually by the Apothecaries’ Company, to such young men as exhibit a superiority in the knowledge of plants 1 in these excursions made by the Demonstrator of the Chelsea Garden” (Pulteney 1790). There is in the British Library a copy of Ray’s Synopsis” (bound in with other of Ray’s works) with Blackstone’s autograph and the date 1736. In 1735 Blackstone’s master died and he was ‘turned over’ to Mr Richard Stevenson for the remainder of his apprenticeship. This concluded on 7th March 1737 “when having served his full term he was sworn and made free””” INFLUENTIAL CONTACTS Before he was out of his apprenticeship the young Blackstone had already been writing to several eminent persons, including his grandfather’s friend Sir Hans Sloane, about his botanisings and had already written a book on the plants of Harefield. His grandfather, Francis Ashby, apparently took an interest in plants and where they grew’', and may well have influenced the young Blackstone’s burgeoning interest in botany and the local flora. By the end of 1736 Blackstone had been commissioned to collect plants for the Apothecaries’ garden at Chelsea as well as the Oxford Physic garden. He mentions this in a letter written on 11 December 1736 to Dr Richard Richardson, a well-to-do physician and well-known botanist of Bierley, Yorkshire. “By good fortune I have discovered the Dentaria baccifera (Cardamine bulbifera) to be a native of this island. The place where it grows is in a large wood near a village call’d Harefield, about eighteen miles from London, where it grows in such plenty, that some acres of ground are cover’d with it. [Old Park Wood, where it still grows]. There are several other rare plants to be found thereabouts, of which a catalogue is now ready for the press... I have communicated this plant and my intentions, to Sir Hans Sloane, Dr. Dillenius (from whom I had a most obliging letter) Mr Rand, and some others... I have orders to get specimens for Chelsea and Oxford Gardens; and wou 'd you give me directions, I shu’d be very glad to send some for your garden, if ’tis not there already”. 42 BM. AND:C.BOWLT A week later he had already received a reply and wrote back with some confidence, giving his book a puff at the same time: “Strand, Dec.18th 1736 I beg leave to return you my humble and hearty thanks for the favour of your obliging letter... I have for these last three years been employ’d in making a collection of the native plants; and having an opportunity of going to see my friends pretty often, I made it my business to see as many of the adjacent places my time would permit, and to collect such plants as offer’d themselves in the course of my walks, without ever intending to publish anything on this subject. But, being detained last Summer by a long illness, near four months on the spot, I found so many rare plants that I thought it worthwhile to make a catalogue of them, and show it to some skillful persons in that science for their approbation, which succeeded to my desire; and I only beg leave to let you know, that the plants there mentioned were gathered almost solely by myself, as also that the catalogue is not general; being only intended as an essay for a more particular search thereabouts. ... I purpose to print my work by subscription, tho’ the price of it cannot exceed five shillings; and I desire no money till the book is deliver’d. I should take it as a great favour if you cou’d procure me a specimen of Epimedium and the Christophoriana, which would be highly acceptable. .. I expect to go to Harefield next week, and will search for some roots of Dentaria, if you please to order it, and will (with God’s leave) send you some in perfection about the latter end of April, when the bulbs will be ready to plant. Be pleased to accept of the compliments due to the ensuing season””>. The catalogue referred to was published in 1737 as Fasciculus Plantarum circa Harefield sponte nascentium (A Little Bundle of Plants freely growing around Harefield). His correspondence with Dr Dillenius, Professor of Botany at Oxford, continued. A letter to him from Blackstone, dated 1744, is laid in with a specimen of Epimedium in the Sherard Herbarium (in the Dept of Plant Sciences, Oxford): “I could not fail sending you a quotation from a letter of that ornament to Botany, Dr Richardson,...Epimedium is a native of England. It grows in the woods near Bingley...and not sparingly”. (Epimedium is not now, of course, considered native, but these were the frontier days of botany with much still undiscovered.) THE BOTANIST With the large number of modern detailed books on plant identification available it is easy to forget the difficulties in identifying plants in the 18th century. The relatively few books that had been published, often long previously, seldom agreed on names and descriptions were not very distinctive. Blackstone included several author names for each plant but in his Fasciculus generally put the name given in Gerard’s Herbal (as revised by Johnson in 1637) first. The Herbal is a hefty tome of more than 1600 pages and must have been difficult to procure and expensive to buy. It is not known whether he possessed a copy or used the Apothecaries’ library at Chelsea. Gerard’s book has woodcut illustrations to the species, which in most cases are recognisable and hugely bolster the rather inadequate descriptions in English. Building up a herbarium, as Blackstone did, would clearly have supplemented the rather limited Floras, and the exchange of specimens with other botanists must have been significant in getting general agreement on the identification of plants. Blackstone’s Fasciculus is an early example of a local Flora. It was a remarkable achievement at that period for a 24 year-old just out of his apprenticeship. It contains a list of 523 plants in alphabetical order of their pre-Linnean Latin names. This is followed by their English names and in many cases the locality (in English), with comments and time of flowering (in Latin). The book ends with an Appendix containing some short notes relating to Harefield. To put this into perspective, without entering into the question of species, it has been estimated that, at the time of Ray’s death in 1704, the total number of plants recorded for the whole of Britain was about 970 (Gilmour & Walters 1954). For the whole of Middlesex, Trimen & Dyer (1869) recorded 859 species, and Kent (1975) recorded 1109 - this last including many recent aliens. A few of Blackstone’s records are from places outside the immediate area of Harefield, such as Iver and Chorleywood, and even as far afield as Windsor and Hampton Court (i.e. up to 15 miles distant). It also includes a few fungi, liverwort and lichen records, but they are not always identifiable. JOHN BLACKSTONE 43 Omitting these, and also what now appear to have been varieties, Blackstone published a list of 464 vascular plants species growing in the parish of Harefield, an area of about 1870 hectares (4621 acres) in the north-west corner of Middlesex. There are, however, some curious apparent omissions of plants that surely must have been common and noticeable in his day. It is difficult to believe, for instance, that Gorse was not present in the 18th century. He does not seem to have deliberately omitted very common plants because he lists some, such as Daisy and Dandelion. In spite of this, including a few extra species included in his second book, the final total of 464 flowering plants recorded by Blackstone must have represented a majority of those growing around Harefield in 1737. It is, therefore, statistically meaningful to examine the extinctions that have occurred in this one small area since then, by comparison with the distribution maps based on the London Natural History Society’s botanical survey 1965-1976, published in the Flora of the London Area (Burton 1983). The parish of Harefield is contained within a block of seven 2 x 2 km squares of the Flora maps. This area has been assumed to be close to that used by Blackstone which was probably not too precise. By the time of the L.N.H.S. survey, 122 flowering plant species recorded by Blackstone were no longer present. That is, 26% of his plants had become extinct in about 240 years. If the area is expanded to include all 2 x 2 km squares immediately surrounding the block of seven (i.e. an additional 16 squares) the number of extinctions is reduced to 85 species, i.e. 18%. These must rank as some of the best estimates over one of the longest periods of time of extinctions from the flora of a relatively small area. For the British Isles as a whole it is estimated that about 15 species (i.e. about 1% of the total) of flowering plants and ferns have been lost in the last 300 years (Rackham 1986). The greater percentage loss in the smaller area of Harefield illustrates the vulnerability of small populations. Inspection of the 85 lost species indicates that most were from wetlands (25%), followed by those from heathland (15%), corn/arable fields (13%) and woods and thickets (13%). These findings agree with those of Kent (1975) for the extinctions in the whole of Middlesex between 1863 and 1968 and confirms many general observations. Some of the plant species no longer growing in Harefield are only local extinctions, but others are no longer to be found in the London area as a whole. For instance, Arnoseris minima has not been seen within two miles of St Paul’s since 1910. Anagallis tenella was last recorded in 1951. Parnassia palustris and Pedicularis palustris are also now extinct in the London area. Among the orchids recorded by Blackstone at Harefield, Orchis militaris seemed to be extinct in Britain by about 1914. Since its rediscovery in the Chilterns in 1947 it remains one of Britain’s rarest orchids. As a Londoner botanising in Middlesex Blackstone was rather fortunate that his grandfather lived in the parish of Harefield. In the extreme north-western corner of the county, the topography and the outcropping geological strata gave rise to a wide range of soils and habitats within a comparatively small area. Blackstone was aware of this, for in an Appendix to his Fasciculus he wrote “The Soil, within this Compass, is various; that of the South and East Parts being a strong Clay, very productive of Elm and Oak, which thrive well here. The North and West Parts, bordering on Hertfordshire and Bucks, is generally a dry Chalk and stony Land, and The great Variety of the Soils here, produces all Sorts of Vegetables, which our Catalogus of such Plants as grow spontaneously in this Neighbourhood, sufficiently shews.” At Harefield the northern edge of the London Basin rises to the surface and is cut into by the River Colne, revealing, under the London Clay, the Reading Beds of clays and sands and the Upper Chalk. The Colne Valley was an area of black alluvium wetlands. Unfortunately, Harefield’s varied geology led to its rape for raw materials. Chalk pits noted by Blackstone continue to the present day. Brickworks have used the clays, sandpits have been dug and, in this century, the vast gravel pits have destroyed the wetlands. To add insult to injury many of the diggings have been filled with rubbish and rubble giving the coup de grace to any plants with the temerity to try and hang on. These activities, along with building, are the principal reasons for the disappearance of species, but there are others. Trimen & Dyer (1869) reproduce a quotation of Peter Collinson’s (1694— 1768) which they say helps to account for the extinction of orchids at Harefield: “There is one Miles, a parson of Cowley, near Uxbridge, who is orchis mad, takes all up, leaves none to seed so extirpates all wherever he comes which is cruel, and deserves chastisement.” In connection with Orchis militaris, Kent (1975) said I “feel that unrestricted collecting by nineteenth-century botanists contributed to the extinction of this rare and beautiful species in the vice-county.” 44 E. M. AND C. BOWLT THE APOTHECARY Having completed his apprenticeship in March 1737, John Blackstone seems to have been in no hurry to set up in business, perhaps because of the ill health frequently mentioned in his letters, or because the collecting and study of plants was proving more alluring. On 8 April 1737, he wrote to Sir Hans Sloane: “The air proves serviceable to me in general, tho’ there are still some symptoms remaining which a perfect state of health ought to be without, particularly the swelling of my legs after walking, and the heat of my hands and feet towards night” His contemporaries, like Dr Richardson, having seen the Fasciculus, certainly encouraged him to pursue his natural history studies and he confessed to Sloane on 29 September 1737 that “the study grows more and more agreeable to me”. Sir Hans had probably been encouraging him to start in business, for Blackstone wrote to him on 26th October 1737, just as he was about to visit Pembroke College, Oxford and relatives at Charlbury: “I hope, Sir, this absence will be no prejudice to your esteem for me, but to engage in business without any foundation would not be deem’d a prudent action, therefore as I am obliged to be depending, I must wait till those I depend on, are pleased to make a beginning for me”*°. In December he said that if his relatives had been willing to fix him in business he would not have left London. In fact it was almost another year before he did set up in business at the sign of the Griffin, Salisbury Court, near St Bride’s, just off Fleet Street, in September 1738. He was admitted to the Society of Apothecaries the same year, and married a widow, a Mrs Webb, in 1739, but she died, probably in 17407’. A William Blackstone, an infant, was buried at St Bride’s in April 1741 and an Elizabeth Blackstone on 11 April 1742, but they cannot be proved to be John Blackstone’s wife and child. We know little about Blackstone’s professional life. He became a subscriber to the Laboratory Stock of the Society of Apothecaries in January 1740, investing £107, but seems to have found difficulty in establishing his business. He complained of trade “being very dull and money very hard to get and seemed to be asking Sir Hans Sloane for a loan in September 1740, when he dispatched a note with a postscript saying that his man would wait on him about noon to know his pleasure~’. The result is unknown, but he felt, himself 1 in a position to take on an apprentice, John Thomas Vernon of Malpas, Cheshire, in 1741°° He married again in October 1742. His second wife was Mary Abbot, the daughter of Richard Abbot, es stainer, and the wedding took place at St Sepulchre’s, Newgate Street on 23 October LATER YEARS The next few years were clouded with tragedy, for his mother and his grandfather both died in April 1743, being buried at Harefield on the same day. The following year his wife gave birth to a daughter, Maria, only for her to die in infancy. On a happier note a second daughter, Sophia, was born in 1748 (and lived until 1827) and his second book, Specimen Botanicum, had been published in 1746. SPECIMEN BOTANICUM The layout of this small 106-page book is the same as for the Fasciculus, but it lists only 367 plants, including 19 non-vascular cryptogams, and their sites in various places in England. A number of records were supplied by other botanists, which he acknowledges - sometimes in a quaint manner, such as, “found by that accurate Botanist, Dr Richardson”, and “with Dr Jarvis, an eminent surgeon and curious Botanist”. The catalogue includes 99 vascular plants from Harefield. All but nine of these had appeared in his earlier Fasciculus. As in his earlier work, most of the non-vascular cryptogam identifications are dubious, presumably a reflection of the state of their taxonomy at that period. The list is of what were considered some of the more interesting plants growing in England. It is in no sense comprehensive in its geographical cover, but was thought worth publishing at a time when few botanical books were being produced. It is interesting that even in Blackstone’s time the loss of certain species from some sites was being noted. He records that Galium cruciata, Crosswort “is JOHN BLACKSTONE 45 lately lost in Hampstead Church-Yard, where it grew in Johnson and Parkinson’s time”. Again of Cypripedium calceolus, Lady’s Slipper, “found by Mr Thornbeck, Surgeon, at Ingleton, who likewise informs me that it has been lost in Helk’s Wood for some years.” Financially, life was probably easier for Blackstone in the 1740s. His wife may have had some money to begin with and she certainly inherited £500 from her father, Richard Abbot, who died in 1746. John inherited £30 from his mother*”, as well as a three-acre field called Maudemeade and the four-acre Further Maudefield, running along the side of Bayhurst Wood, Harefield. He was able to mortgage the mead in 1746 to raise £105 from Pierce Dodd of Red Lyon Square, a doctor of physic’. Did he need the money to publish his book, the Specimen Botanicum? He sold both fields to Anne Cane of Shiplake in 1751**. The fields are famous in botanical history, because it was here that he noted in the Fasciculus the Snake’s Head Fritillary growing “...1n Maud Fields observed above forty years ago by Mr Ashby of Breakspears.” This was its first record as a British plant. In addition, overall Blackstone provided at least 103 first records for Middlesex (v.c. 21). At this same period he was playing a role in the affairs of the Society of Apothecaries, being on the Garden Committee of the Chelsea Physic Garden from 1746 until 1752*, when his health began to fail. It is notable that it was to the botanical activities of the Society, rather than the trading side, that he devoted his time and interest. John Blackstone died on 11 March 1753, not yet 40 years old. He was buried at Harefield a week later on the 18th March*’. He left directions in his will dated | February 1753 “to be buried in the cheapest and most private manner possible...” To his brother, Francis he bequeathed his cornelian seal with two heads engraved on it, the silver can with his grandfather’s coat-of-arms and his mortuary knife. He left his largest punch bowl with a silver ladle and his seal engraved with a man at work to Mr John Abbot, his brother-in-law, “as a mark of my gratitude for many favours” .Everything else was left to his dear wife, Mary’ ’. There is a collection of 44 plants (including 14 mosses and lichens) in the Sloane Herbarium, but in 1947 three volumes of Blackstone’s plants, labelled by him, were discovered at Ripon Museum, Yorkshire. Not all of the plants were localised and only a few were from Harefield. The collection was transferred to the Natural History Museum, where approximately 360 specimens were laid into the general herbarium. BLACKSTONIA Curiously, Blackstone did not record the plant named after him closer to Harefield than Gerrards Cross, although we have photographed it at Harefield on some of the remaining chalkland. Pulteney (1790) noted that when Hudson published his Flora Anglica (1762) and introduced the Linnaean system into English botany, “he separated the Yellow Centaury from the Gentians and gave it the name Blackstonia; which distinction Linnaeus confirmed in the Systema of 1767, but then changed the name to Chlora, an appelation it had received from Rencaulme, in his Specimen Historias Plantarum, published in 1611. It should seem that the discovery of the true place of this plant in the system entitled Hudson to the dispensation of the name, or at least that Blackstone should have been perpetuated in the trivial epithet”. Hudson and Pulteney would be pleased to know that the plant is once more called Blackstonia perfoliata, and it seems a fitting tribute to John Blackstone’s botanical studies, which clearly gained the approval of his contemporaries. REFERENCES BLACKSTONE, J. (1737). Fasciculus plantarum circa Harefield sponte nascentium. BLACKSTONE, J. (1746). Specimen botanicum quo plantarum rarorium Angliae indigenarum loci natales. BurRTON, R. M. (1983). Flora of the London area. London Natural History Society, London. CUTHBERTSON, E. (1992) Gregory King’s Harefield. Hillingdon Borough Library, London. GERARD, J. (1633). The herbal or a general historie of plants. Enlarged and amended by Thomas Johnson. GILMOUR, J. & WALTERS, M. (1973). Wild flowers. Collins, London. HubDsON, W. (1762). Flora Anglica. London. HUNTING, P. (1998). History of the Society of Apothecaries. Society of Apothecaries, London. KENT, D. H. (1949). John Blackstone, Apothecary and Botanist (1712-53). Watsonia 1: 141-148. KENT, D. H. (1975). A historical Flora of Middlesex. Ray Society, London. PULTENEY, R. (1790). Historical and biographical sketches in the progress of botany in England from its origin to the introduction of the Linnean System. London. 46 E. M. AND C. BOWLT RACKHAM, O. (1980). Ancient woodland. Edward Arnold, London. TRIMEN, H. & DYER, W. T. T. (1869). Flora of Middlesex. Hardwick, London. TURNER, D., ed. (1835). Literary and scientific correspondence of Richard Richardson, MD, FRS of Bierley, Yorkshire, pp. 351-355. NOTES Guildhall MS 17607: parish registers St Leonard’s East Cheap Edward Blackstone is described as an ironmonger in the baptismal register of St Leonard’s, East Cheap - Guildhall MS 17607 and in Chancery Proceedings - PRO C 11 1976 3, but does not appear in lists of apprentices or freemen of the Ironmongers’ Company. D. Kent’s Papers: Poll Tax Assessments 1678 and 1690 cxciii. 390 Gentleman’s Magazine: 1827 11 Vol 97 Ibid Guildhall MS 16978 D. Kent’s Notes PRO C11 1584. 7 Guildhall MS 8207: Society of Apothecaries Bindings Book 1694-1836 Harefield Parish Registers transcript: private collection. Guildhall MS 1760718/02/93: parish registers St Leonard’s East Cheap. Guildhall MS 8207: Society of Apothecaries Bindings Book 1694—1836. Guildhall MS 8207: Society of Apothecaries Bindings Book 1694-1836 Guildhall Library: Merchant Taylors School Registers 1561-1934. Guildhall Library: Ibid. Guildhall MS 8207: Society of Apothecaries Bindings Book 1694-1836. Guildhall MS 8228: Garden Committee Minutes 1731-70. Ibid. Brit Lib: 969 f 18 (2). Guildhall MS 8206/2: Society of Apothecaries Freedom and Admissions Book. John Blackstone: Fasciculus plantarum circa Harefield sponte nascentium 1737 Dawson Turner ed: Literary and Scientific Correspondence of Richard Richardson, MD, FRS of Bierley, Yorkshire, pp 351-5 Ibid Brit Lib: Sloane MS 4055 f89 Ibid 4055 ff203-4 Ibid 4055 ff223-4 Collectanea Topographica et Genealogica Vol 5, 1838. Guildhall MS 8226: Society of Apothecaries List of subscribers to Laboratory Stock. Brit Lib: Sloane MS 4056 £305 Guildhall MS 8207: Bindings Book 1694-1836. Guildhall MS 7220/1: parish registers St Sepulchre’ s. PRO: PCC Prob 11 727 London Metropolitan Archive: Acc 1085/M41 LMA: Acc 1085/M42 Guildhall MS 8228/1: Society of Apothecaries Garden Committee Minutes 1731-70. Harefield Parish Registers transcript: private collection. PRO Prob 11 800 71 . (Accepted July 1999) Watsonia 23: 47-58 (2000) AT A history, mainly nomenclatural, of St Dabeoc’s Heath E. C. NELSON* Tippitiwitchet Cottage, Hall Road, Outwell, Wisbech PE14 SPE, Norfolk, UK. ABSTRACT St Dabeoc’s heath was first gathered in Cantabria (northern Spain) by J. P. de Tournefort before 1694, and subsequently during 1700 by Edward Lhuyd in western Ireland. The convoluted history of the Latin names, both pre- and post-Linnaean, for St Dabeoc’s heath is reviewed. Linnaeus was the first to employ, but not validly publish, a binomial (Erica cantabrica) in the second edition (1759) of his Flora Anglica. William Hudson adopted the same specific epithet, but placed St Dabeoc’s heath in Vaccinium; Hudson’s binomial V. cantabricum was validly published and is the basionym for the current name, Daboecia cantabrica (Huds.) K. Koch. The orthography of the generic name, first printed but not validly published by Martyn in 1807, has been a contentious matter for over a century but as nomenclatural stability is desirable, the “corrupt” spelling prevails. The species name is typified, the lectotype being the illustration published by Petiver in 1704. KEYWORDsS: Daboecia cantabrica, Ericaceae, Linnaeus, typification. INTRODUCTION St Dabeoc’s heath is familiar to Irish, French, Spanish and Portuguese field-botanists, and to gardeners in many other places. It belongs to a distinctive genus of the Ericaceae named Daboecia, and that name is derived (“corruptly” to quote Smith 1791) from the vernacular Irish name for this elegant shrub. St Dabeoc’s heath, Daboecia cantabrica (Huds.) K. Koch, inhabits the milder, oceanic regions of western Ireland (counties Galway and Mayo; see Webb & Scannell 1983), western and south- western France, northern Spain (Pyrenees to Galicia) and north-western Portugal (see maps in Woodell 1958; Dupont 1962; Small & Small 1998). There is an isolated population on the Azores, and this is treated either as a discrete species, D. azorica Tutin & E. F. Warb. (Tutin & Warburg 1932; Sealy 1949; McClintock 1969), or as a geographically isolated subspecies of St Dabeoc’s heath, D. cantabrica subsp. azorica (Tutin & E. F. Warb.) D. C. McClint. (McClintock 1989). The purpose of this paper is to describe the discovery and to elucidate the nomenclatural history of Daboecia. During the three centuries which have passed since it was first brought to the attention of botanists, St Dabeoc’s heath has been assigned to various genera — Erica by Linnaeus (1754), and all his predecessors back to Tournefort (1694), and by such of his successors as Thunberg (1785); Vaccinium only by Hudson (1762); Andromeda by Linnaeus (1767, 1770), and by Martyn (1807) who was the first author to list the generic name Daboecia; Boretta by Necker (1791) and some subsequent authors; Menziesia by Salisbury (1781), Jussieu (1802) and de Candolle (1805); Bryanthus only by Merino (1906); and Daboecia (or Dabeocia) by Don (1834), Koch (1872) and most subsequent authorities. As just indicated, there is a second matter of disagreement. The orthography of the generic name is open to dispute (see below). Martyn (1807) and Don (1834) used Daboecia, whereas Dabeocia was the arguably correct orthography (e.g. Babington 1888; Praeger 1925) adopted by Koch (1872) and most authors until the early 1950s (e.g. Moore & More 1866; Colgan & Scully 1898; Webb 1943; Gilbert-Carter 1950; Nelson 1984). *Address for correspondence: E-mail: tippitiwitchet @ zetnet.co.uk 48 E. C. NELSON DISCOVERY IN CANTABRIA AND IRELAND The first mention of a heather from Cantabria (northern Spain) with large flowers and myrtle-like leaves that were white underneath,’ referable to Erica even in pre-Linnaean times, appeared in Joseph Pitton de Tournefort’s Elemens de botanique (1694: 475). The source of Tournefort’s information is not exactly recorded; perhaps he gathered specimens during his exploration of the Pyrenees in 1681, but this is unlikely given his use of the geographical epithet cantabrica. Tournefort distributed specimens of his Cantabrian heath to fellow botanists including William Sherard (see below) although no specimen of the heather collected by Tournefort can now be identified in the Sherardian Herbarium (OXF; S. Marner pers. comm. 1995). In the spring and summer of 1700, the Welsh naturalist Edward Lhuyd (1660-1709; olim Llwyd or Lhwyd) travelled to western Ireland where he collected several hitherto unknown plants including a heath with large purple flowers (Gunther 1945; Mitchell 1975; Nelson & Walsh 1995). Lhuyd brought specimens to Britain and distributed duplicates to several botanical friends including the Revd John Ray, James Petiver and William Sherard. In correspondence written during the summer of 1700, Lhuyd also described the heath (see Mitchell 1975; Nelson 1978). He wrote to Dr Tancred Robinson, among others, about various aspect of Ireland’s natural history; this particular letter was published in 1712, three years after Lhuyd’s death: In most of the Mountains of Galloway and Mayo grows an elegant sort of Heath, bearing large Thyme-leaves, a Spike of fair purple Flowers like some Campanula, and viscous Stalks. I know not whether it be any thing related to the Cisti Ladanifere. [Lhwyd 1712: 525]. To Dr Richard Richardson of North Brierly, near Bradford, Yorkshire, he wrote (see Owen 1922; Dandy 1958): In the moors of ye County of Mayo & Galloway grows a very elegant sort of Heath which so common that ye people have given it ye name of Frych Dabedg 1. Erica [Sti.] Dabeoci & sometimes ye women carry sprigs of it about them as a Preservative against Incontinency. I calld it Erica maxima viscosa, rubra; Rosmarini foliis brevioribus flosculis Campanula minoris ... . That same year, 1700, Tournefort, Professor of Botany in the Royal Garden in Paris, published Institutiones rei herbaria, essentially a Latin edition of Elemens de botanique, repeating the phrase~ that he had first used in 1694 — he did not provide any new information about his knowledge of this plant (Tournefort 1700: 603). Meanwhile, Lhuyd’s specimens provided English authors with material for their publications. The first to notice it was John Ray in Historia plantarum (1704: Ill, Book XXX: 98). The particular volume was issued in the summer of 1704 — on 8 June 1704 Ray told Sir Hans Sloane that the third volume “of my History of Plants is now finished at the press” (quoted by Raven 1950: 300). There is a further reference* to the Irish heath within a section comprising lists of plants that Ray had received from Petiver (Ray 1704: II, Appendix, 244). The most remarkable thing about Ray’s contribution is that he correctly equated Tournefort’s Cantabrian heather with Lhuyd’s Irish one. Was this accurate identification just luck? That question is answered by a letter to Lhuyd from William Sherard written on 29 May 1701 from “Badmington” (Ms Ash. 1817a, original in Bodleian Library, Oxford; Gunther 1945). Sherard told Lhuyd: . As to ye curious plants you have innrich’d me wth I can say little, however shall venture to make some conjectures or queries abt them, since ye are pleas’d to comand it of me. I find (as Mr. Ray complains to me abt my own) ‘tis hard to judg of plants by dry’d specimens, especially where they are not perfect & well preserv’d, wch is not always practicable, especially in travelling. In the accompanying list, which clearly included Lhuyd’s Irish specimens, number 4 was glossed “T had this (or some very so like it, wch upon comparing ye specimens I shall easily judg) from Dr. Tournefort, by ye name of, Erica Cantabrica, fl. max. fol. myrti subtus incanis. Inst. r. herb.” In other words, it was Sherard who identified Lhuyd’s gathering of St Dabeoc’s heath and equated it HISTORY OF ST DABEOC’S HEATH 49 with Tournefort’s specimen. Lhuyd then seems to have labelled his specimens with Tournefort’s name, including whatever material went to Ray, although from a letter written by Ray, dated 11 June 1701 (Gunther 1928: 280-281), it is evident that Lhuyd also used the name Erica S$. Dabeoci: “The plant you intitle Erica S. Dabeoci’, responded Ray, “I am in some doubt whether it be a genuine species of Erica, the flower falls away, & ye fruit seeming to be different.” Shortly after Ray’s Historia plantarum was published in the summer of 1704, if we may judge correctly from his own phrasing, Petiver published an engraving (Fig. 1, p. 53) of Erica HIBERNICA ... S. Dabeoci in the third decade of his part-work Gazophylacium nature & artis (Petiver 1704: plate 27), a work described as a medley of text and illustrations depicting plants, shells, insects, birds and other animals. The unambiguous reference to Ray’s Historia plantarum in the accompanying text’ indicates that the third decade in which the description and illustration of St Dabeoc’s heath were issued was published late in 1704. Jacob Dillenius, editor of the third edition of Ray’s Synopsis stirpium Britannicarum, included St Dabeoc’s heath as the sixth member of his genus Erica, employing the long, elegantly descriptive phrase-name Erica Cantabrica flore maximo, foliis Myrti subtus incanis of Tournefort, and repeating the information first published by Ray (1704; Dillenius 1724). Caleb Threlkeld (1726) listed this western heather as Erica Cantabrica flore maximo, foliis Myrti subtus incanis. Erica Sancti Dabeoci, the name by which it was known to botanists throughout the early 18th century. The only other Irish Flora published before 1753, Botanalogia universalis Hiberniae by John Keogh (1735), did not include any account of St Dabeoc’s heath. LINNAEUS AND ST DABEOC’S HEATH The pre-Linnaean history (prior to 1753 when Linnaeus’ Species plantarum was published) of St Dabeoc’s heath is uncomplicated and of little significance as far as, for example, the nomenclature of the species are concerned. However, the post-1753 history is more complicated. Linnaeus was well informed about the botanical works of earlier decades, as is clearly shown by any page of his own publications. Thus it is not surprising that St Dabeoc’s heath should be included in Linnaeus’ works from an early period. While inexplicably there is no entry for this Hiberno-Cantabrian species in the first edition of Species plantarum (Linnaeus 1753), which is the bench-mark for all modern botanical nomenclature, when Linnaeus turned his attention to the flora of Britain and Ireland, he became aware of Erica S. Dabeoci Hibernica. Linnaeus’ chosen names for St Dabeoc’s heath are listed in Table 1. Erica daboeci of Flora Anglica 1754 In Flora Anglica, a work that is essentially a concordance providing Linnaean binomials for plants described in Dillenius’ edition of Ray’s Synopsis stirpium Britannicarum (1724), and which appeared in the guise of a thesis defended by Isaac Olaus Grufberg dated 3 April 1754, the final entry, under “Dubia’, is for “471 Erica 6 daboeci’”. The numerals refer respectively to the page and the paragraph number in Dillenius (1724) — in fact Linnaeus should have used either 470 because that is the page on which the first Erica appeared, or 472 because Dillenius’ entry for St Dabeoc’s heath is on that page. That combination of Latin words Erica and daboeci was original — neither Dillenius nor Petiver, the two authorities cited by Linnaeus, nor any previous author had used that particular set of words. Did Linnaeus intend this as the binomial for the heather? Why did he list it among the doubtful (“Dubia”)? And, why did he spell it daboeci? The answer to the latter question is perhaps linked with the many typographical errors which occurred in Flora Anglica (cf. Stearn 1973: 44; see also Sealy 1959). The first two questions are discussed below. Stearn (1973) discussed the significance of the specific names in Linnaeus’ Flora Anglica, stating that: Although the nomenclature of Linnaeus’s Flora Anglica (1754) follows closely that of his Species Piantarum (1753) it contains a number of specific names not included in that earlier work, and these must be regarded as new names validly published by reference to the previously published descriptive information given in the cited entries of Ray’s Synopsis, third edition (1724). 50 EC” NELSON TABLE 1. NAMES PUBLISHED BY LINNAEUS AND HIS CONTEMPORARY WILLIAM HUDSON FOR ST DABEOC’S HEATH: THE RIGHT-HAND COLUMN GIVES THE MOSTLY PRE-LINNAEAN SOURCES EXPLICITLY CITED IN EACH PUBLICATION Linnaeus, Flora Anglica (April 1754: 29) a Erica 6 daboeci [Dillenius: Synopsis methodica stirpium Britannicarum] 471 |recte 472|.6 Petiver: Gazophylacii ... t.27.f. 4 Linnaeus, Flora Anglica, in Amoenitates academice ... 1V: 92, 111 (November 1759) Erica cantabrica [Dillenius: Synopsis methodica stirpium Britannicarum| 471 [recte 472].6 Ray: Dendrologia ... 98 Petiver: Gazophylacii ... t. 27. f. 4 Hudson, Flora Anglica (January—June 1762: 143) Vaccinium cantabricum Tournefort: [/nstitutiones rei herbaria] 603 Ray [i.e. Dillenius: Synopsis methodica stirpium Britannicarum|] 472 Ray: Historia Il Dendrologia ... 98 Ray: Historia Ul Appendix ... 244 Petiver: Gazophylacii ... t. 27. f. 4 Linnaeus, Species plantarum ... editio secunda (September 1762: I, 509) Erica daboecii Tournefort: [/nstitutiones rei herbarie| 603 Ray: Historia III Dendrologia ... 98 Petiver: Gazophylacii ... t. 27. f. 4 Vaccinium cantabricum Hudson: Flora Anglica 143 Linnaeus, Systema nature ... editio duodecima (1767: Il, 300) Andromeda dabecia Erica daboecii [Linnaeus]: Species plantarum 509 Hudson, Flora Anglica ... edition altera (1778: I: 166-167) Erica daboecii Linnaeus, Species plantarum [ed. 2] 509 Stearn cited examples of, for example, suppression of names by Linnaeus between the first edition of Flora Anglica and the second edition published in Amoenitates academicae (1759), but he does not comment on the entry for St Dabeoc’s heath. How does the entry for “Erica 6 daboeci” stand with regard to the International code of botanical nomenclature (Tokyo 1993) (hereinafter the JCBN (1993))? Does it fall within Art. 26.6 whereby certain “designations are not to be regarded as specific names’, including Art. 26.6 b, such unspecified “‘other designations of species consisting of a generic name followed by one or more words not intended as specific epithets.” The final section of Flora Anglica headed “Dubia” is not laid out like the previous portion of the fifth section, and most entries in it contain more than one word (a descriptive phrase) which clearly were not intended to be nomina trivialia. Moreover the generic name and the one-word (in this case) phrase are separated by the numeral indicating the paragraph number in Synopsis. Given the use of phrases, and given that the typography is different, there does appear to be cause to reject this designation because it was “not intended as [a] specific epithet.” It is also possible to argue that because Linnaeus did not adopt binomials in the section headed “Dubia’, he did not accept any of the apparent binomials as nomina trivialia, and thus that they are not validly published under Art 34.1 (a): “A name is not validly published (a) when it is not accepted by the author in the original publication ...”.° Thus the first edition of Linnaeus’ Flora Anglica (1754) must be discounted as a source of a binomial for St Dabeoc’s heath. HISTORY OF ST DABEOC’S HEATH 51 Erica cantabrica of Flora Anglica 1759. Linnaeus’ next publication of relevance is the second (1759) edition of Flora Anglica, in which he used a different name, Erica cantabrica, and cited an additional reference, Ray's Dendrologia. In this edition of Flora Anglica, Erica cantabrica is printed twice, once as an undoubtedly deliberate binomial within Linnaeus’ commentary on Ireland (Linnaeus 1759: 92) and later within the section headed “Obscura” in the form “471. Erica 6. cantabrica” (Linnaeus 1759: 111). Linnaeus’ reasons for changing the entry from “471 Erica 6 daboeci” are not known. What is clear, however, is that even in 1759 Linnaeus had not seen any specimens of this plant, only the crude figure in Petiver’s Gazophylacii nature. There can be no doubt that Linnaeus did intend Erica cantabrica to be a binomial (nomen triviale) because he used it deliberately on p. 92, but he was not consistent, and he did not list it with the names of four other species of Erica on p. 100. Was this binomial validly published in the second edition of Flora Anglica? Although the case is different, because of the prior use of Erica cantabrica on p. 92, under Art. 26.6 of the JCBN (1993) (see above) it has to be argued that this name also is not validly published. Erica daboecii of Species plantarum 1762. It is noteworthy that Linnaeus reverted to Erica daboecii in the second edition of Species plantarum (1762), citing William Hudson’s name Vaccinium cantabricum (see below) as a synonym. Linnaeus stated that he had not seen flowers’, but his description suggests that by this time he had been able to examine a fruiting specimen with some leaves; he correctly noted that the capsule was erect and large and had four valves, facts not in previously published descriptions even that by Ray (1704).° Three years after the publication of the new edition of Species plantarum, Linnaeus received a herbarium specimen with flowers from Peter Collinson in September 1765, and in a letter dated 27 December 1765 commented to John Ellis that “Erica Dabeoci was sent by Peter Collinson: a fine specimen which much delighted me. It is truly an Erica, though so unlike the rest” (see O’ Neill & Nelson 1995). WILLIAM HUDSON AND ST DABEOC’S HEATH The Latin binomial for St Dabeoc’s heath, Daboecia cantabrica, has as its basionym Vaccinium cantabricum published by William Hudson F. R. S., a London apothecary and one-time Assistant Librarian in the British Museum, in his Flora Anglica (Hudson 1762), the work which firmly established “Linnean principles of botany in England, and their application to practical use” (Smith 1824). Hudson was planning to compile a British Flora as early as the Summer of 1760. On 3 July 1760 he wrote to Linnaeus (Ms in Linnaeus’ correspondence; Linnean Society of London): I have taken the liberty to send you thro the hands of my worthy friend Mr John Ellis some few of the Plants which are mentioned at the end of your Flora Anglica togather with some others. I should have sent you some others but Mr. Ellis informed me that the ship was to sail in two or three Days, and that he should put up his things the next morning, which hinderd mee from looking out of my collection any more at present, but hope to send as many more before the end of the summer and if there are any plants which are mentioned In R[ay’s] Syn[opsis] which your are in want of if you will honour mee with a list off them, I will do all that laies in my power to procure them for you. I should esteem it as a peculiar Favour if you will be so kind as to honour me with your opinnion concerning the plants but more Especially the grasses which have plagued me much. I should not have thought of giving you so much trouble had I not had some intentions of Publishing a flora Britanica and not having been able to meet with that information which I wanted here. I hope I shall soon have the honour of hearing you have rec[eive]d the plants safe. se . some few of the Plants which are mentioned at the end of your Flora Anglica” would include St Dabeoc’s heath, which was in fact last of all, but we can be fairly certain Hudson did not send a specimen to Linnaeus as none from Hudson is in Linnaeus’ herbarium (LINN). 52 E. C. NELSON In his Flora Anglica Hudson removed St Dabeoc’s heath from Erica and placed it in the genus Vaccinium, but why? What was his actual role in the description and naming of St Dabeoc’s heath? Did he examine specimens of the heather, study the species in details and then write a fresh, original description and, finally, devise a Latin name for the plant without ever referring to Linnaeus’ earlier Flora Anglica? Or did he merely take up a copy of Linnaeus’ Flora Anglica and rework the text and the name to suit his own Flora? Hudson undoubtedly owned, or had direct access to, a copy of the second edition of Linnaeus’ Flora Anglica. Within the bibliography of his own Flora Anglica (the section of the book headed “Nomina auctorum explicata’) Hudson (1762) cited Amoenitates academicae IV in which the second edition of Linnaeus’ Flora Anglica was published during November 1759 (Stearn 1973). Hudson’s description of St Dabeoc’s heath, like that published a few months later by Linnaeus (1762), is clearly derived from Ray’s and is not in any sense original. Hudson added no new facts to the sparse set of words employed by Ray. Thus there is no textual evidence that Hudson had a specimen, alive or pressed, of the heath. I have never seen a herbarium specimen of St Dabeoc’s heath bearing his name or annotations — on the other hand, most of Hudson’s herbarium was destroyed when his house in Panton Street, London, was gutted in a fire in 1783 (Henrey 1975: II, 110; Kent & Allen 1984; Desmond 1994), and it is quite possible whatever specimens he had were burnt then. If he had no specimen to compare with other species of Erica and Vaccinium, it is hard to understand why he made the decision to transfer St Dabeoc’s heath into Vaccinium. I suggest he reasoned that because the illustration published by Petiver (Fig. 1) showed a sprig with broad leaves and large flowers, more like the large-blossomed, broad-leaved Vaccinium than the usually small-flowered, needle-leaved heathers (Erica), this plant belonged therein, like going with like. Thus there is no evidence that Hudson made a detailed analysis of the plant. He employed previously published descriptions and the illustration from Petiver (1704) as the basis for his description of Vaccinium cantabricum. He also simply took up the epithet cantabrica from Tournefort (1694) and the second edition of Linnaeus’ Flora Anglica (1759) as the specific epithet. In terms of the current rules of nomenclature, Hudson’s name Vaccinium cantabricum was the first validly published binomial for St Dabeoc’s heath and thus it has priority, but, I would argue, that has happened by default because he omitted to be directly explicit about his sources. In his Flora Anglica Hudson (1762) occasionally omitted to cite the original author and source of binomials that he published. A good example, one that is discussed by Stearn (1973: 66), is Hypericum elodes which Linnaeus (1759) had published, and there are other instances of Hudson apparently using names from Linnaeus’ earlier Flora Anglica (1759) without acknowledgement, including Trifolium ochroleucon which is in Linnaeus’ work. Comparable examples in the publications of Hudson’s contemporaries, Philip Miller and John Hill, are Helianthemum nummularium and Primula elatior respectively (see Brummitt & Meikle 1993). In the second edition of Flora Anglica, Hudson (1778) accepted Linnaeus’ name Erica daboecii, quoting the second edition of Species plantarum as the source (Linnaeus 1762) and so he abandoned Vaccinium cantabricum without even mentioning it in synonymy. A footnote may be added, by quoting an anonymous reviewer referring to Jussieu’s paper (1802) in which yet another binomial, Menziesia polifolia, a name that held sway for many decades, was published: We cannot refrain from adding, that had our author [Jussieu] thought fit to have retained the trivial name [i.e. specific epithet] before in use, he would in great measure have avoided the confusion necessarily attendant on the change of appellation, as Dabeoci, already bandied from one genus to another, would have been readily recognised under a new generic title. [Anonymous 1806]. DABOECIA AND/OR DABEOCIA The generic name Daboecia — based on the name Dabeoc but with the vowels e and 0 reversed — is conventionally attributed to David Don (1834), but there are several previous works in which it was printed although not validly published, the earliest being in the second part of Thomas Martyn’s “corrected and newly arranged” edition of Miller’s The gardener’s and botanist’s dictionary (the title page of volume | part II is dated 1807; Henrey 1975: II, 91). Martyn included HISTORY OF ST DABEOC’S HEATH 53 vy Vay \\ ey | NT \ A a \A4 \ WWE, ay / \ ve uy e' \ if fe ‘y Sl \h \ Fiera) n AL \\ 7 Lo : \ My AN : ae =“ NA IN ae va ry y f . ‘) } ae A eae Me i FOX BVP { N Age SM dl X Deo Se S \ \\ oe X, n* FIGURE |. St Dabeoc’s heath from Petiver’s Gazophylacii nature et artis decas tertia (1704), tab. 27, fig. 4; this is designated (p. 55) as the lectotype of Daboecia cantabrica (Huds.) K. Koch. 54 E. C. NELSON Daboecia as a generic name at the beginning of D, but simply added “See Andromeda” — Andromeda Daboecia is the ninth species within Andromeda. The implication of Martyn’s edition of The gardener’s ... dictionary is that the name Daboecia had been coined by 1807, and was sometimes employed by gardeners and botanists as early as the first decade of the 19th century. Another intriguing example of its use (with the vowels reversed) before Don formally published the generic name is in a letter from Richard Duppa of Lincoln’s Inn to Sir James Edward Smith, President of the Linnean Society, dated 16 July 1827 (Smith 1832, Britten 1916, Nelson 1978). Duppa wrote: Some where or other, I think, I have met with the assertion that there was but one Heath indigenous to Ireland and that is now removed to the genus Dabeecia. My question is, whether, of the four species of English Heath any one is found wild in Ireland. I have got into a dispute on that point & I cannot find my difficulty solved in any book I have on the subject of Botany, & also if our English heaths are not found in Ireland whether the bogs of Ireland are composed of the roots of the Daboecia, the Salix herbacea, or what? ... And is ye Dabcecia which is peculiar to Ireland, never found in any other country, that we are acquainted with? What was Duppa remembering? Was it Martyn’s edition of Miller’s The gardener’s and botanist’s dictionary, a conversation with a fellow botanist (perhaps David Don, as suggested by Britten (1916: 179 fn.)), or a now-lost source? Smith’s reply is not extant, so the mystery remains. But the more extraordinary thing about Duppa’s letter is that it was published, slightly edited, as early as 1832 by Lady Pleasance Smith in her Memoir and correspondence of the late Sir James Edward Smith (Smith 1832: 2, p. 125). Thus Duppa’s letter was in the public domain two years before David Don validly published the generic name Daboecia. Lady Smith (1832: 2, p. 125) added this footnote to the name Dabecia, echoing her late husband’s opinion (Smith 1791): Mr. Duppa probably meant Menziesia; there is no genus Dabecia. The plant was called Andromeda Dabecia in Linn. Syst. Veg. It is named Dabeoci after St. Dabeoc, whence the Linnean trivial name has been corruptly taken. Given that Daboecia was derived, to quote Smith (1791) again, “corruptly” from the personal name Dabeoc (Nelson 1984), should the spelling be corrected to Dabeocia? This question has exercised botanists for more than a century. Article 60.7 of the JCBN (1993) states that “When changes made in orthography by earlier authors who adopt personal, geographic, or vernacular names in nomenclature are intentional latinizations, they are to be preserved ...” and among examples given is Gleditsia L. (1753) which commemorates Gleditsch “... but the name is ... not to be altered to ... “Gleditschia” ... [because] Linnaeus latinized the names of [this botanist] deliberately as ... Gleditsius ...”. However the ICBN (1993) also states (Art. 60.1) that “The original spelling of a name or epithet is to be retained, except for the correction of typographical or orthographical errors ...”. Was Daboecia an intentional latinization on the part of Martyn (1807) or even Don (1834), or a “typographical or orthographical error’? Clearly those authors based the generic name on the specific epithet taken up by Linnaeus (1754, 1762) from Dillenius’ edition of Synopsis stirpium Britannicarum and thus ultimately from Ray’s Historia plantarum (1704) (see e.g. Sealy 1949). Yet Dillenius used dabeoci, as did his predecessors (see Table 2). We cannot explain why Linnaeus spelled the epithet daboeci (Linnaeus 1754, 1762). Nicholson (1931) suggested that Don “Latinised [the name] to Dabcecia for convenience’, while Sealy (1949) argued that Don “knew perfectly well that the saint’s name was Dabeoc, and ... must have chosen to use the spelling Daboecia quite deliberately.” On the other hand Babington (1889) stated that he did “not see why Dabeocia should be changed into Daboecia because Don’s printers made a not unnatural blunder, and [Don] did not discover it when correcting the press.” Jackson (1889) protested, and argued that Don “clearly meant the name to stand as he wrote it”, Daboecia. That was not Praeger’s opinion (1925): “... when David Don in 1834 placed the plant in a separate genus, he followed the Linnean spelling, and called the genus Dabeecia instead of Dabeocia, being no doubt unaware of the facts of the case.” Alas, no evidence exists to show conclusively whether Martyn or Don deliberately latinized Dabeoc’s name, or never corrected their printers’ errors. HISTORY OF ST DABEOC’S HEATH Sp, TABLE 2. THE TWO VARIANTS OF THE GENERIC NAME AND SPECIFIC EPITHET DERIVED FROM DABEOC, AND SOME OF THE AUTHORS WHO HAVE USED THESE VARIANTS; THE LISTS OF AUTHORS ARE NOT EXHAUSTIVE. Daboecia : daboecil Dabeocia : dabeocii Linnaeus 1754, 1762, 1767 Lhuyd [1700] 1712 Hudson 1778 Ray 1704 Smith 1791 Dillenius 1724 Jussieu 1802 Threlkeld 1726 [as Dabezoci] Martyn 1807 More & Moore 1866 [Duppa 1827] Koch 1872 Smith 1832 Babington 1889 Don 1834 Colgan & Scully 1898 Jackson 1889 Chittenden 1931 Webb 1953: 103 [but Dabeocia in Index: 242] Webb [1943: 215] Woodell 1958 Gilbert-Carter 1950 Webb & Scannell 1983 There is an irrefutable case for correcting the orthography of this name and reinstating Dabeocia (with e followed by 0). Such an action does not conflict with the JCBN (1993). Indeed Dabeocia was used by the principal Irish botanical authors (Table 2) until the early 1950s (see e.g. Webb 1943, 1953). While I argue that the name should be corrected, just as Linnaeus’s orthographic error Stewartia has been corrected to Stuartia (in that case Linnaeus was misinformed about the surname of the earls of Bute, it being Stuart and not Stewart (see e.g. Stearn 1996)), such a move is unlikely to gain acceptance. Stability in nomenclature is preferable to instability. In this context it is worth quoting Chittenden’s (1931) response to Nicholson (1931): there “can be only one recognised way to spell a name, and we must follow the original, unless to do so would lead to confusion. In neither of the instances given can following the original lead to confusion, so we must write Daboecia and Diervilla |after Mons Dierville; see Stearn 1996] in spite of the probable mistakes on the part of the originators of the names.” PRONUNCIATION AND TYPIFICATION Daboecia comes ultimately from an Irish name Dabeoc, and commemorates an obscure saint of the early Irish church (see Nelson 1984). Dabeoc is pronounced as two equal syllables, da-vok. How should Daboecia be pronounced? In one sense it is a nonsensical name, and there can be no correct pronunciation. Webb (1943: 215) suggested DabeOd’cia, while da-b6-ee-kee-a was suggested by Coombes (1985). Bearing in mind the root Dabeoc, da-vok-ee-a seems most appropriate, and has the merit of being simple. Finally, there is no record of typification of the name Daboecia cantabrica. As noted, no herbarium specimens used by Hudson or Linnaeus in the preparation of their respective floras is known. Both Linnaeus (1754, 1759, 1762) and Hudson (1762, 1778) cited the same figure, tab. 27, fig. 4 from Petiver’s Gazophylacii nature et artis decas tertia (1704). This illustration (Fig. 1) can serve as lectotype; it is the only possible choice. Daboecia cantabrica (Huds.) K. Koch, Dendrologie vol. 2 (pt 1): 132 (1872). basionym: Vaccinium cantabricum Huds., Flora Anglica: 143 (1762). synonym: Erica daboecii L., Species plantarum 509. ed. 2. (1762); Hudson, Flora Anglica: 166. ed. 2. (1778), etc. Lectotypus hic designatus: icon, tab. 27 fig. 4, in J. Petiver (1704) Gazophylacii nature et artis decas tertia. 56 EC. NEESON ACKNOWLEDGMENTS My particular thanks are due to Gina Douglas, Librarian, Linnean Society, London, for making available the many eighteenth century sources that form the basis of this paper. I am also grateful to Dr R. K. Brummitt, Royal Botanic Gardens, Kew, for his invaluable comments and criticisms of my opinions. REFERENCES ANONYMOUS (1806). [Review of] Annales du Muséum d Histoire Naturelle ... an xi (1802) ... If Memoir on the plant called by Botanists Erica Daboecia [more properly Dabeoci] ... By Jussieu. Annals of botany 2: 167-169. BABINGTON, C. C. (1888). On botanical nomenclature. Journal of botany 26: 369-371. BRITTEN, J. (1916). An overlooked Irish botanist. Journal of botany 54: 173-180. BRUMMITT, R. K. & MEIKLE, R. D. (1993). The correct Latin names for the Primrose and the Oxlip. Watsonia 19: 181-184. CANDOLLE, A. P. DE (1805). Flora Francaise. Desray, Paris. CHITTENDEN, F. J. (1931). [Daboecia]. Gardening illustrated 53 (14 March): 164. COLGAN, N. & SCULLY, R. (1898). Contributions towards a Cybele Hibernica. 2nd ed. E. Ponsonby, Dublin. COOMBES, A. J. (1985). The Collingridge dictionary of plant names. Collingridge, London. DANDY, J. E. (1958). The Sloane herbarium. British Museum (Natural History), London. DILLENIUS, J. (1724). Synopsis methodicum stirpium Britannicarum. (Facsimile edition 1973. The Ray Society, London:). Don, D. (1834). An attempt at a new arrangement of the Ericaceae. Edinburgh philosophical journal 14: 150- 160. DUPONT, P. (1962). La flore Atlantique Européenne. Introduction a |’ étude phytogéographique du secteur ibéro-atlantique. (Theses présentées a la faculté des sciences de l’université de Toulouse. 1) Toulouse: Imprimerie Edouard Privat. GILBERT-CARTER, H. (1950). Glossary of the British flora. Cambridge University Press, Cambridge. GUNTHER, R. W. T. ed. (1928). Further correspondence of John Ray. The Ray Society, London. GUNTHER, R. W. T. (1945). Early science in Oxford Vol. XIV Life and letters of Edward Lhwyd. Oxford University Press, Oxford. HENREY, B. (1975). British botanical and horticultural literature before 1800. Oxford University Press, London. HUDSON, W. (1762). Flora Anglica, exhibens plants per regnum Angliae sponte crescentes ... The author, London. HuDSON, W. (1778). Flora Anglica, exhibens plants per regnum Britanniae sponte crescentes ... 2nd edition. The author, London. JACKSON, B. D. (1889). Daboecia. Journal of botany 27: 50. JUSSIEU, A. DE (1802). Mémoire sur la plante nommée par les botanistes Erica daboecia, et sur la nécessité de la rapporter a un autre genre et a une autre famille. Annales du Muséum d'Histoire Naturelle, Paris 1: 52-56. KEOGH, J. (1735). Botanalogia universalis Hiberniae. Cork. KOCH, K. (1872). Dabeocia in Dendrologie 2(1). LHwyD, E. (1712). Some father observations relating to the antiquities and natural history of Ireland. In a letter from the late Mr. Edw. Lhwyd ... to Dr Tancred Robinson, F. R. S. Philosophical transactions 27: 524-526. LINNAEUS, C. (1753). Species plantarum ... L. Salvius, Stockholm. (Facsimile edition. 1957. The Ray Society, London). LINNAEUS, C. (1754). Flora Anglica. L. M+ Hojer, Uppsala. (Facsimile edition, with Dillenius (1724). 1973. The Ray Society, London). LINNAEUS, C. (1759). Flora Anglica, in Amoenitates academicae ... 4: 111. (Facsimile edition, with Dillenius (1724). 1973. The Ray Society, London). LINNAEUS, C. (1762). Species plantarum ... 2nd ed. Tom. 1, p. 509. L. Salvius, Stockholm. LINNAEUS, C. (1767). Systema naturae per regna tria naturae ... 12th ed. Tom 2, p. 299-300. L. Salvius, Stockholm. LINNAEUS, C. (1770). De Erica. J. Edmann, Uppsala. MCCLINTOCK, D. C. (1969). Daboecia azorica and its hybrids with D. cantabrica. Journal of the Royal Horticultural Society 94: 449-453. MCCLINTOCK, D. C. (1989). The heathers of Europe and adjacent areas. Botanical journal of the Linnean Society 101: 279-289. HISTORY OF ST DABEOC’S HEATH 2) MARTYN, T. (1807). The gardener’s and botanists dictionary ... by the late Philip Miller ... corrected and newly arranged. Vol. 1, pt. Il. Rivington et alii, London. MERINO, B. (1906). Flora descriptiva é illustrada de Galicia. Vol. 2, pp. 250-251. Santiago. MITCHELL, M. E. (1975). Irish botany in the seventeenth century. Proceedings of the Royal Irish Academy 75 B (13): 275-284. Moore, D. & Mork, A. G. (1866). Contributions towards a Cybele Hibernica. Hodges, Smith, Dublin. NECKER, N. J. DE (1791). Elementa botanica. Paris. NELSON, E. C. (1978). Historical records of the Irish Ericaceae, with particular reference to the discovery and naming of Erica mackaiana Bab. Journal of the Society for the Bibliography of Natural History 9: 289- DOO: NELSON, E. C. (1984). Dabeoc — a saint and his heather. Yearbook of the Heather Society 3 (2) : 41-46. NELSON, E. C. & WALSH, W. F. (1995). The flowers of Mayo. Dr Patrick Browne’s Fasciculus plantarum Hiberniae (1788). Eamonn de Burca, Dublin. NICHOLSON, C. (1931). Erroneous plant names. Gardening illustrated 53 (12 February): 113. O’ NEILL, J. & NELSON, E. C. (1995). Introduction of St Dabeoc’s heath into English gardens, 1763. Yearbook of the Heather Society 1995: 27-32. OWEN, E. (1922). A catalogue of the manuscripts relating to Wales in the British Museum. The Honourable Society of Cymmrodorion, London. PETIVER, J. (1704). Gazophylacii nature et artis decas tertia: ... tab. 27, p. 42. Christopher Bateman, London. PRAEGER, R. LL. (1925). St Dabeoc’s heath. The garden 89: 180. RAVEN, C. E. (1950). (reissued 1986) John Ray naturalist his life and works. Cambridge University Press, Cambridge. RAY, J. (1704). Historia plantarum ... Tom. 3. p. 98. Samuel Smith & Benjamin Walford, London. SALISBURY, R. (1781). Species of Erica. Transactions of the Linnean Society 6: 316-338. SEALY, J. R. (1949). Daboecia azorica. Curtis’s botanical magazine 166 (new series): tab. 46. SMALL, D. J. & SMALL, A. (1998). The Heather Society handy guide to heathers. Denbeigh Heather Nurseries, Creeting St Mary. SMITH, J. E. (1791). Erica dabeoci. English botany tab. 35. SMITH, J. E. (1824). Preface in The English flora. Vol. 1, pp. xili-xiv. London. SMITH, P. (1832). Memoir and correspondence of the late Sir James Edward Smith ... London. Vol. 2. pp. 124-127. STEARN, W. T. (1973). [Introduction to facsimile edition of] John Ray Synopsis methodica stirpium Britannicarum editio tertia 1724 ... The Ray Society, London. STEARN, W. T. (1996). Stearn’s dictionary of plant names for gardeners. (‘Paperback, with some emendations’ ) Cassell, London. THRELKELD, C. (1726). Synopsis stirpium Hibernicarum. Dublin: the author. (Facsimile edition. 1988. Boethius Press, Kilkenny). TOURNEFORT, J. P. (1694). Elemens de botanique. Imprimerie royale, Paris. TOURNEFORT, J. P. (1700). Institutiones rei herbaria. Editio altera, Gallica longe auctior ... Tom. 1, p. 603. Typographia Regia, Paris. THUNBERG, C. P. (1785). De Erica. Joh. Edman, Uppsala. TUTIN, T. G. & WARBURG, E. F. (1932). Contributions from the University Herbarium, Cambridge. — Notes on the flora of the Azores. Journal of botany 70: 7-13. WEBB, D. A. (1943). An Irish flora. \st ed. Dundalgan Press, Dundalk. WEBB, D. A. (1953). An Irish flora. 2nd ed. Dundalgan Press, Dundalk:. WEBB, D. A. & SCANNELL, M. J. P. (1983). Flora of Connemara and The Burren. Royal Dublin Society and Cambridge University Press, Cambridge. WOODELL, S. R. J. (1958). Biological flora of the British Isles. Daboecia cantabrica. Journal of ecology 46: 205-216. NOTES 1 Erica Cantabrica flore maximo, foliis Myrti subtus incanis. 2 Erica Cantabrica, flore maximo, foliis Myrti subtus incanis. 3. 30. Erica S. Dabeoci Hibernica D. Lhwyd. Er. Cantabrica flore maximo, foliis Myrti, subtus incanis Tournefort. Inst. rei Herb. Erice species genuina esse videtur quoad floris formam, cum Erica tenuifolia Unedonii flore conveniens. Vasculi seminalis forma Geranium imitari videtur. Cauliculi autem ubi florere incipiunt non minus viscosi sunt quam Muscipulz, In montibus Mayo squalido & spongioso solo frequens est, ut & per totum Hiar-Connacht in Gallovidia. Mulericule superstitiose surculos ejus secum circumferunt adversus incontinentiam. 58 EC. NELSON 4 which simply reads ‘Erica Hibernica foliis myrti pilosis, subtus incanis’ 5 4.Erica HIBERNICA fol. Myrti pilosis subtus incanis Hort. nost. sicc. 244. E. 20. Erica S. Dabeoci D. Lhwyd Ray Dend. V. 3. p. 98. 30. Erica Cantabrica Flore maximo, foliis Myrti subtus incanis Instit. Rei Herb. 603 5. Elem. Botan. 475. I had a specimen of this elegant Plant from my generous Friend Mr Edward Lhwyd, Keeper of the Museum Ashmoleanum, who gathered it in Jreland in the places Mr Ray mentions. 6 Neither of the two editions of Linnaeus’ Flora Anglica (1754, 1759) is listed in the Appendix V (Opera utique oppressa) of the ICBN (1993) so names published in them are not liable to be regarded as not validly published under the same CBN (1993, Art. 32.8). 7 ‘Flores non vidi, quos describant autopte.’ In Linnaeus’ own, interleaved copy of this edition (now in the Linnean Society, London), he has written a very detailed description of the flowers and scored out the sentence quoted, as well as the phrase ‘Rami breviores.’, and inserted in manuscript after ‘lanceolata’ (referring to the leaves) the additional phrase ‘s[eu] elliptica’. The manuscript description in this copy was published in Systema naturae (Linnaeus 1767) ; the printed text is almost exactly the same. 8 Quoted from the original manuscript; Smith correspondence, Linnean Society, London. (Accepted April 1999) Watsonia 23: 59-81 (2000) 59 Engulfed by suburbia or destroyed by the plough: the ecology of extinction in Middlesex and Cambridgeshire C.D, PRESTON Institute of Terrestrial Ecology, Monks Wood, Abbots Ripton, Huntingdon, Cambs., PE17 2LS ABSTRACT Middlesex (v.c. 21) has lost 18% of the recorded native vascular plant species through extinction; the corresponding loss in Cambridgeshire (v.c. 29) is 13%. If species which became extinct before 1750 are excluded, this amounts to the loss of one species every 1-7 years in Middlesex and one every 2-0 years in Cambridgeshire. The extinct species share many ecological similarities: they tend to be small, to grow in open habitats and to be characteristic of environments with low fertility. A disproportionate number of calcifuge species has been lost in both areas; in Middlesex, where calcareous soils are rare, many marked calcicoles have also been lost. In both counties the percentage extinction in the small Boreal and Oceanic groups has been greater than that in the larger phytogeographical elements. The peak periods of extinction in Cambridgeshire have been those characterised by major agricultural change: in Middlesex they coincide with the spread of the London conurbation. The overall totals suggest that, judging by the number of extinctions, agricultural intensification is almost as damaging to the native flora as urbanisation. KEYWORDS: Life-form, Ellenberg values, phytogeography. INTRODUCTION One of the many ways to study the changes in the native flora of Britain during recent centuries is to look at extinctions at a local level. Over a century ago, West (1898) suggested that no counties had lost more native species than Middlesex (v.c. 21) and Cambridgeshire (v.c. 29), two areas Where the landscape had been drastically modified by agricultural and urban change. A comparison of the nature of the species which have become extinct in these counties ought therefore to be particularly illuminating. The aim of this paper is to investigate whether the vascular plant species which disappear when an area undergoes major environmental change have anything in common either ecologically or phytogeographically, or whether they approximate to a random sample of the species present in the area. Middlesex (734 km?) is only a third the size of Cambridgeshire (2124 km7), but the counties share important geographical similarities . Both areas are exclusively lowland: the highest point in Middlesex scarcely exceeds 150 metres and nowhere in Cambridgeshire reaches even this altitude. Neither county has a coastline, though a few coastal species penetrate into Cambridgeshire along the tidal portions of the River Nene and reach Middlesex along the tidal Thames. There are, however, important differences in the superficial geology of the two areas. Middlesex is made up of the predominantly acidic clays, sands and gravels of the London Basin with a small area of chalk at the edge of the county near Harefield. In Cambridgeshire calcareous soils predominate; acidic ground is rare. Southern Cambridgeshire is dominated by chalk and calcareous boulder clay, with only small areas of acidic, sandy soil over the Lower Greensand near Gamlingay at the western edge of the county and on the fringe of Breckland to the east. The extensive Fenland of northern Cambridgeshire, which falls below sea-level in places, has no parallel in Middlesex. A comparison of extinctions in Middlesex and Cambridgeshire is particularly appropriate because of certain similarities of botanical history. Most importantly, the counties share a long tradition of field study, sustained in Middlesex by the presence of London with its metropolitan population and institutions and in Cambridgeshire by the University of Cambridge. In Cambridgeshire the first thorough investigation of the flora was Ray’s Catalogus plantarum circa * Like Kent (1975), I have taken the area of the Watsonian vice-counties from the Agricultural Returns for 1873, the year of publication of Watson's Topographical botany. In this paper the names Middlesex and Cambridgeshire refer to the Watsonian vice-counties unless the administrative counties are explicitly referred to. 60 C. D. PRESTON Cantabrigiam nascentium of 1660, followed by another detailed Flora, Relhan’s Flora Cantabrigiensis (1785). Rather surprisingly, there was no Flora of Middlesex in this period but very many records from sites in Middlesex were published in national works from those of Turner (1548) onwards, and these are augmented by local publications (e.g. Blackstone 1737) and specimens in the London herbaria. The botanical history of the two counties subsequently converged, with the publication in the mid-19th century of Floras which included both historical records and the results of recent fieldwork (Babington 1860; Trimen & Dyer 1869). In the 20th century, Floras of the two counties were again published littlke more than a decade apart, Cambridgeshire (Perring et al. 1964) again preceding Middlesex (Kent 1975). These have been supplemented by updated checklists (Crompton & Whitehouse 1983; Kent, in press). Kent’s book The historical flora of Middlesex (1975) was especially notable for its thorough coverage of the voluminous historical records available from the county. IDENTIFICATION OF NATIVE AND EXTINCT NATIVE SPECIES In investigating the characteristics of the extinct species, I have attempted in this paper to characterise the native flora of each county as a whole with respect to each ecological or phytogeographical factor, then to compare this overall pattern with that shown by the species which are regarded as extinct. Lists of native and of extinct native species for each county were drawn up from published and unpublished sources, namely Kent (1975 and in press) for Middlesex and Perring ef al. (1964), Crompton & Whitehouse (1983), Crompton & Wells (1996) and Preston (1997) for Cambridgeshire. These draft lists were then submitted to the vice-county recorders, R. M. Burton (v.c. 21) and Mrs G. Crompton & D. A. Wells (v.c. 29) for vetting. Nomenclature follows Stace (1997). DEFINITION OF NATIVE SPECIES The list of British native species provided by Preston & Hill (1997) was used as the basis of the study. Where there is doubt about whether or not a species is native in Middlesex or Cambridgeshire, the species was accepted as native if its history, habitat and distribution in the county is similar to that in those areas of Britain where it was assumed native by Preston & Hill (1997). Kent (1975) classifies Middlesex plants as native, denizen, colonist, alien or introduced, and the list of native species accepted here is very similar to the combined list of natives, denizens and colonists. The Cambridgeshire Floras pay less attention to the question of whether or not species are native. IDENTIFICATION OF EXTINCT SPECIES Species were regarded as extinct if they had not been reported from the county from 1970 onwards, or had been seen between 1970 and 1989 but were known to have disappeared subsequently. No species seen from 1990 onwards was regarded as extinct, on the grounds that such a judgement would be premature. Critical segregates which have been recognised recently from herbarium material collected before 1970 (e.g. Erophila majuscula) were not regarded as extinct if there has not yet been time to search for them in the wild. In compiling the list for Cambridgeshire an unpublished list of extinct species prepared by Crompton & Wells (1996) was particularly helpful, but this needed considerable modification as it includes both native and alien taxa. The last year in which a species was recorded was noted for each species. By convention, species which were known in Cambridgeshire to Relhan (1785, 1802, 1820) and Babington (1860) but not to later authors were regarded as having been last recorded in 1820 and 1860 respectively in the absence of an exact date for the last record. In some cases species are extinct as natives but survive as aliens or introductions. These have been treated as extinct, with the date of extinction the year when a native population was last recorded. A particular difficulty in Middlesex was the treatment of a number of arable weeds which are extinct as natives but persist as casuals. The species in this category were treated as extinct (Anchusa arvensis, Anthemis cotula, Chrysanthemum segetum, Scandix pecten-veneris, Silene noctiflora, Stachys arvensis). The last record as a persistent weed of cultivated land has been taken as the last native record, though this is difficult to establish with certainty. The extinct natives in Middlesex and Cambridgeshire are listed in Appendix 1, with the date of the last native record. EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 61 NUMBER AND RATE OF EXTINCTIONS A total of 146 native species have become extinct in Middlesex since records began, representing 18% of the 816 species recorded. For Cambridgeshire the number of extinct native species is 120, 13% of the 897 species recorded. There are 45 species which have become extinct in both counties (Table 1). The number of species last recorded in successive 30-year intervals from 1750 is shown in Table 2. The rate of extinction, as measured by the number of species lost per decade, can be calculated on the assumption that a species became extinct in the period when it was last recorded. The year 1750 has been chosen rather arbitrarily as the starting point for these calculations. From 1750 to 1809 the rate of extinction was low, with 2-3 species lost from Middlesex per decade and 1-5 from Cambridgeshire. In the next 30 years the rate of extinction increased dramatically in Cambridgeshire, with a species lost on average every |-3 years. By contrast the rate actually fell in Middlesex during this period. This pattern was reversed between 1870 and 1929, when extinctions in Middlesex were almost twice as numerous as those in Cambridgeshire. Extinctions continued at approximately the same rate in Middlesex until 1989; in Cambridgeshire the rate increased from 1900 to 1959 to match that in Middlesex. From 1750 onwards 5-8 species have been lost per decade from Middlesex and 4-9 from Cambridgeshire; the rates for the 20th century (1900-1989) are 8-4 and 7-3 respectively. One species has been lost from Middlesex every 1-7 years and from Cambridgeshire every 2-0 years since 1750. TABLE 1. ACOMPARISON OF EXTINCT AND SURVIVING SPECIES IN MIDDLESEX (V. C. 21) AND CAMBRIDGESHIRE (V.C. 29) Extinct in v.c.21 | Never recorded in v.c. 21 Still present in v.c. 21 Total Extinct in v.c. 29 45 43 52. 120 Never recorded in v.c. 29 39 - 28 67 Still present in v.c. 29 62 105 610 vee Total 146 148 670 964 TABLE 2) KATE OF EXTINCTION OF SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) Species are assumed to have become extinct during the period when they were last recorded. The overall rate of extinction is calculated for the period 1750-1989, and excludes species last recorded before 1750. Time period Number of extinct species Rate of extinction (species per decade) V.C21 Vic. 29 Wecs 21 WeGs 20 pre-1750 8 2 - - 1750-1779 3 e] 1-0 1-0 1780-1809 11 6 3-7 2-0 1810-1839 6 24 2-0 8-0 1840-1869 14 10 4.7 3-3 1870-1899 Di 9 9-0 3-0 1900-1929 23 18 7-7 6-0 1930-1959 24 pH | 8-0 9-0 1960-1989 29 al OF7 7-0 Date unknown ] 0 = : Total 146 120 5-8 4-9 62 C.D. PRESTON ECOLOGICAL CHARACTERISTICS OF EXTINCT SPECIES In this section seven ecological characteristics of the extinct species are examined. Data on plant height and life-form are derived from the Ecological Flora Database (see Fitter & Peat 1994) with appropriate additions and corrections. The light, moisture, pH and nitrogen requirements of the species studied and the typical vegetation in which they grow are assessed by Ellenberg’s indicator values (Ellenberg 1988; Lindacher 1995). Ellenberg values are available for a maximum of 778 (95%) species in Middlesex and 870 (97%) species in Cambridgeshire, and the results are therefore based on these species. The values refer to the behaviour of species in central Europe. Species which have a different habitat in central Europe from that in eastern England will be misclassified from a British perspective but there are relatively few examples of such differences (some are discussed in the section on vegetation below). A modified set of Ellenberg indicator values, corrected where appropriate and applied to all British and Irish species, has recently been prepared and will be available for future analyses (Hill er al. 1999). PLANT HEIGHT Data on the typical maximum height of a species have been used to group terrestrial species into eight height categories. (Submerged and floating aquatics are excluded from this analysis.) These categories are those defined for canopy height by Grime ef al. (1988). In both Middlesex and Cambridgeshire the proportion of short species which have become extinct greatly exceeds the proportion of taller species (Table 3, Fig. 1). TABLE 3. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) RELATED TO PLANT HEIGHT AND RAUNKIAER LIFE FORM The totals for plant height exclude aquatic species. Species which occur as two Raunkiaer life- forms are scored as (} 5 under each. Middlesex (v.c. 21) Cambridgeshire (v.c. 29) no. spp. no. extinct % extinct no. spp. no. extinct % extinct Plant height (cm) 0-10 17 5 29 16 5 3] 11-29 78 24 3] 88 ZA 24 30-59 204 46 23 234 42 18 60-99 209 30 14 224 22 10 100-300 220 28 13 243 19 8 301-600 ibe) 7 pe 0 0 601-1500 13 0 0 12 0 0 >1500 16 0 0 16 0 0 Total 772 134 iy 848 109 13 Life form Bulbous geophytes 8 3 38 8 2 2a Other geophytes 52 12 2 64-5 15 de) Hemicryptophytes S275 47 14 370-5 35 9 Chamaephytes ot 8 22 46 7:5 16 Nanophanerophytes a2 7 22 32 =) 16 Larger phanerophytes 38-5 0 0 37:5 0 0 Epiphytes | 0 0 ] 0 0 Therophytes 197 43 22 ZA TS 30 14 Helophytes 52 12 23 61-5 12 20 Hydrophytes 61 14 23 64-5 13-5 ZA Total 816 146 18 897 120 13 EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 63 Height Light ho r=) % extinet vy extinet i [=] | a 4 2 3 4 5 6 7 8 | 2 3 4 5 6 7 8 9g Height class Ellenberg L value Middlesex Middlesex Oo % extinet % extinet meee CU [ 2k Sb eG Ellenberg L value Height class Cambridgeshire Cambridgeshire FIGURE |. The proportion of extinct species in Middlesex and Cambridgeshire in relation to typical maximum plant height classes and to Ellenberg light values. Plant height increases from class | to 8 (see Table 3) and light requirements increase from value | to 9 (see Table 4). In both cases x represents species with a wide amplitude. LIFE-FORM The Raunkiaer life-form system classifies species by the height of the winter buds rather than the summer foliage (Clapham er al. 1987). Geophytes have resting buds below ground level, hemicryptophytes at ground level, chamaephytes up to 25 cm above the soil surface and phanerophytes over 25 cm above the ground. For the purposes of this analysis, nanophanerophytes (with resting buds 25 cm to 2 m above soil level) are separated from larger phanerophytes (which have resting buds more than 2 m above the ground). In the Raunkiaer system annuals (therophytes) form a separate category, as they pass through the unfavourable season as seeds. Marsh plants (helophytes), water plants (hydrophytes) and epiphytes are also treated separately. The overall life- form spectrum and the proportion of species in each life-form which have become extinct are very similar in the two counties (Table 3). A disproportionate number of geophytes, chamaephytes and nanophanerophytes have become extinct, and in Middlesex therophytes show a greater extinction than average. Extinctions in the wetland groups also exceed the overall average. The proportion of extinctions is below average for the hemicryptophytes, the preponderant life-form in both counties, and there have been no extinctions amongst the larger phanerophytes in either county. The single epiphyte, the hemiparasitic Viscum album, has survived in both areas. LIGHT Ellenberg L values for light range from 1 (plants of deep shade) to 9 (plants only found in full light); there is also a category for species with a wide amplitude (x). The values for species in Middlesex and Cambridgeshire (Table 4, Fig. 1) show that the light-loving species in categories 8 and 9 have suffered a disproportionate number of extinctions. In Cambridgeshire there is also a high level of extinction in one group of shade plants (category 3), attributable to the extinction of five woodland species (Blechnum spicant, Carex strigosa, Equisetum sylvaticum, Lathraea squamaria, Polystichum aculeatum) in this small group. Rather surprisingly, the small group of species of wide amplitude (x) has also lost a disproportionate number of species. This group includes ecologically tolerant species such as Anemone nemorosa, Festuca rubra and Urtica dioica which have survived in both counties, but also parasitic or hemiparasitic species such as Cuscuta epithymum, Melampyrum pratense and Orobanche rapum-genistae which contribute almost all the extinctions. 64 C2 D: PRESTON TABLE 4. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) RELATED TO ELLENBERG VALUES FOR LIGHT, WATER, pH AND NITROGEN Ellenberg value Middlesex (v.c. 21) Cambridgeshire (v.c. 29) no. spp. no. extinct % extinct no. spp. no. extinct % extinct Light 1 Deep shade i 0 0 1 0 0 2 8 l 13 8 1 1 3 Shade 20 ys 10 18 5 28 4 38 3 8 37 - 11 5 Half shade A7 - . 49 3 6 6 122 14 Lt 123 5 4 7 Partial shade/full light 0 be) AO 15 313 30 10 8 Light-loving 202 al 25 250 43 18 9 Full light only 51 14 ZI 69 i19/ 25 x Wide amplitude 14 3 Zz 14 o 29 Total 778 132 17 869 112. 13 Water 1 Extremely dry 0 0 - 0 0 - 2 16 . Zz 20 1 5 3 Dry 46 14 30 66 - 6 4 143 a | 19 166 18 11 5 Moist 168 15 9 169 13 8 6 68 a 6 73 12 16 7 Damp 58 15 26 65 12 18 8 78 16 21 87 12 14 9 Wet 66 20 30 80 24 30 10 Occasionally flooded 36 6 17 39 6 15 11 Emergent or floating 19 - 21 eA 2 10 12 Submerged 24 | 21 28 i, 25 x Wide amplitude 54 2 - 54 2 ~ Total 776 132 17 868 113 13 pH | Extremely acid 8 5 63 8 a 88 2 24 5 pA 24 9 38 3 Acid 51 18 35 a2 PA AO a 46 11 24 A8 6 13 5 Fairly acid 48 7 15 53 i 13 6 71 8 i 77 aT 9 7 Weakly acid or weakly basic 210 32 15 ZN. 24 10 8 126 24 19 160 18 11 9 Calcareous 15 - at 26 3 2 x Wide amplitude 176 16 9 183 10 5 Total TD 130 17 868 112 13 Nitrogen | Nitrogen-poor 33 11 33 ae 13 35 2 87 29 a8 122 35 29 3 Prefers nitrogen-deficient 89 : 24 ra | 104 17 16 - 90 13 14 96 10 10 5 Average 109 ig 16 128 13 10 6 107 8 7 110 8 7. 7 Prefers nitrogen-rich 105 13 12 jig bs 7 6 8 i 5 7 70 3 4 9 Extremely nitrogen-rich 21 3 14 2A 3 14 x Wide amplitude 59 7 he 61 3 5 Total : vig 130 LZ, 863 112 13 EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 65 WATER Ellenberg F values for water range from | (indicators of extreme dryness) to 12 (submerged plants); there is also a category for species with a wide amplitude (x). The values for species in Middlesex and Cambridgeshire are set out in Table 4. None of the species are classified as indicators of extreme dryness. In Middlesex more extinctions than average have taken place in a wide range of groups, especially those of dry-site (values 2, 3) and damp- or wet-site (7, 8, 9) species. Extinct species of dry sites include some calcicoles (e.g. Helianthemum nummularium, Scabiosa columbaria) but most are calcifuges (e.g. Potentilla argentea, Teesdalia nudicaulis). The extinct species in groups 7-9 (e.g. Anagallis minima, Carex binervis, C. dioica, Cicuta virosa, Cirsium dissectum, Drosera rotundifolia, Mentha pulegium, Oenanthe lachenalii, Parnassia palustris) would have been found in a wide range of damp or wet habitats. The lowest rates of extinction are in the species of moist sites (5, 6), large categories which include many woodland species, and those with a wide amplitude (x). In Cambridgeshire, as in Middlesex, there is a particularly high proportion of extinctions in the wet-site species (9) and a particularly low proportion in the species of wide amplitude (x). However, few species in the driest sites (2, 3) have become extinct. pH Ellenberg R values for pH range from | (indicators of extreme acidity) to 9 (always found on calcareous soils); there is also a category for species with a wide amplitude (x). Both mm Middlesex and more especially in Cambridgeshire (Table 4, Fig. 2) there has been a massive loss of species which indicate extreme acidity (value 1). The same eight species were originally present in both counties. Calluna vulgaris is the only one to survive in Cambridgeshire, whereas Juncus squarrosus and Pedicularis sylvatica have also survived in Middlesex; Carex binervis, Drosera rotundifolia, Erica tetralix, Teesdalia nudicaulis and Trichophorum cespitosum are extinct in both. The extinctions in the other acidic categories (2, 3) are also well above the mean, especially in Cambridgeshire. In Middlesex the species of basic habitats (9) also show a high level of extinction, but this is not the case in Cambridgeshire. In both counties the lowest proportion of extinct species is in the wide-amplitude category (x). NITROGEN Ellenberg N values for nitrogen range from | (indicators of sites poor in available nitrogen) to 9 (in extremely rich situations such as areas where cattle congregate); there is also a category for species with a wide amplitude (x). The proportion of extinct species in Cambridgeshire and Middlesex (Table 4, Fig. 2) is highly correlated with nitrogen value, especially in Cambridgeshire: 33-35% of the indicators of sites poor in available nitrogen (value 1) have become extinct, compared to 4—-7% of species characterised by value 8. The exception to this trend is the small group of species characteristic of extremely rich situations (value 9), which show a higher proportion of extinctions than those species of somewhat less rich soils. The extinct species are Beta vulgaris, a coastal plant, and Bidens cernua, Chenopodium glaucum and C. vulvaria. VEGETATION Ellenberg’s (1988) classification of species by the characteristic vegetation in which they grow is based on phytosociological categories, with each species classified by its occurrence in one of eight major groups. The proportion of extinct species in each vegetation type is again similar in the two counties (Table 5). The marked losses of coastal and wetland species in both counties is evident, whereas species of wide tolerance, broadleaved woods and (in Cambridgeshire) scrub and woodland edges show the least losses. The tiny group of conifer woodland species shows the highest percentage loss, but of the four species which fall into this category only one (Pyrola minor) appears from a British perspective to be correctly classified; two of the other three (Blechnum_ spicant, Monotropa hypopitys) could equally well be classified as species of broadleaved woodland and the fourth, Carex ericetorum, is a continental species which in Britain is confined to calcareous grassland. As both counties have lacked native conifer woodland in historic times they would not be expected to possess native species typical of this habitat. Pyrola minor may have colonised the county naturally by wind-borne seed, but failed to persist. In Table 5 the species in the three largest vegetation types are subdivided by phytosociological class. This shows marked variation in the proportion of extinct species in different aquatic and wetland classes. The few species of bogs (Oxycocco-Sphagnetea) are all extinct in both counties and the larger groups typical of acidic mires (Scheuchzerio-Caricetea nigrae) and shallow, acidic 66 C. D. PRESTON pH Nitrogen “% extinct % extinct | 2 3 4 5 6 £ 8 g 1 7 3 4 5 6 7 8 9 Ellenberg R value Ellenberg N value Middlesex Middlesex % extinct % extinct { 2: @ 4 36 °@ 7 “8 o ox 1, 2 St AL Sh, Oe ee Ellenberg R value Ellenberg N value Cambridgeshire Cambridgeshine FIGURE 2. The proportion of extinct species in Middlesex and Cambridgeshire in relation to Ellenberg pH and nitrogen values. As Ellenberg pH values increase the species become increasingly characteristic of base-rich soils; nitrogen requirements increase from Ellenberg value | to 9 (see Table 4). In both cases x represents species with a wide amplitude. waters (Littorelletea) also show disproportionate extinction. In contrast, species in communities characteristic of more basic and more eutrophic waters (Potamogetonetea, Phragmitetea) tend to have survived. There are no marked differences between the two counties. The disturbed ground community which has lost most species is the Isoéto-Nanojuncetea, found on seasonally flooded terrain. Otherwise the different communities of disturbed ground show less variation in the proportion of extinct species within each county. The main differences between counties are the higher percentage of species in the Secalietea (arable weeds) and Agrostietea stoloniferae (pioneer plants of damp or flooded sites) which have become extinct in Middlesex. There are also marked differences between the two counties in the plants of heaths, meadows and pastures. In Cambridgeshire the greatest losses have been amongst the calcifuge plants in the Nardo- Callunetea, whereas in Middlesex the calcicolous Festuco-Brometea has lost most species. Losses in the Molinio-Arrhenatheretea, which contains many species characteristic of moister and often more nutrient-rich soils, have been low, especially in Cambridgeshire. DISTRIBUTION AND PHYTOGEOGRAPHY OF EXTINCT SPECIES DISTRIBUTION WITHIN THE COUNTIES Have the species which have become extinct in Middlesex and Cambridgeshire always been rare, or have more widespread species become extinct as well? Answering this question is not straightforward, as distributional data are presented in different ways in the Floras of the two counties. In Middlesex, Kent (1975 and in press) followed Trimen & Dyer (1869) in dividing the county into seven divisions. In Cambridgeshire, Perring et al. (1964) and Crompton & Whitehouse (1983) list records in 10-km squares. To make the Cambridgeshire data comparable to those available for Middlesex, seven divisions based on blocks of 5 to 6 10-km squares were defined (Appendix 2). Thése were broadly based on the eight divisions defined by Babington (1860), with two of the smaller divisions amalgamated and the areas redefined in terms of 10-km squares. Details of the number of divisions in which the native and extinct native species have been EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 67 recorded are provided in Table 6. If a species is recorded as a native in some divisions and as an alien in others, only the native occurrences have been counted, if possible. However, the introduced or even casual occurrences of some species are not separable from the native records. In Middlesex over 50% of extinct species were recorded from only |—2 divisions and over 80% from 1 divisions. In Cambridgeshire the tendency of rarer species to become extinct is even more marked, with over 50% of extinct species recorded from a single division and over 80% from 1-3 divisions. Despite this clear general trend some widespread species have become extinct, especially in Middlesex. Seven of the nine species which were recorded from all seven divisions in v.c. 21 but are now extinct are arable or other weeds (Anchusa arvensis, Anthemis cotula, Chenopodium vulvaria, Chrysanthemum segetum, Lithospermum arvense, Ranunculus arvensis, Scandix pecten-veneris), the exceptions being Ranunculus hederaceus and Spiranthes spiralis. The only extinct Cambridgeshire species to have been recorded from all divisions is Marrubium vulgare. TABLE 5. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) RELATED TO -TYPICAL VEGETATION TYPE Vegetation type Middlesex (v.c. 21) Cambridgeshire (v.c. 29) no. extinct % extinct no. extinct % extinct no. spp. spp. spp. no. spp. spp. spp. 1 Freshwater and mires 133 33 pas) 155 oy 24 1.1 Lemnetea 6 1 17 & i tps) 1.2 Utricularietea i i 100 2 Z 100 1.3 Potamogetonetea 5 f) 6 7, 33 3 9 1.4 Littorelletea | 6 Ds) 10 6 60 1.5 Phragmitetea 4§ 6 L3 D2 2 + 1.6 Montio-Cardaminetea 7 1 14 7. Z 29 1.7 Scheuchzerio-Caricetea nigrae 2 9 43 SE) IS 43 1.8 Oxycocco-Sphagnetea 3 5) 100 D 2) 100 2 Saltwater and sea coasts 5) py) 40 28 10 36 3 Frequently disturbed sites 2D 37 16 231 24 10 3.1 Isoéto-Nanojuncetea 14 4 29 1 Bi ©) 38 3.2 Bidentetea 19 2 Dy 19 3 16 3.3 Chenopodietea 52 3} 10 >)! Z d 3.4 Secalietea 34 9 26 45 ) ne 3.5 Artemisietea 59 9 tS D7: 3 e) 3.6 Agropyretea 6 1 IPE 6 if TZ 3.7 Plantaginetea 7; 0 0 ve 1 14 3.8 Agrostietea stoloniferae 24 ii 29 27. | bl 4 Stony sites and walls 8 | 13 8 | 13 5 Heaths and grasslands 189 B)5) 19 DAS 23 11 5.1 Nardo-Callunetea 25 6 24 PAE it 4] 5.2 Sedo-Scleranthetea 36 & 22 40 4 10 5.3 Festuco-Brometea 30 bk GWE 4§ 5) 6 5.4 Molinio-Arrhenatheretea 59 10 ilel 9] 5) ») 6 Scrub and wood edges 27 4 15 32 Z 6 7 Conifer woods and allied heaths 2 1 50 3 yy 67 8 Broadleaved woods 120 11 9 124 9 4 x Wide amplitude 64 6 !) 63 3 5) Total 773 130 17 855 sila 13 68 C. D. PRESTON TABLE 6. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) RELATED TO THE NUMBER OF DIVISIONS OF THE COUNTY FROM WHICH THEY WERE RECORDED Number of divisions Middlesex (v.c. 21) Cambridgeshire (v.c. 29) no. spp. no. extinct % extinct no. spp. no. extinct % extinct Spp.- Spp. Spp. Spp. | 71 54 76 109 61 56 2 36 20 56 56 22 39 3 37 22 59 Wp 18 25 4 66 29 38 72 11 15 5 58 7 2 88 6 I 6 83 9 ie 93 1 i 7 465 9 2 408 1 <1 Total 816 146 18 897 120 13 For details of the divisions, see Kent (1975) for Middlesex and Appendix 2 of this paper fc Cambridgeshire. TABLE 7. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) RELATED TO NATIONAL RARITY National rarity Middlesex (v.c. 21) Cambridgeshire (v.c. 29) no. spp. no. extinct % extinct no. spp. no. extinct % extinct SPP. Spp. SPP. SPP. Extinct l l 100 3 3 100 Rare De 20 | Sal 13 42 Scarce 46 ps) 63 67 23 34 Other 747 96 13 796 81 10 Total 816 146 18 S97 120 13 There are two reasons why these results should be treated with some caution. It is doubtfi whether recording of species in the 17th, 18th and early 19th centuries was sufficiently thorough | ensure that species which became extinct at an early date were recorded in all the districts in whic they grew. There may therefore be a tendency for extinct species to appear to have been le: widespread than they were, although the habitat requirements of many of the extinct specie suggest that they would always have been rare and restricted. In any case, the pattern in bo! counties is so clear that one has to conclude that the rarer a species was originally, the more like. it is to have become extinct. The species which appear to have been widespread are mainly weed and the records in some divisions may have been only casual occurrences. RARITY OF SPECIES IN BRITAIN : Three terms are conventionally used to describe the rarest species in Britain: extinct, rare (prese in | to 15 10-km squares) and scarce (present in 16 to 100 squares). The native and extinct nati species in the two counties are classified in these categories in Table 7; details of the nation status are taken from Stewart et al. (1994) and Wigginton (1999). One species formerly recorde in Middlesex (Arnoseris minima) and three formerly found in Cambridgeshire (Arnoseris minim Bromus interruptus, Tephroseris palustris) are now extinct in Britain as a whole. Of 22 ra species which have been recorded from Middlesex only two survive (Arabis glabra, Cyper fuscus) and over 60% of the scarce species have been lost. A smaller proportion of nationally ra and scarce species has been lost in Cambridgeshire, although even in Cambridgeshire losses these species have been much greater than those of species which are more frequent in Britain as whole. | EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 69 Whereas the analysis of the rarity of species within the counties is based on all records of the species, this analysis is based on current national status. It therefore contains an element of circularity, as species may be rare or scarce because they have declined in counties such as Middlesex and Cambridgeshire. However, it does at least demonstrate these changes at the local scale. PHYTOGEOGRAPHY Preston & Hill (1997) classify native species according to their latitudinal and longitudinal distribution in the northern hemisphere. The extinct flora of both counties is analysed in relation to these phytogeographical elements in Table 8. There is a clear relationship between the phytogeographical elements and the proportion of extinct species in both counties. The greatest proportion of extinctions has taken place in the small Boreal and Oceanic elements. A single Boreo-arctic species (Carex dioica) occurred in both counties, but has now been lost. In Middlesex five out of six Boreal species have been lost; Vaccinium myrtillus survives but is reduced to a single plant on Hampstead Heath (R. M. Burton, in litt. 1999). Similarly eight of the 13 Boreal species in Cambridgeshire have become extinct; the surviving species are Carex lasiocarpa, Coeloglossum viride, Potamogeton alpinus, P. praelongus and Salix myrsinifolia. Although less severe, losses of Oceanic and Suboceanic species are also much greater than the average values. TABLE 8. EXTINCT SPECIES IN MIDDLESEX (V.C. 21) AND CAMBRIDGESHIRE (V.C. 29) INRELATION TO PHYTOGEOGRAPHICAL AFFINITIES The percentage of extinct species in each phytogeographical group is given. The values for groups containing less than 11 species in total are bracketed. For details of the phytogeographical classification, see Preston & Hill (1997). Eastern limit category Major | 2 3 4 5 6 biome category Oceanic Suboceanic European Eurosiberian Eurasian Circumpolar Total Middlesex 1 Arctic-montane - - : - - - - 2 Boreo-arctic Montane - - - : - (100) (100) 3 Wide-boreal - - (O) (20) (18) - (100) (0) - (100) (80) 4 Boreal-montane - 5 Boreo-temperate (100) (67) 14 12 19 16 17 6 Wide-temperate - - (QO) 0 (O) (O) 0 7 Temperate oi 30 16 18 14 6 18 8 Southern-temperate (20) 35 17 16 0 (10) 18 9 Mediterranean (44) 8 - - - - 20 Total 50 28 16 15 13 16 18 Cambridgeshire 1 Arctic-montane . - - - - - - 2 Boreo-arctic Montane - - . - - (100) (100) 3 Wide-boreal - - - (O) (0) 25 21 4 Boreal-montane - - (100) (O) - 64 62 5 Boreo-temperate (50) (50) 19 ») 16 24 16 6 Wide-temperate - - - 0 (0) 1 17 7 Temperate 33 By) ) 7 10 13 11 8 Southern-temperate (60) 26 8 8 0 0 11 9 Mediterranean (3) 7 - - - - 21 Total 39 Zi 10 6 10 8) 13 70 CD. PRESTON EXTINCTIONS AND LAND-USE CHANGE Although the ecological and phytogeographical characteristics of the extinct species of Cambridgeshire and Middlesex are broadly similar, the above analysis revealed marked differences in the timing of the extinctions. The differences are particularly marked in the 19th and early 20th centuries: extinctions in Cambridgeshire exceeded those in Middlesex between 1810 and 1839, whereas the rate in Middlesex was markedly greater from 1870 to 1929. There is no obvious explanation of these figures in terms of recording intensity: the recording history of both counties is broadly similar, and a major Flora of each was published in the 1860s. It is therefore worth seeking an explanation of these differences in terms of the land-use history of both counties. CAMBRIDGESHIRE Cambridgeshire has always been a predominantly agricultural county, and the land-use history which is most relevant to the botanist is the history of its agriculture. The county lies in one of the areas of Britain which are most suitable for cereal growing, and, once drained, the fenland soils support a range of arable crops. Only the clays of the west were once more marginal as agricultural land. The period of just over a century from 1640 to 1750, the first for which numerous detailed botanical records are available, was one in which agriculture was in relative decline in Britain. Agricultural output increased much more rapidly than did the population. Agricultural prices, which had increased six-fold between 1500 and 1640, increased by only 2% between 1640 and 1750 (Jackson 1985; Thirsk 1984). Farmers attempted to compensate for this by growing new crops, and there was a boom in ‘alternative agriculture’ (Thirsk 1997). It was during this period that drainage of the “Great Level’ of the Fens was accomplished in Cambridgeshire (Darby 1956; Taylor 1973). The Bedford Level was declared drained in 1652 and Soham Mere, a large Fenland lake, was drained, enclosed and brought into cultivation in 1664. The fact that ‘alternative’ crops such as flax, hemp, rapeseed and vegetables could be grown in the newly drained fenland perhaps encouraged these schemes for agricultural improvement. Continued technical problems resulted in the deterioration of some of the drained land, especially in the first half of the 18th century when the incentive for capital investment was low, but these problems were eventually solved by the application of steam power to pumping engines. From 1750 onwards the growth in agricultural output in England and Wales slowed, whereas population grew rapidly, from 6 million in 1750 to 18 million in 1850. Improvements in transport made it easier for farmers to get their products to the growing urban centres (Mingay 1989). Consequently there was a prolonged agricultural boom. As agriculture prospered, farmers brought - more and more ‘waste’ into intensive cultivation. At first the increase in the area of cultivated grassland exceeded that of arable, but after 1800 arable expanded more rapidly and by 1840 the acreage of arable in England and Wales actually exceeded that of pasture (Prince 1989). Interest in alternative crops waned as the rewards of mainstream farming increased. In Cambridgeshire the medieval open field system survived in many parishes until the end of the 18th century. The first great wave of parliamentary enclosure in the 1760s and 1770s scarcely affected the county. However, this was not true of the second period when enclosure was “particularly rampant’, between 1793 and 1815. During this period bad harvests and insecurity engendered by the Revolutionary and Napoleonic Wars led to great pressure for agricultural improvement (Turner 1980). A ‘general view’ of the agriculture of the county prepared by Charles Vancouver (1794) for the newly established Board of Agriculture provided detailed and forceful arguments for agricultural reform (cf. Grigg 1967). Vancouver’s report was followed by a period in which many parishes were enclosed. Although the rate of enclosure slowed markedly after the: end of war with France in 1815, as it did elsewhere in England, there was a period of renewed activity in the 1820s and 1830s, a feature virtually unique to the county (Turner 1980). Cambridgeshire is described by Turner as “the county of parliamentary enclosure”. A higher percentage of land (53%) was enclosed than in any other county except Oxfordshire (54%), but the Cambridgeshire figure is reduced by the fact that there was little parliamentary enclosure in the northern, fenland areas. In the south of the county over 70% of the land was enclosed, representing “the most concentrated agricultural organisational change that there had ever been, certainly in that county and probably in any county” (Turner 1980). EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE Til The enclosures in Cambridgeshire were followed by the widespread agricultural improvement that Vancouver had recommended. Enclosure had a marked effect on the flora of the county because in its immediate aftermath semi-natural habitats were taken into cultivation: fens were drained and chalk grassland and heathland ploughed up and converted to arable land. These reforms provide an explanation of the large number of extinctions between 1810 and 1839. Corroborating evidence is not hard to find. The Journal of Natural History kept by the Rev. Leonard Jenyns from 1823 to 1846 reveals “ a recurring concern about the effects of Parliamentary Enclosures on the extent of heathland and wetland” (Crompton 1997). Babington (1860) had no doubt why he was unable to refind some species at localities known to Relhan before 1820. Drosera anglica at Sawston and Hinton Moors and D. intermedia at Teversham and Sawston Moors had “not [been] found since these places were drained”, Pulsatilla vulgaris and Ulex minor formerly grew on Barrington Hill, Linton, “a place now ploughed up”, and Vicia sylvatica had been recorded at Hall Wood, Wood Ditton, which “does not now exist’. There is no evidence from the study of extinctions to suggest that the intensification of agriculture on existing arable land had a marked effect on the flora of the county at this period. The introductory comments in Babington’s Flora (1860) are often cited. In reviewing the county habitat by habitat, he cited major changes which had happened to each. The Chalk Country which “until recently (within 60 years)” was “open and covered with a beautiful coating of turf...is now converted to arable land” and “even the tumuli, entrenchments, and other interesting works of the ancient inhabitants have seldom escaped the rapacity of the modern agriculturist”. The plants of the Clayey District “have suffered nearly as much”, and the Fens “have undergone an equally if not more destructive change than the Chalk district” as steam drainage had rendered the whole level “a pattern in farming”. The statistics on extinction allow these statements to be seen in their historical perspective. They are not the exaggerations of a congenitally conservative don, but a reaction to the effects of a prolonged period of agricultural prosperity and expansion and in particular a marked increase in the proportion of arable land. This had produced effects on the native flora which no generation of botanists in Britain had hitherto experienced. Babington’s Flora of Cambridgeshire is the first local flora in which environmental change is a major theme. The agricultural prosperity of the 18th and 19th centuries ended relatively abruptly in the 1870s, when an increasing amount of grain and other agricultural produce became available on world markets (Perren 1995). The resulting agricultural depression lasted with minor interruptions until 1940. By 1880 the area of arable in England and Wales was again less than that of pasture, and it continued to fall so that by 1914 East Anglia was the only region in which the proportion of arable consistently exceeded that of pasture (Whetham 1978). The better land in Cambridgeshire escaped the worst of the depression, and the agriculture of the county remained overwhelmingly arable (Fig. 3). However, by the 1930s much of the permanent pasture in the western claylands was “derelict or nearly so” and the view taken of farming in the area was “necessarily gloomy” (Pettit 1941). This prolonged depression ended with the Second World War and the subsequent Agriculture Act of 1947, which guaranteed agricultural prices and markets. Further support was offered to farmers under the Common Agricultural Policy after Britain’s entry into the European Economic Community in 1973. Financial prosperity was combined with technological advances to produce a prolonged period of agricultural intensification. The resulting changes in eastern England included the expansion and intensification of arable farming, with the removal of hedges to increase field sizes, a switch from spring to autumn sowing of crops, the use of more productive cereal varieties and increased applications of fertilisers, herbicides and pesticides. Remaining areas of grassland were also converted from species-rich semi-natural communities to intensively managed, heavily fertilised, species-poor swards (Stoate 1996). A detailed account of the changes in one Cambridgeshire parish during this period is given by Sell (1989). The rate of extinction of species in Cambridgeshire was very low between 1840 and 1899. This doubtless reflects the completion of agricultural enclosure followed by the end of the long period of agricultural prosperity. It may also be influenced by the fact that there was little systematic plant recording in southern Cambridgeshire after the completion of Babington’s Flora. The rate of extinction increased in the period from 1900 and 1929, and it further increased from 1930 onwards. The increase from 1900 to 1929 has no clear explanation in terms of national land-use history, but the later increase coincides with the post-war agricultural revolution. The two decades in the 20th century in which most species were lost were the 1950s (twelve species) and the 1960s (eleven species), although it could be argued that species which were last seen in this period are 2 C. D. PRESTON 100 50 = & i = 60 7 = | Non-agricultural op 1 Grass = 40 @ Arable YS o all 20 @) - (oO Lo) LY) LO LO LO LO LO) LO) LO LO ce) ~~ co Oo O = —N fo8) wT ire) © co co co co Oo oO) >) Oo (>) Oo [S)) Year FIGURE 3. The percentage of arable land, permanent grassland (including rough grazing) and non-agricultural land in the administrative county (or counties) of Cambridgeshire and the Isle of Ely, 1866-1970. Values are plotted for 1866 and then at 5-year intervals from 1870 to 1970. They are based on official agricultural returns extracted by Stamp (1941) for the period between 1866 and 1910 and by C.D.P. from 1915 onwards. Radical revisions to the administrative county boundary took place in 1974. more likely to be rediscovered than those last seen in earlier decades. In some cases the extinctions which have occurred since 1930 may have been the long-delayed result of earlier changes: Potamogeton polygonifolius was last seen at Gamlingay in 1948, for example, long after the extinction of the more exacting calcifuges there. In other cases the loss of species can be attributed directly to agricultural intensification, such as the ploughing up of the county’s only site for Cirsium tuberosum in 1973 (Pigott 1988). However, most extinctions are probably the result of less dramatic, gradual changes brought about by factors such as eutrophication and falling water tables. MIDDLESEX The periods of maximum extinction in Cambridgeshire are closely related to periods of agricultural improvement. Can the same relationship between the number of extinct species and land-use history be demonstrated in Middlesex? Until it was swamped by the growth of London, Middlesex, like Cambridgeshire, was an almost exclusively agricultural county (Tanner 1911). The more fertile soils are suitable for arable cultivation, and in Elizabethan times this land, “fat and fertile and full of profite’, produced “most excellent good wheate” (Norden 1593). However, although arable cultivation persisted on the best soils, agriculture in the county became predominantly pastoral between 1550 and 1750 in response to the insatiable demand for hay in London. Thus Perivale, which was praised for its wheat by Norden and in Michael Drayton’s poem Poly-Olbion (1612), was to be remembered by John Betjeman in his poem Middlesex as a “parish of enormous hayfields” (Betjeman 1954). Hay was required for the city’s horses and for cows kept in urban cowsheds; the name Haymarket (Piccadilly) survives from one of four new hay markets set up in the city in the 17th century. The refuse of the city provided manure which was returned to the countryside to sustain production. Close to London extensive areas of market gardens developed, so that the area west of St Martin-in-the-Fields appeared to be “a continuous garden” (Richardson 1984). Pehr Kalm, visiting London in 1748, explained that “the frightful number of people which there crawl in the streets pays the market gardeners many fold their labour and outlay” (Lucas 1892). A detailed account of the market gardens of Middlesex, and of other aspects of the changing land-use of the county, is given by Bull (1957). EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 13 Some of the open countryside of Middlesex was lost before Parliamentary enclosure to non- agricultural uses such as roads, parks and buildings. Parliamentary enclosure took place earlier than in Cambridgeshire: 28% of the total area enclosed was dealt with by 1793 and 91% by 1815, compared to 6% and 60% in Cambridgeshire (Turner 1980). The proportion of the total area of Middlesex subject to enclosure, 28%, was much less than the 53% figure for Cambridgeshire. A notable feature of the enclosure of Middlesex was the conversion of commons and wastes, including much heathland, to agricultural use. The agricultural improvers of the late 18th century regarded the presence so close to the capital of unimproved land such as Hounslow Heath as “disgraceful to the county, and insulting to the inhabitants of the metropolis” (Middleton 1798). The enclosure of these commons was the culmination of a long process of attrition that reduced the area of common land from approximately 45,000 acres in 1500 to 30,000 acres in 1760; most of these 30,000 acres had been enclosed by 1825 (Kent 1975). After enclosure the assured profits obtained by catering for the predictable needs of London led to an agricultural regime which appears to have been relatively stable. In Trimen & Dyer’s (1869) Flora the farmers of Middlesex bear no relationship to the rapacious agriculturalists of Babington’s Cambridgeshire. Basing their account on Clutterbuck (1869), they describe the hayfields over London clay as “small and not very convenient enclosures, sometimes overgrown with timber in the hedgerows’. Clutterbuck (1869) applied Norden’s (1593) words to the hay farmers of the 1860s: “they only covet to maintaine their ancient course of life and observe the husbandrie of their fathers without adding any thing to their greater profit”. By 1869, however, it was apparent to Trimen & Dyer that the market gardens and orchards close to London were “giving way to the advancing wave of buildings more rapidly than they are replaced”. The population of London was approximately 400,000 in 1650, 575,000 in 1700 (when it overtook Paris as the largest city in Europe), 675,500 in 1700 and 900,000 in 1800 (Wrigley 1967). But in the 19th century its population more than trebled in size, reaching 3-56 million in 1901 and growing further to 5 million by 1961. It was in the second half of the 19th century and more particularly the first half of the 20th that the large town expanded to become the “immense continuous urban area” of Greater London (Figs 4, 5). 100 - 80 60 (1 Non-agricultural C1 Grass @ Arable 40 Percentage of area 20 - 0 © LY} “ly Ly} Ly Lo) Lo} is) Lt} “uy vt i] (ey) GO (ey) Da 5 ia ba L, Tae a 5 a eee — ae 2 ae bi, ia Year FIGURE 4. The percentage of arable land, permanent grassland (including rough grazing) and non-agricultural land in the administrative counties of Middlesex (1866-1891) and Middlesex plus London (1892-1964). Values are plotted for 1866 and then at 5-year intervals from 1870 to 1960 and finally for 1964. They are based on official agricultural returns extracted by Willatts (1937) for the period between 1866 and 1935 and by C.D.P. from 1940 onwards. From 1949 onwards the agricultural land in the county of London was apportioned between Kent, Middlesex and Surrey rather than included in the Middlesex total, but by then it represented less than 1.5% of the total area of the county of London so that the change has no appreciable effect on the diagram. 74 C. D. PRESTON 6000 50G0 4000 3000 2000 1000 People per square kilometre 0 be — se B pees —_— ie = i aad bo L ics © N w [) (o>) (o>) [e>) [@>) = bs ae = ed rT raed boa geo bd b ae Year FIGURE 5. Population density (people per square kilometre) in Middlesex (M), extra-metropolitan Middlesex (EM) and Cambridgeshire (C), derived from decadal census returns from 1801 onwards. There was no enumeration in 1941 and this value was replaced by a retrospective estimate for 1939. Figures for Middlesex refer to the ‘ancient geographical county’, in effect the vice-county, until 1901. Subsequently the Middlesex figures are derived by adding the values for the revised administrative county of Middlesex to the figures for those areas of the county of London which were formerly in Middlesex (the City of London and 18 metropolitan boroughs). It is not possible to derive an estimate for 1939 by this method and the value plotted is therefore the mean of the 1931 and 1951 values. Extra-metropolitan Middlesex is the area of the county outside the former county of London; retrospective population estimates for extra-metropolitan Middlesex were published with the 1961 census returns. Figures for Cambridgeshire refer to the administrative county until 1971, and subsequently to four districts (Cambridge, E. Cambs., S. Cambs., Fenland) which approximate to the administrative county. The administrative and vice-counties are broadly similar (Crompton & Whitehouse 1983). The expansion of London was made possible by improved rail and then road transport, and by the ability of engineers to solve the water supply problems which had hampered urban development on the London Clay (Smailes 1964; Willatts 1937). The growth of London “resulted in the greater part of the vice-county being virtually obliterated by a huge torrent of bricks and mortar and the creation of vast new suburbs which engulfed most of the old Middlesex villages and almost eradicated the London Clay grass plain... The rapid spread of suburbia has so reduced the agricultural areas of the vice-county that today only a minimum remains” (Kent 1975). In addition to the spread of suburbia, the county has been affected by other changes including gravel extraction from the river valleys, which has been directly responsible for the extinction of some species, and the growth of Heathrow Airport. The land-use history of the county therefore ties in well with the statistics on extinctions. The initially higher rate of extinction when compared to Cambridgeshire is explicable in terms of early agricultural change, but during the period 1810—1839 the rate fell behind as the intense double wave of enclosures in Cambridgeshire was not matched by such radical developments in Middlesex. Thereafter the extinction rate in Middlesex picked up again, reflecting the growth of London: the primary effect of habitat destruction and all the secondary consequences of urbanisation. EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE es) DISCUSSION ACCURACY OF THE LISTS The analysis in this paper rests on lists of the native and extinct native species of Middlesex and Cambridgeshire. In some cases where the evidence of the occurrence of a species rests solely on old records there is an element of doubt about the correct identification of the species or its geographical location. Limonium bellidifolium, for example, is traditionally included in Floras of Cambridgeshire on the basis of a locality discovered by W. Skrimshire, the salt marshes below Wisbech, but three vice-counties meet in this area and there is considerable doubt as to whether the plant actually grew in Cambridgeshire (G. Crompton, in litt. 1999); the species is therefore excluded from this paper. The flora of both Middlesex and Cambridgeshire is very well-known, and the rate at which new native species are discovered is now very low. In over 20 years since the publication of Kent’s Flora in 1975, Galium constrictum and Salvia pratensis have been added to the Middlesex flora as a result of the discovery of overlooked herbarium specimens, Erophila glabrescens, E. majuscula and Juncus ambiguus were also found in herbaria following taxonomic revisions, and two new native species were detected in the wild, the ferns Dryopteris affinis and Polypodium interjectum (Kent, in press; Rich & Lewis 1999). Similarly in Cambridgeshire the only native vascular plants to be added to those listed in the 1983 Checklist have been three species represented by 18th or 19th century herbarium specimens (Eriophorum vaginatum, Rosa pumpinellifolia and Ruppia cirrhosa, the last known previously from literature references which were misinterpreted as R. maritima) and a single species discovered in the wild, Dactylorhiza maculata. The definition of what constitutes a native species is a greater area of uncertainty than under-recording of species in the field. In this analysis the list of native species made use of that given by Preston & Hill (1997), which in turn largely followed Stace (1991), in order to maintain comparability with the national study. There are certainly changes which could and perhaps should be made to this list, but they are very unlikely to alter any of the conclusions reached in this paper. Although the list of native species can be regarded as relatively stable, experience suggests that the list of supposedly extinct species will change. At the vice-county level, the slogan ‘extinction is for ever’ is manifestly false. Species may be discovered at new sites or rediscovered at sites where they have been overlooked, they may arise from seed banks after vegetation change or they may reinvade the county by natural means (although recolonisation 1s less likely to be a significant factor for flowering plants than for cryptogams, which have more mobile propagules). Babington (1860, pp. 314-315) listed 61 “lost plants” which had not been recorded in Cambridgeshire for many years or which, although recorded more recently, were “also probably now extirpated’. Only 42 of these 61 species are natives of Cambridgeshire and, of these 42, 22 (53%) are still thought to have become extinct before 1860, 6 (14%) were rediscovered but are now regarded as extinct and 14 (33%) are still present in the county. Similarly, Trimen & Dyer (1869, pp. 345-346) reported as extinct 38 species which are accepted as natives of Middlesex in this paper. Of these 38, only 17 (45%) are still thought to have become extinct before 1869; the remaining 21 species were later rediscovered and 13 (34%) of these are still present, although 8 (21%) are again thought to be extinct. The rediscovery of supposedly extinct species has continued into modern times: 13 native species which Kent (1975) regarded as extinct had been rediscovered by the time he prepared the Supplement (Kent, in press). James (1997) has documented the same phenomenon in Hertfordshire. One must therefore expect some species listed as extinct in Appendix | to be rediscovered. There may also be species which have been seen in one or other county since 1990 and are therefore regarded as extant, but which will never be seen there again. The extent to which the dates of the last records of extinct species reflect the actual timing of extinctions also needs to be assessed. These dates might potentially be influenced by the history of plant recording in each county. Even if the rate of extinction in a county was absolutely constant, fluctuations in the intensity of recording would nevertheless lead to an apparent concentration of last records in periods of intensive recording. It is, however, unlikely that the main fluctuations in extinction rate suggested by the data in Table 2 can be explained in this way. In Middlesex the rate in the 1840-1869 period was low despite the publication of Trimen & Dyer’s Flora in 1869; subsequently the rate has been surprisingly constant. In Cambridgeshire the high rate from 1810 to 1839 cannot be explained by a subsequent falling off of recording intensity as the next period, 76 C. D. PRESTON from 1840 to 1869, was that in which Babington prepared his Flora. There is scope for a more detailed study of extinction in each county, which would attempt to estimate dates of extinction of those species which have succumbed to habitat destruction by the date of the destruction of the last site rather than by the date of the last botanical record. Even if the date of extinction of a species is known with certainty, it will not necessarily coincide with the date of the causal event. Plants may persist for many decades after gradual changes which render sites unsuitable for the sexual or vegetative reproduction of a species but are insufficient to eradicate it completely (Summerfield 1972). However, catastrophic changes such as total habitat destruction may eliminate a species at once. Such changes were characteristic of the agricultural revolution which accompanied parliamentary enclosure in Cambridgeshire, and doubtless explain why this period stands out so strongly in the record of extinctions. In the absence of radical change, the correlation between land-use history and the rate of extinction will be less clear-cut. ECOLOGICAL AND PHYTOGEOGRAPHICAL CHARACTERISTICS OF EXTINCT SPECIES A high proportion of the extinct species are small, grow in open habitats, favour substrates with low pH or are characteristic of environments with low fertility. This is true of both counties, although the patterns are sharper in Cambridgeshire than in Middlesex (Figs 1, 2). Although these attributes have been treated separately, there is clearly a correlation between them. Habitats which have low pH, for example, generally have low fertility and a vegetation composed of low-growing species. Such semi-natural habitats have often been destroyed by agricultural improvement and urban expansion, and any remaining remnants have been indirectly modified by factors such as the reduction in grazing in an increasingly arable or urban landscape, eutrophication, falling water tables and increasing disturbance from a larger and more mobile population. Twelve species combine all four attributes of the least successful species (typical maximum height <30 cm and Ellenberg L=8—9, R=1-3, N=1-—2). Of these, four species are extinct in both counties (Drosera intermedia, D. rotundifolia, Lycopodiella inundata, Lycopodium clavatum), four more are extinct in the only county from which they are recorded (Antennaria dioica, Drosera anglica, Hammarbya paludosa, Radiola linoides) and two are extinct in one county but survive in the other (Lythrum portula extinct in Cambridgeshire and Vicia lathyroides in Middlesex). Surprisingly, two species in this group not only survive in both counties but even persist in central London, Aira praecox and Trifolium arvense (Kent, in press). In an analysis of a completely different dataset, the results of the B.S.B.I. Monitoring Scheme, Thompson (1994) demonstrated a marked correlation between low canopy height and a tendency to decline in England. He pointed out that canopy height is correlated with competitive ability, so the declining species tend to be the less competitive. He commented on the “increasing restriction of slow-growing plants of infertile, relatively undisturbed habitats to fragmented islands of suitable habitat, many of them in nature reserves, surrounded by a sea of unsuitable landscape”’. Thompson’s conclusions about the nature of the species which have declined in the modern landscape are strikingly similar to those of this survey. In addition to these trends, there is a consistent indication in Cambridgeshire and Middlesex that wetland and coastal species have suffered more than plants in other habitats. The decrease in wetland species has been noted in other studies (e.g. Dony 1977; James 1997; Mountford 1994). The tendency of coastal species to decrease almost certainly reflects the very small area of suitable habitat for such species in these counties rather than a national trend. Judging by trends shown by nationally scarce species, the distribution of coastal species is relatively stable (Stewart ef al. 1994). The persistence of the woodland flora of both counties is striking. Some of the differences between Cambridgeshire and Middlesex reflect the differing geology of the two counties. Only in Middlesex do the species of basic habitats show a high level of extinction, a reflection of the small area of calcareous soil in that county. One might expect a correspondingly greater survival of calcifuge species in Middlesex, but this is only true to a limited extent. There has been massive extinction in calcifuge species and plants characteristic of plant communities of acidic soils in Cambridgeshire. However, the loss of these species has also been greater than average in Middlesex. Even though Middlesex is dominated by acidic soils, the heathland habitats of many species have been destroyed or grossly altered. The most striking feature of the phytogeographical analysis is the loss of species from groups at the edge of their geographical range, the Boreal, Mediterranean-Atlantic and Oceanic plants. The EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE ig Same pattern is revealed by a recent analysis of extinctions in Bedfordshire (Boon 1999). If such changes are indeed characteristic of south-east England as a whole, they must have led to a reduction in the phytogeographical diversity of the flora and the increasing predominance of species in the already dominant major biome classes. However, it must be remembered that this applies only to native species. The increase in alien species of diverse origins is an aspect of change which has not been examined in this paper, but must be considered in an overall assessment of the phytogeographical diversity of the area. EFFECTS OF URBANISATION AND AGRICULTURAL IMPROVEMENT The similarities in the extent and nature of extinctions in Cambridgeshire and Middlesex are much more striking than the differences. This is remarkable in view of the differing land-use history of the counties: Middlesex has become a suburban and urban county, whereas Cambridgeshire is still rural although very intensively farmed. The fact that 18% of species have been lost from Middlesex and 13% from Cambridgeshire suggests that agricultural intensification is almost as damaging to the native flora as urbanisation. In making these comparisons it is difficult to account for the effect of the differing size of the counties. If the counties were uniform, one might expect fewer extinctions in the larger county, but the highly heterogeneous nature of Cambridgeshire with distinct areas of chalk, clay and fenland negates such a simplistic approach. Thompson & Jones (1999) found little relationship between the number of extinct scarce species and the area of a vice-county as such. An alternative explanation for the similarity of the changes in the two counties might be that the changes in each have the same cause — that in both counties the primary cause is urbanisation. Thompson & Jones (1999) found a strong relationship between the number of nationally scarce species that have been lost from vice-counties and their current population density. They point out that population density is an indicator of many aspects of land-use (including road-building, recreation and urbanisation) which have a negative impact on plant survival. They go on to suggest that, whereas intensive agriculture “is frequently assumed to be the main cause of decline in Britain’s native wildlife’, these effects of population pressure may be “more pervasive and widespread than suspected” and “a major cause of local plant extinction”. They conclude that a more detailed search for the proximate causes of local plant extinction would be rewarding. The study reported in this paper differs from that of Thompson & Jones (1999) in being more limited spatially, but benefits from incorporating a temporal dimension. There has been a marked variation in the rate of extinction in differing time periods in both Middlesex and Cambridgeshire, and this provides a powerful insight into the possible causes of such extinctions. There is evidence to support Thompson & Jones’ view that urbanisation has been a major cause for species loss in Middlesex (indeed, the urbanisation of that county is so extreme that no other conclusion would be credible). However, the analysis for Cambridgeshire has demonstrated the importance of agricultural change as an agent of local extinction. These variations in the rate of extinction also highlight the desirability of separating historic and recent extinctions when considering the implications of previous changes for the conservation of the modern flora. Causal factors as well as rates may have varied over time, and the fact that historic extinctions in Cambridgeshire have been primarily caused by agricultural improvement does not mean that the effects of population pressure described by Thompson & Jones (1999) are not important now, or will not become apparent in the future. Further detailed studies of extinction at the vice-county level can be expected to throw more light on the changes which have taken place in the British flora. The vice-county is a convenient unit to use in the analysis of such changes, and such analyses are completely dependent on critical compilations of historic records such as that undertaken by Duggie Kent for The historical Flora of Middlesex. ACKNOWLEDGMENTS I am very grateful to R. M. Burton, Mrs G. Crompton and D. A. Wells for help in drawing up the lists of native and extinct native species, although I take responsibility for the many, sometimes somewhat arbitrary, decisions which have had to be made in finalising the lists. The ecological and phytogeographical data used in this paper have been taken from ORACLE databases held at LT.E. 78 C. D. PRESTON Monks Wood and I thank H. R. Arnold, D. B. Roy and Dr M. G. Telfer for help with computing. C. R. Boon, R. M. Burton, Prof. C. A. Stace and Dr K. Thompson kindly helped me obtain copies of unpublished books or papers, and Julie Gaunt generated the diagrams. R. M. Burton, Mrs J. M. Croft, Mrs G. Crompton, P. T. Harding, Dr M. O. Hill and P. H. Oswald kindly commented on a draft of this paper. I am also grateful to Dr O. Rackham and Dr D. L. Kelly for their advice as referee and editor, respectively. I appreciate the opportunity of preparing this paper for an issue dedicated to D. H. Kent, whose detailed work on the Middlesex flora provides one basis for the study. I was grateful for the cheerful welcome this doyen of the B.S.B.I. extended to me as a naive newcomer, often benefited from his irreverent observations and gradually came to appreciate the enormous amount of work he undertook for the Society with so little fuss. REFERENCES BABINGTON, C. C. (1860). Flora of Cambridgeshire. John van Voorst, London. BETJEMAN, J. (1954). A few late chrysanthemums. John Murray, London. BLACKSTONE, J. (1737). Fasciculus plantarum circa Harefield sponte nascentium. London. Boon, C. R. (1999). British and Irish floristic elements applied to the Bedfordshire flora. Bedfordshire naturalist 52: 78-91. BULL, G. B. G. (1957). The changing landscape of rural Middlesex 1500-1850. Ph.D. thesis, University of London. CLAPHAM, A. R., TUTIN, T. G. & Moore, D. M. (1987). Flora of the British Isles, 3rd ed. Cambridge University Press, Cambridge. CLUTTERBUCK, J. C. (1869). The farming of Middlesex. Journal of the Royal Agricultural Society of England 2nd series 5: 2—27. CROMPTON, G. (1997). Botanizing in Cambridgeshire in the 1820s. Nature in Cambridgeshire 39: 59-73. CROMPTON, G. & WELLS, D. (1996, unpublished). “Extinct” and rare species in vc 29 (Cambs.). Document circulated to Cambridgeshire Flora Group. CROMPTON, G. & WHITEHOUSE, H. L. K. (1983). A checklist of the flora of Cambridgeshire. Privately published, Cambridge. DARBY, H. C. (1956). The draining of the fens, ed. 2. Cambridge University Press, Cambridge. Dony, J. G. (1977). Change in the flora of Bedfordshire, England, from 1798 to 1976. Biological conservation 11: 307-320. DRAYTON, M. [1612]. Poly-Olbion. London. ELLENBERG, H. (1988). Vegetation ecology of central Europe. English edition, translated from 4th German edition. Cambridge University Press, Cambridge. FITTER, A. H. & PEAT, H. J. (1994). The ecological flora database. Journal of ecology 82: 415-425. GRIGG, D. B. (1967). The changing agricultural geography of England: a commentary on the sources available for the reconstruction of the agricultural geography of England, 1770-1850. Transactions of the Institute of British Geographers 41: 73-96. GRIME, J. P., HODGSON, J. G. & HUNT, R. (1988). Comparative plant ecology. Unwin Hyman, London. HILL, M. O., MOUNTFORD, J. O., Roy, D. B. & BUNCE, R. G. H. (1999). Ellenberg’s indicator values for British plants. Institute of Terrestrial Ecology, Huntingdon. JACKSON, R. V. (1985). Growth and deceleration in English agriculture, 1660-1790. Economic history review 2nd series, 38: 333-351. JAMES, T. J. (1997). The changing flora of Hertfordshire. Transactions of the Hertfordshire Natural History Society 33: 62-84. . KENT, D. H. (1975). The historical Flora of Middlesex. Ray Society, London. KENT, D. H. (in press). Supplement to The historical Flora of Middlesex. Ray Society, London. LINDACHER, R. (1995). PHANART: Datenbank der Gefasspflanzen Mitteleuropas. Verdéffentlichungen des geobotanischen Institutes der eidgendssische technische Hochschule, Stiftung Riibel, in Ziirich 125: 1—- 436. Lucas, J., trans. (1892). Kalm’s account of his visit to England on his way to America in 1748. Macmillan & Co., London. MIDDLETON, J. (1798). View of the agriculture of Middlesex. B. Macmillan, London. MINGAY, G. E. (1989). Introduction, in MINGAY, G. E., ed. The agrarian history of England and Wales, vol. 6, 1750-1850, pp. 1-5. Cambridge University Press, Cambridge. MOUNTFORD, J. O. (1994). Floristic change in English grazing marshes: the impact of 150 years of drainage and land-use change. Watsonia 20: 3-24. NORDEN, J. (1593). Speculum Britanniae. The first parte. An historicall, & chorographicall discription of Middlesex. EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 79 PERREN, R. (1995). Agriculture in depression, 1870-1940. Cambridge University Press, Cambridge. Pettit, G. H. N. (1941). Cambridgeshire (excluding the Isle of Ely), in STAMP, L. Dudley, ed. The land of Britain, part 74, pp. 379-404. Geographical Publications, London. PERRING, F. H., SELL, P. D., WALTERS, S. M. & WHITEHOUSE, H. L. K. (1964). A Flora of Cambridgeshire. Cambridge University Press, Cambridge. PiGoTT, C. D. (1988). The reintroduction of Cirsium tuberosum (L.) All. in Cambridgeshire. Watsonia 17: 149-152. PRESTON, C. D. (1997). The genus Rosa in Cambridgeshire (v.c. 29): an interim account. Nature in Cambridgeshire 39: 40-53. PRESTON, C. D. & HILL, M. O. (1997). The geographical relationships of British and Irish vascular plants. Botanical journal of the Linnean Society 124: 1-120. PRINCE, H. C. (1989). The changing rural landscape, 1750-1850, in MINGAY, G. E., ed. The agrarian history of England and Wales, vol. 6, 1750-1850, pp. 7-83. Cambridge University Press, Cambridge. RAY, J. (1660). Catalogus plantarum circa Cantabrigiam nascentium. Cambridge & London. RELHAN, R. (1785). Flora Cantabrigiensis. Cambridge. RELHAN, R. (1802). Flora Cantabrigiensis, 2nd ed. Cambridge. RELHAN, R. (1820). Flora Cantabrigiensis, 3rd ed. Cambridge. RICH, T. C. G. & LEwis, J. (1999). Use of herbarium material for mapping the distribution of Erophila (Brassicaceae) taxa sensu Filfilan & Elkington in Britain and Ireland. Watsonia 22: 377-385. RICHARDSON, R. C. (1984). Metropolitan counties: Bedfordshire, Hertfordshire and Middlesex, in THIRSK, J., ed. The agrarian history of England and Wales, vol. 5, 1640-1750, 1, Regional farming systems, pp. 239- 269. Cambridge University Press, Cambridge. SELL, P. (1989). The changing face of nature in Bassingbourn (1930s—1980s). Nature in Cambridgeshire 31: 12-18. SMAILES, A. E. (1964). The site, growth, and changing face of London, in CLAYTON, R., ed. The geography of Greater London, pp. 1-52. G. Philip & Son, London. STACE, C. A. (1991). New Flora of the British Isles. Cambridge University Press, Cambridge. STACE, C. A. (1997). New Flora of the British Isles, ed. 2. Cambridge University Press, Cambridge. STAMP, L. DUDLEY (1941). Note on land use changes in Cambridgeshire 1794-1938, in STAMP, L. Dudley, ed. The land of Britain, part 74, pp. 405-408. Geographical Publications, London. STEWART, A., PEARMAN, D. A. & PRESTON, C. D., comps & eds. (1994). Scarce plants in Britain. Joint Nature Conservancy Committee, Peterborough. STOATE, C. (1996). The changing face of lowland farming and wildlife, part 2: 1945-1995. British wildlife 7: 162-172. SUMMERFIELD, R. J. (1972). Biological inertia - an example. Journal of ecology 60: 793-798. TANNER, M. E. (1911). Social and economic history, in PAGE, W., ed. The Victoria history of the county of Middlesex 2, pp. 61—109. Reprinted in 1970 by Dawsons of Pall Mall, Folkestone. TAYLOR, C. (1973). The Cambridgeshire landscape. Hodder & Stoughton, London. THIRSK, J. (1984). Introduction, in THIRSK, J., ed. The agrarian history of England and Wales, vol. 5, 1640- 1750, 1, Regional farming systems, pp. Xix—xxxi. Cambridge University Press, Cambridge. THIRSK, J. (1997). Alternative agriculture. Oxford University Press, Oxford. THOMPSON, K. (1994). Predicting the fate of temperate species in response to human disturbance and global change, in BOYLE, T. J. B. & BOYLE, C. E. B., eds. Biodiversity, temperate ecosystems, and global change, pp. 61—76. Springer-Verlag, Berlin. THOMPSON, K. & JONES, A. (1999). Human population density and prediction of local plant extinction in Britain. Conservation biology 13: 185-189. TRIMEN, H. & DYER, W. T. T. (1869). Flora of Middlesex. Robert Hardwicke, London. TURNER, M. (1980). English parliamentary enclosure. W. Dawson & Sons, Folkestone. TURNER, W. [1548]. The names of herbes in Greke, Latin, Englishe, Duche & Frenche wyth the commune names that herbaries and apotecaries use. Reprinted in 1965 by the Ray Society, London. VANCOUVER, C. (1794). General view of the agriculture in the county of Cambridge. W. Smith, London. WEST, W. (1898). Notes on Cambridgeshire plants. Journal of botany 36: 246-259. WHETHAM, E. H., ed. (1978). The agrarian history of England and Wales, vol. 8, 1914-1939. Cambridge University Press, Cambridge. WIGGINTON, M. J., comp. & ed. (1999). British Red Data Books 1 Vascular plants, 3rd ed. Joint Nature Conservation Committee, Peterborough. WILLATTS, E. C. (1937). Middlesex and the London region, in STAMP, L. Dudley, ed. The land of Britain, part 79, pp. 117-304. Geographical Publications, London. WRIGLEY, E. A. (1967). A simple model of London’s importance in changing English society and economy 1650-1750. Past & present 37: 44-70. (Accepted June 1999) 80 C. D. PRESTON APPENDIX 1. EXTINCT NATIVE SPECIES OF MIDDLESEX AND CAMBRIDGESHIRE The extinct species are listed below, with the date of the last record. MIDDLESEX (V.C. 21) Agrimonia procera, 1874; Alchemilla filicaulis, 1970; Allium oleraceum, 1856; Althaea officinalis, 1760; Anagallis minima, 1800; A. tenella, 1873; Anchusa arvensis, 1966; Anthemis cotula, date not known; Aquilegia vulgaris, 1737; Arnoseris minima, 1778; Barbarea stricta, 1965; Beta vulgaris, 1887; Blysmus compressus, 1830; Bupleurum tenuissimum, 1860; Campanula glomerata, 1980; Carduus tenuiflorus, 1929; Carex appropinquata, 1936; C. binervis, 1983; C. dioica, 1792; C. filiformis, 1960; C. laevigata, 1830; C. pulicaris, 1946; C. rostrata, 1792; Carlina vulgaris, 1730; Chenopodium glaucum, 1953; C. vulvaria, 1970; Chrysanthemum segetum, 1978; Chrysosplenium alternifolium, 1838; Cicuta virosa, 1780; Cirsium dissectum, 1884; Cochlearia anglica, 1869; Coeloglossum viride, 1913; Crepis foetida, 1873; Cuscuta epithymum, 1887; C. europaea, 1960; Cynoglossum germanicum, 1710; C. officinale, 1955; Damasonium alisma, 1886; Dianthus armeria, 1948; D. deltoides, 1939; Drosera intermedia, 1746; D. rotundifolia, 1935; Eleocharis multicaulis, 1960; Erica cinerea, 1981; E. tetralix, 1970; Eriophorum angustifolium, 1980; Euphorbia platyphyllos, 1884; Euphrasia arctica, 1884; E. confusa, 1887; Filago pyramidata, 1891; Fritillaria meleagris, 1950; Fumaria purpurea, 1892; Galeopsis angustifolia, 1945; Galium constrictum, 1780; Gentiana pneumonanthe, 1795; Gentianella amarella, 1975; G. germanica, 1977; Gnaphalium sylvaticum, 1945; Gymnadenia conopsea, 1965; Helianthemum nummularium, 1907; Hordelymus europaeus, 1951; Hypericum androsaemum, 1947; H. elodes, 1826; Hypochaeris glabra, 1843; Jasione montana, 1935; Juncus ambiguus, 1842; Juniperus communis, 1746; Lactuca saligna, 1800; Legousia hybrida, 1914; Leucojum aestivum, 1845; Lithospermum arvense, 1888; L. officinale, 1914; Littorella uniflora, 1935; Lycopodiella inundata, 1869; Lycopodium clavatum, 1865; Lythrum hyssopifolia, 1778; Maianthemum bifolium, 1924; Marrubium vulgare, 1949; Melampyrum arvense, 1870; Mentha pulegium, 1871; Myosotis secunda, 1847; Myosurus minimus, 1914; Myrica gale, 1901; Myriophyllum alterniflorum, 1957; Neottia nidus-avis, 1946; Nepeta cataria, 1978; Oenanthe lachenalii, 1847; Ophrys insectifera, 1907; Orchis militaris, 1900; O. ustulata, 1737; Orobanche elatior, 1902; O. rapum-genistae, 1873; Osmunda_ regalis, 1794; Parnassia_ palustris, 1900; Pedicularis palustris, 1920; Petroselinum segetum, 1930; Pilularia globulifera, 1800; Platanthera bifolia, 1887; P. chlorantha, 1890; Polystichum aculeatum, 1965; Potamogeton acutifolius, 1981; P. alpinus, 1886; P. compressus, 1917; P. polygonifolius, 1884; P. trichoides, 1966; Potentilla argentea, 1916; Pulicaria vulgaris, 1908; Pyrola minor, 1908; Radiola linoides, 1890; Ranunculus aquatilis, 1918; R. arvensis, 1966; R. fluitans, 1977; R. hederaceus, 1974; R. parviflorus, 1885; R. sardous, 1912; Rosa agrestis, 1818; R. micrantha, 1910; R. pimpinellifolia, 1887; Sagina nodosa, 1906; S. subulata, 1783; Salvia pratensis, 1901; Samolus valerandi, 1866; Scabiosa columbaria, 1939; Scandix pecten-veneris, 1888; Schoenoplectus triqueter, 1887; Scrophularia umbrosa, 1841; Silene noctiflora, 1967; Sium latifolium, 1965; Sonchus palustris, 1835; Spiranthes spiralis, 1936; Stachys arvensis, 1954; Stellaria palustris, 1977; Teesdalia nudicaulis, 1932; Trichophorum cespitosum, 1780; Trifolium ornithopodioides, 1885; T. squamosum, 1721; Utricularia minor, 1746; Valeriana dioica, 1970; Valerianella dentata, 1950; V. rimosa, 1941; Vicia lathyroides, 1866; Viola canina, 1966; V. lactea, 1868; V. palustris, 1912; Vulpia unilateralis, 1969; Wolffia arrhiza, 1898. EXTINCTION IN MIDDLESEX AND CAMBRIDGESHIRE 81 CAMBRIDGESHIRE (V.C. 29) Alchemilla filicaulis, 1953; Alisma gramineum, 1972; Althaea officinalis, 1959; Anagallis minima, 1820; Antennaria dioica, 1864; Aphanes inexspectata, 1977; Armeria maritima, 1930; Arnoseris minima, 1914; Atriplex pedunculata, 1826; Beta vulgaris, 1928; Bidens cernua, 1964; Blechnum spicant, 1660; Botrychium lunaria, 1833; Bromus interruptus, 1972; Campanula latifolia, 1852; Carex binervis, 1961; C. curta, 1853; C. diandra, 1860; C. dioica, 1841; C. echinata, 1954; C. pilulifera, 1949; C. rostrata, 1966; C. strigosa, 1795; Catapodium marinum, 1785; Ceterach officinarum, 1967; Chenopodium urbicum, 1837; C. vulvaria, 1958; Cicuta virosa, 1763; Cirsium tuberosum, 1974; Colchicum autumnale, 1926; Crassula tillaea, 1930; Cuscuta epithymum, 1975; Drosera anglica, 1840; D. intermedia, 1820; D. rotundifolia, 1913; Eleocharis multicaulis, 1853; Eleogiton fluitans, 1975; Epilobium palustre, 1969; Epipactis phyllanthes, 1987; E. purpurata, 1962; Egquisetum sylvaticum, 1763; Erica cinerea, 1914; E. tetralix, 1920; Eriophorum angustifolium, 1960; E. latifolium, 1887; Festuca filiformis, 1964; Frankenia laevis, 1820; Fritillaria meleagris, 1820; Genista anglica, 1932; Gnaphalium luteoalbum, 1802; Hammarbya paludosa, 1855; Herminium monorchis, 1825; Hordeum marinum, 1881; Hypericum elodes, 1930; Hypochaeris glabra, 1964; Jasione montana, 1910; Juncus squarrosus, 1833; Lactuca saligna, 1953; Lathraea squamaria, 1889; Lepidium heterophyllum, 1965; L. latifolium, 1795; L. ruderale, 1831; Limosella aquatica, 1877; Liparis loeselii, 1945; Littorella uniflora, 1820; Lycopodiella inundata, 1853; Lycopodium clavatum, 1831; Lythrum portula, 1878; Marrubium vulgare, 1930; Melampyrum arvense, 1862; M. pratense, 1930; Mentha pulegium, 1903; Misopates orontium, 1873; Moenchia erecta, 1928; Monotropa hypopitys, 1946; Montia fontana, 1945; Myriophyllum alterniflorum, 1919; Narthecium ossifragum, 1833; Ophrys sphegodes, 1837; Orchis ustulata, 1955; Oreopteris limbosperma, 1820; Ornithogalum pyrenaicum, 1774; Orobanche artemisiae- campestris, 1935; O. rapum-genistae, 1913; Osmunda regalis, 1685; Parnassia palustris, 1980; Pedicularis sylvatica, 1912; Pinguicula vulgaris, 1984; Polygala serpyllifolia, 1954; Polystichum aculeatum, 1820; Potamogeton compressus, 1912; P. polygonifolius, 1948; Potentilla anglica, 1945; P. palustris, 1886; Pulicaria vulgaris, 1833; Ranunculus baudotii, 1955; R. hederaceus, 1900; R. parviflorus, 1928; Rhynchospora alba, 1839; Rosa pimpinellifolia, 1827; R. sherardii, 1910; Ruppia cirrhosa, 1795; Sagina maritima, 1983; Salicornia dolichostachya, 1959; Silene gallica, 1951; Solidago virgaurea, 1916; Sonchus palustris, 1843; Stratiotes aloides, 1960; Teesdalia nudicaulis, 1954; Tephroseris palustris, 1830; Trichophorum cespitosum, 1820; Triglochin maritima, 1930; Ulex minor, 1832; Utricularia australis, 1899; U. intermedia, 1898; U. minor, 1951; Vaccinium oxycoccos, 1859; Valerianella rimosa, 1933; Vicia sylvatica, 1792; Zostera marina, 1908. APPENDIX 2. DIVISIONS OF CAMBRIDGESHIRE The following seven divisions were defined to assist in comparison of data presented by Kent (1975 and in press) for seven divisons of Middlesex. They correspond very approximately to the divisions recognised by Babington (1860), which are named after each list of 10 km squares: 1: TL 54, 55, 64, 65, 75 (Cambridge); 2: TL 23, 24, 33, 34, 43, 44 (Royston); 3: TL 15, 25, 26, 35, 36, 45 (Cottenham and Wimpole); 4: TL 46, 56, 57, 66, 67, 76 (Burwell); 5: TL 47, 48, 58, 59, 68, 69 (Ely); 6: TL 29, 37, 38, 39, 49 (Chatteris); 7: TF 20, 30, 31, 40, 41, 51 (Wisbech). Pie Watsonia 23: 83—92 (2000) 83 Hybridisation between Rumex rupestris Le Gall (Polygonaceae) and other docks D. T. HOLYOAK S Edward Street, Tuckingmill, Camborne, Cornwall, TR14 SPA ABSTRACT Hybrids of Rumex rupestris with R. conglomeratus are reported for the first time and named as R. xX rosemurphyae D. T. Holyoak, hybr. nov. Other hybrids involving R. rupestris have been found at the same locality in West Cornwall (v.c. la), involving R. pulcher (= R. X trimenii Camus), probably R. crispus, and possibly R. obtusifolius. Evidence of introgressive hybridisation was found resulting from R. x rosemurphyae backcrossing with R. conglomeratus, but there was no evidence of introgression with other coexisting Rumex species. KEYWORDS: conservation, Shore Dock, introgression, taxonomy. INTRODUCTION Shore Dock Rumex rupestris Le Gall is a rare European endemic species occurring on and near coasts from Wales southwards to north-western Spain (Jalas & Suominen 1979; Daniels ef al. 1998). Many of its localities are on rocky sea cliffs where few other dock species grow, so that opportunities for it to be involved in interspecific hybridisation are less prevalent than with congeners that commonly grow together on disturbed ground inland. Indeed, the reviews by Lousley & Williams (1975) and Lousley & Kent (1981) reported few hybrids of R. rupestris, and those only with R. pulcher L. and R. crispus L. There do not appear to be any reports of hybrids involving R. rupestris from outside Britain. R. rupestris was investigated from 1994—1998 in dune-slack like habitats at Penhale Camp, West Cornwall. During this period its population there increased from about 60 to 137 mature plants. At this site, four other dock species (Rumex subgenus Rumex) and several of their interspecific hybrids grow intermingled with, or close to, R. rupestris. It was expected that under these circumstances hybrids involving R. rupestris would occur, and, over the five years of study, a total of eight such hybrid plants was found, apparently representing four different hybrid combinations. This paper extends and partly revises the preliminary notes (Holyoak 1995, 1996) on the hybrids at Penhale Camp by giving descriptions of each of the hybrid taxa and naming the hybrid with R. conglomeratus Murray which has not been reported from elsewhere. In addition, previous records of hybrids involving R. rupestris are reviewed. Results are also described of biometric investigation of apparent introgression of R. rupestris and R. conglomeratus at Penhale Camp. Potential threats to the survival of R. rupestris from introgressive hybridisation are discussed on the basis of these data. METHODS Penhale Camp (West Cornwall; c. SW/770.570) occupies an extensive area of coastal sand-dunes. In most winters water stands in several large and small pools and flows along an ephemeral stream, but all of these areas are usually dry in summer. These dune-slack like areas support vegetation characteristic of dune-slacks, fens and pool-margins, with locally dominant plants of different areas including Eleocharis palustris (L.) Roemer & Schultes, Epilobium hirsutum L., Equisetum palustre L., Mentha aquatica L., Pulicaria dysenterica (L.) Bernh. and Rorippa nasturtium-aquaticum (L.) Hayek. Rumex rupestris grows intermixed with these wetland plants in four separate small colonies. 84 D. T. HOLYOAK TABLE 1. HYBRID FREQUENCY INDEX MEASUREMENTS OF RUMEX RUPESTRIS, R. CONGLOMERATUS AND INTERMEDIATE PLANTS FROM LOCALITIES IN WEST CORNWALL, AUGUST 1996 Hybrid-index score Site 0 1 2 22 4 3 (6 7 8 9 0D Pity ibrsAais Penhale Camp Ze Be 2 - - - 1 - = 2A oO 22) 28 SW/769.569 Mount Field - - - - - - - - - - . - (ea ela) 20 SW/781.571 near Ventongimps - - - - - - - - - - - - Sesto. 20 Moor SW/779.511 Bonython Estate - - - - - - - - - - - - SU Ome Mees SW/696.207 See Appendix | for details of the five characters used and the scoring system, with which a typical plant of R. rupestris scores 0 and a typical plant of R. conglomeratus scores 15. R. conglomeratus occurred at all four of these sites; R. rupestris was present with it only at Penhale Camp. Other Rumex species present in the same wet areas are R. conglomeratus (hundreds of plants), R. crispus subsp. littoreus (J. Hardy) Akeroyd (many hundreds of plants, but most of them growing in drier edges of wetland vegetation) and R. obtusifolius L. var. obtusifolius (2 plants seen). In addition, R. pulcher occurs in very small quantity on dry slopes nearby. Other dock hybrids recorded in the same area were R. conglomeratus X R. crispus (5), R. conglomeratus X R. pulcher (1) and R. crispus X R. obtusifolius (c. 47) (Holyoak 1996 and subsequent pers. obs.). The area was visited several times in August of each of the years 1994 to 1998, so that virtually all of the docks present could be identified and counted as their fruits ripened. Some plants were individually marked from 1995 onwards and by 1998 all plants of R. rupestris and its hybrids had been individually marked. In 1995, 1996 and 1998 specimens were collected from each of the marked hybrids involving R. rupestris at times when they had at least some mature fruits. During August 1996 it was noticed that some fruiting plants of R. conglomeratus showed characters approaching those of the R. rupestris growing near them. Because these intermediate characters seemed likely to have resulted from hybridisation, their morphology and those of the closest plants of R. conglomeratus and R. rupestris were investigated using the “hybrid frequency index” technique of Anderson (1936). The scoring system used is explained in Appendix | and other details are given with the results in Table 1. Comparative data were obtained from three populations of R. conglomeratus growing at localities in West Cornwall that lacked R. rupestris. Several counts of chromosomes at mitotic metaphase were obtained from root tips of seedlings germinated on moist filter paper in petri dishes. Excised root tips were fixed overnight in Farmer’ s fluid before squash preparations were prepared using acetic orcein stain. RESULTS Rumex X rosemurphyae D. T. Holyoak, hybr. nov. (Rumex conglomeratus Murray X R. rupestris Le Gall) (Fig. 1) Hybrida a Rumice conglomerato Murray et R. rupestri Le Gall genita et characteribus plerlsque intermedia; ab ambobus fructibus pro parte maxima abortivis et statura nonnunquam multo majore differt. A hybrid between Rumex conglomeratus and R. rupestris, found within 2 m of plants of those two species. It is intermediate between them in most characters but differs from both in being mostly but not completely infertile and sometimes in its much greater size. A robust perennial growing in a compact clump from a stout rootstock. The holotype was much larger than accompanying plants of R. conglomeratus and R. rupestris, on 8 August 1995 it had 35 RUMEX RUPESTRIS HYBRIDS 85 (G; iy 7 Q ORD (Cy (ONY By AY ce ow N NY NY \" ee Gf 6 Hye | OD Ay. ys AD BY) i) FIGURE 1. Rumex x rosemurphyae. A. Fruit (i.e. perianth enclosing nutlet) B. Diagrammatic section through fruit. C. Single whorl of inflorescence. D. Inflorescence. E. Leaf from lower part of stem. Scale bars represent 1 mm (A-C) or 10 mm (D, E). 86 Di T.HOLYOAK stems up to 1-4 m long, the same plant on 18 July 1998 had 77 flowering stems up to 1-5 m long, in both years most stems had become decumbent before fruits ripened. Lower stem leaves with lamina at flowering up to 20 x 8 cm or more, but most lower leaves wither before fruits ripen. Upper stem leaves smaller, oblong-lanceolate to lanceolate. At least some leaves thicker than in accompanying plants of R. conglomeratus, but less thick than in some R. rupestris. Panicles with many branches that mostly arise at about 45° from main stem, the branches more numerous than is usual in R. rupestris, but with branching at more acute angles than typical of R. conglomeratus. Each of the lower whorls of flowers on each branch subtended by an ovate- lanceolate to narrowly lanceolate leafy bract, the bracts becoming abruptly smaller towards the middle of each branch and absent near the branch apex, as in R. conglomeratus. On those parts of the inflorescence where fertile fruits occur, whorls of flowers appear less close and congested than in R. rupestris, more like those of R. conglomeratus. Where nutlets develop, inner perianth- segments oblong to oblong-lanceolate with bluntly rounded apex and sides subparallel in upper part, always entire. Where nutlets develop, inner perianth-segments very variable in size, some as small as in typical R. conglomeratus (length 1-7—2-5 mm) others as long as in typical R. rupestris (2-8-3-7 mm), but with many of intermediate lengths. Where nutlets develop, each of the inner perianth-segments has a swollen, rounded tubercle that varies from 70-120% of the maximum width of perianth segment and 40-65% of its length. The few well developed nutlets seen were trigonous, 1-4—1-6 mm long, brown, glossy, with acute angles. Counts of mitotic chromosomes from three seedlings grown from seed collected from the Holotype were all n = 20. Identical counts were obtained from seedlings of R. conglomeratus (4) and R. rupestris (5) grown from seed collected from plants growing within 10 m of the holotype. Counts of n = 20 have been reported previously for both of these species (e.g. Degraeve 1975; Rechinger 1993; Kay 1996). Named for Miss Rosaline J. Murphy in recognition of her work on the Cornish flora and as thanks for introducing the author to Penhale Camp. HOLOTYPUS: W. Cornwall, v.c. la, Penhale Camp (SW/768.569), among Epilobium hirsutum in fen along course of ephemeral stream, 7-8 August 1995 and 18 July 1998, D. T. Holyoak, field labels C and 28 (RNG). A smaller plant growing 6 m away from the holotype (field label D) was also identified as R. x rosemurphyae. On 8 August 1995 this plant had only two flowering stems, the longest 70 cm tall. It resembled the Holotype closely in other respects, including its low fertility. In addition to the two plants described above as R. X rosemurphyae and interpreted as F; hybrids between R. conglomeratus and R. rupestris, at least six (and perhaps as many as 13) of the 40 plants of R. conglomeratus growing within 20 m of them in August 1996 showed characters that somewhat approached those of R. rupestris (Table 1, Appendix 1). As discussed below, these are believed to represent back-crosses between R. x rosemurphyae and R. conglomeratus. On 13 September 1998 two more plants attributed to R. x rosemurphyae were seen in a dune- slack at Gear Sands (SW/7.5), about 700 m from the Penhale plants (marked by C. J. Neil as numbers 117 and 122; vouchers were given field labels DTH 4 and 6 respectively). They were again close to plants of both R. conglomeratus and R. rupestris. Both of these hybrids also had low fertility (<20% of nutlets developed) and both showed evidence of “hybrid vigour’, one plant having about 20 flowering stems up to 1-3 m tall, the other 39 flowering stems up to 1-4 m tall. Details of the inflorescence and inner perianth-segments were similar to those described for the Holotype from Penhale Camp. Their (fertile) inner perianth-segments measured (2-0)3-0—-3-3(3-7) mm on one plant and (2-2)3-0—3-2(4-0) mm on the other. Probable Rumex crispus L. x R. rupestris Le Gall A single plant tentatively attributed to this hybrid was found at Penhale Camp on 24 July 1996, close to both of the supposed parent species. It was about 60 cm tall with a single main stem. The lower stem leaves were lanceolate, up to 23 x 5 cm, rather thick and with somewhat crisped margins. The panicle had nine rather upright branches diverging at 10—20° from the main stem, several of the longer branches having several whorls in the lower half of the branch subtended by a narrowly lanceolate, petiolate bract. The whorls of the inflorescence appeared less crowded than in R. crispus, but this impression apparently resulted from low fertility, with many of the inner RUMEX RUPESTRIS HYBRIDS 87 perianth-segments failing to enlarge after flowering. The minority of perianth segments that had enlarged were smaller and narrower than in R. crispus, but broader than in R. rupestris (reaching 3-7 mm long X 2-8 mm wide), with wider apices, three swollen tubercles and entire margins (occasionally with a few short or indistinct teeth). While the balance of probabilities would suggest that this is a hybrid between R. crispus and R. rupestris, it may be impossible to discount other hybrid combinations using morphological characters. In particular, the parent with wide inner perianth segments might have been R. crispus x R. obtusifolius (= R. x pratensis Mert. & Koch) rather than R. crispus, a possibility strengthened by the presence of short teeth on some inner perianth segments of the hybrid. Varied forms of R. x pratensis were growing nearby, some of them with up to 40% of nutlets well developed. It is even possible that this hybrid plant represents an extreme form of R. X pratensis or a backcross between it and R. crispus, but some of its enlarged inner perianth segments appear too narrow for that to be likely. The second parent with narrow inner perianth segments might have been R. conglomeratus rather than R. rupestris, but that seems less likely in view of the upright branches of the panicle in the hybrid and the large size of some inner perianth segments and their tubercles. However, involvement of R. conglomeratus would explain the rather prominent bracts on some branches of the panicle. Because of these doubts about identification a new name for the hybrid combination of R. crispus x R. rupestris is not introduced here. The possible occurrence of “triple” hybrids involving R. X pratensis is discussed further below. Rumex crispus X R. rupestris has been reported from the Isles of Scilly (v.c. 1b) and Kenfig, Glamorgan (v.c. 41) (Lousley & Williams 1975; Lousley & Kent 1981; Stace 1991); there are specimens from both vice-counties at RNG. Dr J. R. Akeroyd (pers. comm.) has located an additional specimen collected above rocks just above HWM at Pendower Beach, E. Cornwall (v.c. 2) by Olga Stewart 277/82 on 13 September 1982 (FE). Kay (1996) mentions instances of R. crispus subsp. littoreus being mistaken for this hybrid. However, the Penhale plant and those discussed by Lousley & Williams (/oc. cit.) differed from R. crispus subsp. littoreus not only in being largely infertile but also in having at least some inner perianth-segments narrower overall, or narrower apically, than in R. crispus. Possible Rumex obtusifolius L. x R. rupestris Le Gall Three dock plants growing close together in the edge of a fen area at Penhale Camp were studied on 8 August 1995 and on 24 July 1996. Two of them that survived were studied again on several visits during July-September 1998, allowing herbarium material to be collected at various stages of development. Their puzzling combination of morphological characters and consistently low fertility (with less than 20% of nutlets developing) implied they were hybrids, but although the three plants are rather similar to each other, they show an odd mixture of features that has prevented confident inference of the parent species. All three plants grew as compact patches from stout rootstocks, with strong, erect flowering stems. In August 1995 one plant (field label E) had about 25 flowering stems up to 1-1 m tall, the other plant (G) had 11 stems up to 1-2 m tall. The basal and lower stem leaves were thick and fleshy, with strongly undulate margins; an immature basal leaf had petiole 7 cm, lamina 11 cm; the longest stem leaves were oblong-lanceolate and had petiole 6:5 cm, lamina 20-5 cm. The underside of the leaf midrib and some of its strongest veins had low conical papillae, recalling those in R. obtusifolius, but much smaller and less developed. The panicles were similar in habit to those of R. rupestris, with branches mainly rather erect (at 20—30° from main stem) and none widely divaricate. The inner perianth segments of the minority of fruits that ripen were wider than in R. rupestris, but narrower than in R. crispus (reaching 4-6 mm long and 3-0 mm wide), with a longer and more attenuate apex than in R. crispus, mostly with one or two short teeth on the basal margins. All three inner perianth segments on each well-developed fruit had a large swollen tubercle, that on one perianth segment being larger than those on the other two perianth segments. Dock species growing within 20 m of these plants were R. crispus, R. rupestris and R. conglomeratus, along with numerous R. X pratensis and two R. X rosemurphyae; the only other dock species within many hundreds of metres being two plants of R. obtusifolius and a few of R. pulcher. Nevertheless, the distinct teeth on the inner perianth segments imply that among the dock species occurring nearby, either R. obtusifolius or R. pulcher was one of the likely parents of the hybrids. However, involvement of R. pulcher seems unlikely, as the hybrids gave no evidence of 88 DoT HOLYOSK the divaricate branching, warty tubercles or other characters of that species. On the other hand, involvement of R. obtusifolius might be deduced from the presence of papillae on the back of the leaf midribs, albeit that these and other features of R. obtusifolius appear poorly developed. Other features of these plants imply that R. rupestris was one parent, including the thick leaves, large to very large tubercles and rather narrow inner perianth segments with more or less attenuate apex. Nevertheless, a supposed parentage of R. obtusifolius x R. rupestris does not account for the strongly undulate leaf margins, for which it is tempting to infer some involvement of R. crispus. Since numerous plants of R. crispus x R. obtusifolius (R. X pratensis) were present nearby. whereas the only two plants of R. obtusifolius found were several hundreds of metres distant, the characters of the three hybrids might therefore be best explained by inferring their parentage as (R. crispus X R. obtusifolius) X R. rupestris. Hybrids between R. obtusifolius or R. X pratensis and R. rupestris have not been reported before. Indeed, no “triple” hybrids have been reported for wild docks in Britain, although crosses involving three species have been produced experimentally and they are known in Europe (Lousley & Williams 1975). The absence of British reports of “triple” hybrids in Rumex might therefore result not from their absence but from the almost insuperable difficulties in identifying them from morphological characters. Williams (1971) suspected from field observations that R. x pratensis back-crosses with both parental species and this suspicion was strengthened because R. crispus X R. X pratensis has been produced in cultivation. Holyoak (1996) noted that the numerous R. x pratensis at Penhale Camp vary widely in fertility and in characters of the inner perianth segments, concluding that it is uncertain to what extent their marked variability is due to back-crossing or merely the expression in F,; hybrids of an independent assortment of varied characters from the parental genotypes. Overall, it seems likely that the three puzzling plants described above originated either from R. obtusifolius or R. xX pratensis hybridising with R. rupestris. Because analysis of their morphological characters alone may provide an insufficient basis to choose between these alternatives the hybrid combination is not named here. Rumex x trimenii Camus (Rumex pulcher L. xX R. rupestris Le Gall) A single plant of this hybrid grew close to numerous plants of R. rupestris at Penhale Camp from 1994-1996 (RNG); it was described and illustrated by Holyoak (1995). A similar, but smaller, plant was found in 1998 close to a different colony of R. rupestris. The nearest plants of R. pulcher to both of these hybrids were 200 m away and few in number. Both of the Penhale hybrids had low fertility although at least some apparently fertile fruits were surrounded by perianth-segments that enlarged after flowering. Their widely divaricate branches resembled those of R. pulcher and the influence of that species was also apparent from the reticulate venation of the perianth segments, the presence on them of marginal teeth and the warty surface of their tubercles. Influence of R. rupestris was apparent in the hybrids from the strong stems, the rather thick leaves of broadly lanceolate shape and the narrowly lingulate inner perianth-segments. Lousley & Kent (1981) describe a similar combination of characters in R. Xx trimenii. There appear to be three previous records of this hybrid in the wild, each of single plants: in v.c. 1b from east coast of Samson, Isles of Scilly (RNG) and New Grimsby, Tresco, Isles of Scilly (RNG) , and in v.c. 2 at Whitesand Bay, E. Cornwall (BM); it also arose spontaneously in the garden of the South London Botanical Institute (Lousley 1971, 1983; Lousley & Williams 1975; Lousley & Kent 1981; Margetts & David 1981; Holyoak 1995, 1996). DISCUSSION Rumex rupestris at Penhale Camp has apparently produced hybrids involving all four of the other dock species that grow in the same area. The total population of R. rupestris there was about 60 mature (fruiting) plants in 1994, but it had increased to 132 by 1998. Although the number of hybrids involving this species at Penhale was small (eight plants), they apparently exceeded 10 % of the total R. rupestris population during 1994-1996. Moreover, six of the eight hybrids were found in one small fen area (SW/768.569) that supported a maximum of 18 plants of “pure” R. RUMEX RUPESTRIS HYBRIDS 89 rupestris. Two more hybrids were found at Gear Sands alongside a colony of about 27 plants of R. rupestris. Most, if not all, of these hybrids produce at least some pollen and at least small amounts of viable seed. Hence, given the rather high frequencies of F; hybrids that have been found, there may be opportunities for introgression to occur between the dock species involved. The following discussion explores various possibilities of introgression between Rumex species at Penhale. POSSIBLE INTROGRESSION OF R. RUPESTRIS GENES INTO R. CRISPUS Lousley and Kent (1981) reported possible introgression of R. rupestris into R. crispus in the Isles of Scilly and at Kenfig, Glamorgan, noting that “in both these localities plants referred to R. crispus occur with exceptionally large, elongate tubercles recalling those of R. rupestris and indicating possible introgression”. However, these may have been merely the coastal taxon now treated as R. crispus subsp. littoreus (e.g. in Stace 1991). Nevertheless, plants of R. crispus subsp. littoreus growing at Penhale Camp in the same areas as R. rupestris also include some with unusually elongate tubercles and it was tempting at first sight to suspect introgression had occurred between these species. However, the Penhale plants also grew alongside numerous R. X pratensis (26 were counted in 1995) and possible back-crosses of these to R. crispus, the plants showing considerable variability in morphology and in fertility (cf. Holyoak 1996). Hence there is no need to invoke introgression from R. rupestris to explain variability in the R. crispus growing close to it at Penhale Camp. INTROGRESSION OF R. RUPESTRIS GENES INTO R. CONGLOMERATUS In August 1996 the small fen area at Penhale Camp with the two plants of R. x rosemurphyae also had about 40 plants of R. conglomeratus and at least six and perhaps as many as 13 of these showed characters approaching those of R. rupestris (Table 1). Because no evidence of R. conglomeratus showing similar characters was found in 39 plants studied from other parts of Penhale Camp or in a total of 125 plants from three localities elsewhere in West Cornwall, the presence of R. rupestris seems to be associated with occurrence of some of its characters in the coexisting population of R. conglomeratus (Table 1). Since two F, hybrid plants (R. x rosemurphyae) were present at the site it seems likely that some of the plants of R. conglomeratus had acquired genes from R. rupestris, presumably as a result of back-crossing from F, hybrids. Although fully adequate data on pollen fertility of R. x rosemurphyae are not available, freshly collected pollen from its Holotype mainly appeared well formed when examined microscopically in July 1998. The F, hybrid (R. x rosemurphyae) produces few fertile fruits so that few F, plants would be expected to occur, although some of its fruit has been successfully germinated in cultivation. In contrast, the plants of R. conglomeratus putatively introgressed with R. rupestris appear to produce mainly fertile fruits so these back-crosses would be expected to persist once established and this may explain the occurrence of at least six and perhaps as many as 13 such plants in one small fen area at Penhale Camp. RISK OF INTROGRESSION OF R. CONGLOMERATUS GENES INTO R. RUPESTRIS R. rupestris is regarded as a globally threatened species, which is included in the British Red Data Book (Wigginton 1999), placed on the “Biodiversity Short List” (B.S.G.R. 1995) and included in Schedule 8 of the Wildlife and Countryside Act, 1981. It is included in English Nature’s “Species Recovery Programme” which is being undertaken collaboratively with Plantlife’s “Back from the Brink” Project. Its population at Penhale Camp (137 fruiting plants in 1998) apparently includes at least 20 % of all R. rupestris plants currently known in the British Isles (cf. Daniels et al. 1998). In these circumstances any “leakage” of genes from other docks into R. rupestris populations would complicate attempts to maintain a favourable conservation status for “genotypically pure” R. rupestris. The potential danger is evident from several well documented instances of the loss of flowering plant taxa through introgressive hybridisation, either locally (DePamphilis & Wyatt 1990; Klier et al. 1991), or over the whole range of geographically restricted taxa (Reiseberg et al. 1989). However, although the data presented in this paper appear to show “leakage” of genes out of R. rupestris into R. conglomeratus there is no direct evidence for genes of R. conglomeratus entering the R. rupestris genotype. This asymmetry might be genuine and have arisen because hybridisation was asymmetrical with respect to male and female parentage, or because of differential mortality in hybrid products. 90 D. T. HOLYOAK: However, this apparently fortunate result might be seriously misleading, both regarding the situation at Penhale Camp and elsewhere. Some doubts arise because few R. rupestris plants could be studied (although at least 130 have now been studied in detail, of which 35 were scored for hybrid index frequency in 1996), despite the Penhale Camp population being one of the largest known. Thus, relatively rare, introgressed, fruiting plants of R. rupestris might be absent merely as a result of stochastic processes operating during germination or causing mortality during growth, so that viable seed with an introgressed genotype might nonetheless be present at low frequency in the local seed-bank. It is also possible that the mode of inheritance or mode of phenotypic expression of the characters used in deriving the hybrid frequency index means it is easier to recognise introgressed R. conglomeratus than introgressed R. rupestris. Although polygenic traits can be expected to show intermediate expression in hybrids, traits governed by one or two genes are more likely to show parental expression, which could include strong maternal effects. This may explain why, in a survey of morphological patterns in natural and experimental hybrids of flowering plants, Reiseberg & Ellstrand (1993) found that hybrids are no more likely to display intermediate morphological features than parental ones. The classic “hybrid frequency index” technique of Anderson (1936) can thus be viewed as applicable in only the minority of situations where hybrids are intermediate. Considered against this background, the present lack of morphological evidence for introgression of R. conglomeratus genes into R. rupestris might well be misleading. It seems likely that if a theoretical introgressed R. rupestris was once established in a population, further back- crossing and introgression into that species could easily occur, much as appears to have happened with introgressed R. conglomeratus. In view of such dangers, further work on hybridisation of R. rupestris with other docks is desirable, for which morphological studies might usefully be supplemented by genetic studies that should provide additional characters. Daniels et al. (1998) reported “preliminary” results of isozyme electrophoresis on samples of R. rupestris from south-western England, disclosing a high level of genetic diversity within some populations and significant differences between certain groups of populations. Nevertheless, it remains unclear how much of that genetic variation is intrinsic to R. rupestris and how much of it might be derived from introgressive hybridisation with congeners. This doubt becomes important if the species is to be introduced or reintroduced into the wild since it is then desirable to establish that genotypically “pure” R. rupestris plants are used. Otherwise, attempts to maximise genetic variation in the introduced populations might result in introgressed plants being chosen for introduction attempts. ACKNOWLEDGMENTS Thanks are due to the Ministry of Defence for access to Penhale Training Camp and to Major B. Andrews and Lt Col R. C. Taylor for assistance and information. The following botanists helped in field surveys of docks at Penhale or commented on specimens: Dr J. R. Akeroyd, M. Atkinson, T. Atkinson, I. Bennallick, Dr C. N. French, Dr P. A. Gainey, E. C. M. Haes, G. A. Holyoak, G. Kitchener, E. J. McDonnell, H. M. Meredith, R. J. Murphy, C. J. Neil and J. Stewart. Dr Akeroyd and Dr S. L. Jury (RNG) provided information on specimens in herbaria and A. Jones and Dr Q. O. N. Kay provided other unpublished information. T. L. Blockeel kindly prepared the Latin diagnosis. Thanks are also due to Zowie Keating for drawing the figures of Rumex xX rosemurphyae and to the B.S.B.I. for assisting with the cost of the drawings. REFERENCES ANDERSON, E. (1936). Hybridization in American Tradescantias. Annals of the Missouri Botanic Garden 23: 511-525. [BSGR] BIODIVERSITY: THE U.K. STEERING GROUP REPORT (1995). Volume 2: Action Plans. H.M.S.O.., London. DANIELS, R. E., MCDONNELL, E. J. & RAYBOULD, A. F. (1998). The current status of Rumex rupestris Le Gall (Polygonaceae) in England and Wales, and threats to its survival and genetic diversity. Watsonia 22: 33— RUMEX RUPESTRIS HYBRIDS 91 39. ; DEGRAEVE, N. (1975). Contribution a l'étude cytotaxonomique des Rumex. 1. Le genre Rumex L. sensu stricto. Caryologia 28: 187-201. s DEPAMPHILIS, C. W. & WYATT, R. (1990). Electrophoretic confirmation of interspecific hybridization in Aesculus (Hippocastanaceae) and the genetic structure of a broad hybrid zone. Evolution 44: 1295-1317. HOLYOAK, D. T. (1995). A rare hybrid dock Rumex xX trimenii Camus at Penhale Camp, West Cornwall. Sanctuary 24: 35. a HOLYOAK, D. T. (1996). Hybridization in docks (Rumex subgenus Rumex) in Cornwall. Botanical Cornwall 7: 8-26. JALAS, J. & SUOMINEN, J. (1979). Atlas Florae Europaeae. Distribution of vascular plants in Europe. 4 Polygonaceae. The Committee for Mapping the Flora of Europe & Societas Biologica Fennica Vanamo, Helsinki. . Kay, Q. O. N. (1996). The conservation of Rumex rupestris (Shore Dock) in Wales. Past, present and possible future sites and habitats for Rumex rupestris in South and West Wales. Unpublished Report of Countryside Council for Wales Contract Survey. KLIER, K., LEOSCHKE, M. J. & WENDEL, J. F. (1991). Hybridization and introgression in white and yellow ladyslipper orchids (Cypripedium candidum and C. pubescens). Journal of heredity 82: 305-319. LOUSLEY, J. E. (1971). Flora of the Isles of Scilly. David & Charles, Newton Abbot. LOUSLEY, J. E. (1983). Flowering plants and ferns in the Isles of Scilly, 2nd ed. (1975), revised by Clare Harvey (1983). Isles of Scilly Museum Publication no. 4. St Mary’s. LOUSLEY, J. E. & KENT, D. H. (1981). Docks and knotweeds of the British Isles. Botanical Society of the British Isles, London. LOUSLEY, J. E. & WILLIAMS, J. T. (1975). Rumex L., in STACE, C. A., ed. Hybridization and the flora of the British Isles. Academic Press, London. MARGETTS, L. & DAVID, R. W. (1981). A review of the Cornish Flora, 1980. Institute of Cornish Studies, Redruth. RECHINGER, K. H. (1993). Rumex L. (revised by J. R. AKEROYD), in TUTIN, T. G. et al., eds. Flora Europaea Vol. 1, 2nd ed. Cambridge University Press, Cambridge. REISEBERG, L. H. & ELLSTRAND, N. C. (1993). What can molecular and morphological markers tell us about plant hybridization? Critical reviews in plant sciences 12: 213-241. REISEBERG, L. H., ZONA, S., ABERBOM, L. & MARTIN, T. D. (1989). Hybridization in the island endemic, Catalina Mahogany. Conservation biology 3: 52-58. STACE, C. A. (1991). New Flora of the British Isles. Cambridge University Press, Cambridge. WIGGINTON, M. J., ed. (1999). British Red Data Books. 1. Vascular plants, 3rd ed. J.N.C.C., Peterborough. WILLIAMS, J. T. (1971). Seed polymorphism and germination, 2. The role of hybridization in the germination polymorphism of Rumex crispus and R. obtusifolius. Weed research 11: 12-21. (Accepted October 1999) o2 D. T. HOLYOAK APPENDIX 1 Methods used for hybrid frequency index measurements of Rumex rupestris, R. conglomeratus and intermediate plants from localities in West Cornwall, August 1996. The hybrid index was derived by scoring each of five characters on each plant. A score of 0 was assigned for character-states typical of R. rupestris, 3 for those typical of R. conglomeratus and 1— 2 for intermediate states. Hence, summing data for five characters, a typical plant of R. rupestris would score 0, a typical plant of R. conglomeratus would score 15. Data were scored only from undamaged plants with mature (drying) perianths. The characters and scoring systems were as follows: Length of longest inner-perianth segment: mean of 10 perianths examined from middle part of inflorescence; measurements made with eyepiece graticule to accuracy of + 0-05 mm; 0 = > 3-0 mm, | = 2-5—3-0 mm, 2 = 2-0—2-:5 mm, 3 = < 2:0 mm; Length of tubercle as per cent of length of longest inner-perianth segment: mean of 10 perianths examined from middle part of inflorescence (same perianths as for preceding character); measurements made with eyepiece graticule to accuracy of + 0-05 mm; 0 = > 60%, 1 = 55-60%, 2 = 50-55%, 3 =< 50%: Angle of main branches of inflorescence: modal value; angle measured from main stem (not from vertical); measured only for branches > 5 cm long; 0 = < 50°, 1 = 50-60°, 2 = 60-70°, 3 = > 70°; Number of bracts on longest three branches of inflorescence: using only branches > 10 cm long; 0 = 1-5 bracts, 1 = 6 or 7 bracts, 2 = 8 or 9 bracts, 3 = > 9 bracts; Crowding of whorls on main branches of inflorescence: estimated as ratio of inflorescence whorl width (flowers + their pedicels)/ interwhorl width along the inflorescence axis (length of stem clear of all flowers + pedicels); modal value for all of main inflorescences: 0 = ratio <1-0/1, | = ratio 1 -0—1-25/1, 2 = ratio 1-25—1-5/1, 3 = ratio > 1-5/1. Watsonia 23: 93—110 (2000) 93 Prize-winners to pariahs - A history of Japanese Knotweed s./. (Polygonaceae) in the British Isles J.P. BAILEY and A. P. CONOLLY Biology Department, University of Leicester, Leicester, LE] 7RH ABSTRACT The early history of the discovery and introduction to the British Isles of the alien invasive perennials belonging to Fallopia section Reynoutria (hereafter referred to as Japanese Knotweed s./.) are described from a study of documentary evidence and herbarium material. The role of the various agencies involved in the dispersal of these plants is discussed, as is the influence that contemporaneous gardening trends had on the ornamental use of Japanese Knotweed s. /. KEYWORDS: Fallopia japonica, Fallopia x bohemica, Fallopia sachalinensis, invasive plants, garden escapes, plant collectors. INTRODUCTION The group of plants referred to here as Japanese Knotweed sensu lato needs little introduction. Details of reproductive biology, cytology and distribution are given in Conolly (1977), Bailey & Conolly (1985, 1991), Bailey (1989, 1994), Bailey & Stace (1992), Beerling et al. (1994) and Bailey et al. (1996). The term Japanese Knotweed s./., may need further explanation. Since the amalgamation of Reynoutria and Fallopia there is no convenient way of referring to the complex of taxa formerly circumscribed by the name Reynoutria, which comprises F. japonica var. Japonica, F. japonica var. compacta, F. sachalinensis, the hybrids between F. japonica and F. sachalinensis (F. x bohemica) and between var. japonica and var. compacta as well as any backcrosses between the hybrids and their parents. It is generally known that these plants were originally garden plants, but it is not so well appreciated just how valuable an addition to the garden they were once considered to be. This paper seeks to build on the initial contribution by Conolly (1977), by giving a more detailed examination of just those taxa transferred from Reynoutria to Fallopia and using the year 1914 as a general cut-off point. It is our intention to deal with each taxon in turn, to detail its original discovery and the most probable source of the accessions that were introduced to Europe, and to try and unravel the somewhat bewildering array of synonyms that litter the scientific and horticultural literature. Using a combination of type and other early herbarium specimens, archive material, scientific and horticultural literature, nursery garden catalogues and accounts of plant collecting expeditions, the discovery and spread of each taxon is given in as much detail as possible. There follows a consideration of the gardening movements that contributed to the dissemination of Japanese Knotweeds to every corner of the British Isles. Since we are dealing with long-lived perennials spreading primarily by vegetative means, and easily outliving gardener and garden alike, it is important to have records of where these plants were once grown or available for sale. Large private gardens or estates, nursery gardens and botanic gardens are likely to constitute the initial loci from which further dispersal by exchange, gift, sale or disposal took place. With regard to the hybrid F. x bohemica, it is useful to know where both parental taxa were grown together, especially where the F. sachalinensis involved is male-fertile. We will now consider in detail, by taxon and in chronological order, the early history of these plants as aliens in the British Isles, as we track the riotous career of F. japonica from gold medal holder in 1847 to proscription in the 1981 Wildlife & Countryside Act. 94 J.P. BAILEY AND-AjP2CONOELY FALLOPIA JAPONICA VAR. JAPONICA (HOUTT.) RONSE DECRAENE A MOST PROMISING START Fallopia japonica was awarded a gold medal in 1847 by the Society of Agriculture & Horticulture at Utrecht for the most interesting new ornamental plant of the year, according to the 1848 catalogue of Von Siebold & Company of Leiden, the specialist importers of plants from the East (Siebold 1848). An accompanying seven line foot-note in this price-list extols its great vigour, its combination of ornamental and medicinal use, its value in protecting young plantations from wind and sun, its edible young stems and leaves and a rhizome highly valued in Japanese and Chinese medicine. Such a plant, newly imported from Japan, would obviously command a high price. 500 francs would purchase a “mother” plant and 25 strong plants. In order to put this sum into perspective, the same catalogue offered 250 plants of Hydrangea japonica or 500 plants of Wisteria sinensis for the same amount! If, by the time of Siebold’s 1856 catalogue, the price had plummeted (100 plants for 25 francs), the rhetoric certainly had not. In addition to the benefits listed above were added stabilisation of sand dunes, gracious flowers suitable for bouquets, a forage which cattle loved to eat, melliferous flowers which gave bees their winter sustenance and, last but not least, dead stems that could be used to make matches! There was, however, no element of exaggeration in the claim that the plant was “inextirpable”. SIEBOLD, HIS EXPEDITION AND HIS COMPANY Phillipe von Siebold had for his time absolutely unparalleled access to the then closed country of Japan. Appointed as doctor in residence to the Dutch trading settlement in Japan, he stayed there from 1823-1829. His medical knowledge gained him many friends and high level contacts in Japan. Siebold used every opportunity to collect plant and animal specimens, and had friends and students collecting material from all over Japan. After a series of fascinating peregrinations (shipwrecks, imprisonment and civil war), he was not finally reunited with all his live Japanese plant specimens back in Leiden until 1841 (Kimura & Kitamura 1977). Here he founded, in 1842, a Royal Society for Encouragement of Horticulture (Anderson 1951) - apparently just a respectable cover for the commercial activities of his Von Siebold & Co. The horticultural company owned not only the specimens brought back from his sojourn in Japan but also a governmental monopoly on any subsequent introductions from Japan and Java. By 1847 this Royal Society had severed its connections with Von Siebold & Co. Fortuitously, this connection allowed the survival of a more-or-less complete collection of sales lists in the pages of their journal: Jaarboek van de Koninklijke Nederlandsche Maatschappij tot Aanmoediging van den Tuinbouw. From a study of these lists it is possible to reconstruct the very early days of this plant in Europe. The name Polygonum sieboldii Reinw. first appears in 1848, and this name is given as a synonym of Polygonum cuspidatum Sieb. & Zucc. by De Vriese (1849) (see Table 1). Von Siebold & Co. presumably considered the eponymous P. sieboldii to be a better advertisement than the name given to it by Siebold and Zuccarini in 1846. In the 19th century gardening literature the plant is generally known as P. sieboldii. UNTANGLING THE KNOTWEED NOMENCLATURE The history of Fallopia japonica really begins in the 18th century, when it was described as Reynoutria japonica by Houttuyn (1777), presumably from dried material brought back from Japan by Thunberg. Considering that both Houttuyn and Thunberg wrote substantial botanical books, it is surprising that the world in general was unaware of the plant until it was described by Siebold & Zuccarini in 1846. Even as late as 1895 the first edition of Index Kewensis lists Reynoutria as incertae sedis. It was not until 1901 that Makino realised that Reynoutria japonica of Houttuyn and Polygonum cuspidatum of Siebold and Zuccarini were one and the same; on these grounds he made the new combination Polygonum reynoutria, but without sufficient grounds for replacing P. cuspidatum. There are various reasons why this important taxon was lost to science for seven decades: one was that Houttuyn’s Natuurlijke Historie (1777) was written in Dutch, not the most accessible of languages. He also erroneously described the plant as having ten stamens, thus putting Reynoutria under “Decandria Trigyna”’, isolated from the other Polygonums. However, Houttuyn does have a good clear illustration of the plant in his work. Thunberg (1784) does not recognise Reynoutria in 95 HISTORY OF JAPANESE KNOTWEED os ee ih MP Gf A 7 VUGYCTV Lt Wb KL, Sf + FIGURE |. Engraving from De Vriese (1849) of Siebold’s plant. 96 J.P. BAILEY AND A. P. CONOLLY TABLE |. EARLY SYNONOMY OF FALLOPIA JAPONICA Name Authority/user Current name Reynoutria japonica Houttuyn 1777 Fallopia japonica Polygonum multiflorum © Thunberg 1784 Fallopia japonica P. pictum Siebold 1841 = F. japonica var. compacta ? P. cuspidatum Siebold & Zuccarini 1846 F. japonica var. japonica P. sieboldii used by de Vriese 1849 F. japonica var. japonica P. reynoutria Makino 1901 F. japonica his Flora Japonica, but has his own taxon, Polygonum multiflorum (Thunb.), which is obviously not in Natuurlijke Historie. This is a perfectly good species, now known as Fallopia multiflora, and one of the link species between the climbing and non-climbing members of the genus Fallopia. Photocopies of some of the Thunberg specimens from Uppsala (UPS) have been seen. Herb. Thunberg 9706 is labelled Polygonum multiflorum (probably in Thunberg’s hand) and is indeed that taxon. However, specimen 9705 bears the label “Polygonum multiflorum a’, but is clearly a specimen of Fallopia japonica. Ironically, the Houttuyn specimen of F. multiflora (Herb. Delessert collection Burman s.n., G) is labelled “Polygonum (Helxine) Japonicum Mihi. Vid Nat His. VII Pl. 49 Fig 3” in Houttuyn’s hand (?). However, the name does not tally with the text reference, which refers to the “Japansch” version of Polygonum chinense L., whereas the figure depicts F. multiflora. In summary, although Houttuyn and Thunberg both handled specimens of F. Japonica and F. multiflora, the former regarded F. multiflora as the Japanese form of P. chinense, whilst the latter regarded F. japonica as a variant of F. multiflora. Further, it was only possible to appreciate Thunberg’s circumscription of P. multiflorum by reference to his herbarium specimens! Since there is no evidence of live F. japonica being in Europe in the 18th century, in order to find the source of the live introductions to Europe, we need to look at the Siebold’s collections. The Rijksherbarium at Leiden sent us four specimens; the one labelled “Herb. Lugd. Batav. No. 908.174 826 Polygonum cuspidatum S. & Z. Legit v.S.” is probably the holotype of P. cuspidatum. The specimen is male-sterile and very similar to our British plants. Further evidence that this is the source of our plants is provided by De Vriese (1849) and the excellent detailed engraving (see Fig. 1). De Vriese heaps praises upon the plant and states that it is the sole property of Siebold who has had it growing for many years [at Leiden]. The red spotted stem and the zig- zag appearance of the stem are also mentioned. The figure (Fig. 1), drawn from life from Siebold’s plant, shows leaves with truncate bases closely matching the appearance of our plants. It is interesting to note that it is also male-sterile and bears good seed with embryos; this raises the question of what could have pollinated it in the absence of F. sachalinensis? The most likely source is Polygonum pictum (= F. japonica var. compacta?) which had been in cultivation at Leiden for some time. THE ARRIVAL OF FALLOPIA JAPONICA IN BRITAIN Although the first date that the plant was commercially available in Europe is undeniably 1848, the date that it actually crossed the Channel to Britain was not known until now. The literature contains two conflicting dates: 1825 (Conolly 1977 ex Synge 1956) and the late 1840s (Bailey 1994). It appears that in a sense, both dates are correct! Morren (1849) gives a résumé of the DeVriese paper of 1849 which describes Polygonum cuspidatum Sieb. & Zucc., and says that it was introduced by Phillipe von Siebold and that it was his sole property. Lindley & Paxton (1850), however, in reply to the Morren article, declare that the plant was actually introduced by the Horticultural Society of London around 1825. The plant had been thought to be Houttuynia cordata, and had been planted in an artificial swamp at their Chiswick garden; not surprisingly, it had failed to thrive, though an illustration is provided of the one time that it did manage to flower. Lindley & Paxton ended with the comment that they considered it unlikely that such a plant would bind blowing sand together in the manner described by Dr Siebold. Significantly, they record that the Chiswick plant came from China and indeed their illustration shows quite clearly the acuminate leaf base so characteristic of Chinese accessions of F. japonica. This was not the end of HISTORY OF JAPANESE KNOTWEED 97 the cross-channel dialogue. Groenland (1858) writing in La Revue Horticole (Paris), expresses surprise that “Fallopia japonica” could remain unnoticed for 20 years in the garden at Chiswick, without even flowering, adding that in Japan the plant is used to bind sand-dunes. With the knowledge that F. japonica had no name in 1825, since this predates the Sieb. & Zucc. name, and the earlier Reynoutria combination had been lost until 1901, one of the authors (J.P.B.) had severe reservations regarding an introduction date of 1825. In the event, the mis-identification of the Chiswick plant as Houttuynia neatly side-stepped this objection; (that it should have been confounded with a Houttuyn eponym is rather a nice irony!). So, although the earliest recorded live plant of F. japonica in Europe was in the Chiswick garden of the Horticultural Society, this was a Chinese accession. Chinese plants are morphologically distinguishable from the Siebold clone of F. japonica, and, even if the Chiswick plant was ever distributed to any extent, it is doubtful that any has survived; it has had no discernible role in the Japanese Knotweed invasion of Europe. Siebold’s nursery at Leiden, where the plant was actively sold and heavily promoted, even to the extent of offering a 25% discount to botanical gardens, is considered the initial locus of the F. japonica clone that has so effectively invaded much of Europe. The manuscript “Inwards Book” 1848-1858 at the Royal Botanic Gardens Kew records that Polygonum sieboldii was received on 9 August 1850 as part of a large batch of plants from M. Siebold of Leyden. There is appended the following note: “They are intended to be in exchange for new China and Japan plants. But on account of the bad selection he is written to, telling him that only 6 of them are probably new to us”. The plants were probably sent unsolicited in the hope of receiving further novelties in return. The Kew herbarium has a male-sterile F. japonica var. Japonica specimen labelled “Hort, Kew 1857”, which is presumably the Siebold introduction. The Kew archive also contains sets of “outward Books” which record a commerce absolutely unimaginable to modern-day gardeners attempting to obtain plants from the Kew collection. Collections and cabinets of plants were being dispatched alike to Honourables, military men and commercial nursery gardens apparently without distinction. However, a search of the Outward Book 1853-1856 has failed to reveal a single reference to Japanese Knotweed. This does not mean that Kew took no further role in the distribution of F. japonica, but that only the choicer plants were listed individually, there being plenty of examples of collections of un-named herbaceous plants being sent out. The archive at the Royal Botanic Garden Edinburgh also has a number of manuscript accession ledgers known as “Plant Books”. The 1849-55 Plant Book revealed that, on 26 April 1854, Polygonum sieboldii was received from Messrs Jackson & Son Nurserymen of Kingston. This, incidentally, is the earliest record of a British nurseryman offering Japanese Knotweed for sale by a very long way (see Table 2), and it is tempting to think that, with the proximity of Kew and Kingston upon Thames, the plant was supplied to Jacksons by Kew. Thus in the year 1850 began the inexorable spread of Japanese Knotweed throughout the British Isles. Before moving on, it is worth considering the fate of Siebold’s Garden of Acclimatisation at Leiden. In the year of his death, 1866, it boasted nearly a thousand different species and varieties. By the time the English horticulturists F. W. Burbidge and P. Barr visited in 1883 they found a neglected jungle, overrun by none other than - Japanese Knotweed (Sharman 1990). F, JAPONICA VAR. COMPACTA (HOOK. F.) J. P. BAILEY The 1844 catalogue of Von Siebold & Company records that a taxon called Polygonum pictum Sieb. was first introduced to Europe in 1841. It is thought that Siebold’s Polygonum pictum is actually the dwarf variant of F. japonica. This is based on two supporting pieces of evidence. Firstly its height is given by Siebold as 60 cm, and secondly the name is given as a synonym of P. cuspidatum by De Vriese in 1849. That aside, little is known about the early history of var. compacta - it is certain that the name P. pictum has not yet been found in the British botanical literature. Hooker (1880) gives the taxon specific rank as P. compactum, but this treatment has not been widely followed. It is a decorative garden plant with a dwarf habit, crimped leathery “square shaped” leaves, and usually growing up to about one metre, but with occasional records of much shorter plants in cultivation (c. 30 cm high). Male-sterile plants often have a conspicuous red pigmentation, making them more desirable in the garden. Unlike var. japonica, both male-sterile and hermaphrodite individuals are found. It was recommended by Gertrude Jekyll, who included 98 J. P=BAILEY AND A. -PZCONOLEY. “this dwarf Japanese Knotweed” in some of her planting schemes, such as the garden at Long Aston, Birmingham (Bisgrove 1992). The earliest record for cultivation in the British Isles is from John Wood’s garden at Kirkstall Abbey, Leeds in 1881. In 1883 it was grown in Mr Scott’s garden, Denzell (Dunham Massey, near Altrincham, Cheshire). Two 1911 herbarium specimens of plants grown at Charles Bailey’s garden at Cleeve Hill near Cheltenham (v.c. 33) are labelled as having been supplied by Smith’s nursery. Richard Smith of Worcester was a well known nurseryman at this time (see Table 2). These latter are labelled as P. alpinum, but it is not clear whether this error was Bailey’s or Smith’s. (All these herbarium specimens are now at MANCH). It is now thought that the plants we refer to as var. compacta are actually particular alpine variants of F. japonica. Shiosaka & Shibata (1993) in their transplant experiments clearly show that there is a genetic component governing both height and time of flower initiation in this taxon. As the altitude from which the plant was collected increased, so the final stature of the plant diminished and the initiation of flowering became earlier. Plants from 2,400 m on Mt Fuji only grew 10 cm tall and initiated flower buds in May. We have noted very small var. compacta plants in Britain at Arno’s Castle, Bristol (LTR) and Wisley. L. H. Bailey (1949) lists “Polygonum Reynoutria Makino” as a ground cover plant 4-6 inches high. Synge (1956) distinguishes “Polygonum Reynoutria Makino”, a synonym of P. cuspidatum, from “Polygonum ‘Reynoutria’ of gardens, non Makino”, a male-sterile var. compactum with red fruiting perianths. Amos Perry in 1936 received an R.H.S. Award of Merit for his “P. Reynoutria”, 18 inches high (according to his 1937 catalogue). Its crimson fruiting sprays were much admired in the Wisley September notes in the 1940s (Anon. 1943). The “P. Reynoutria” at E. A. Bowles’ rockery at Myddleton House, Enfield is reported as having come from this source. It is quite possible that there were multiple introductions, as these dwarf variants would be encountered by any botanical expedition to the mountains. The earliest recorded escape was 1915 Melrose Abbey (E). The plant is still rather rare, and there are few instances of naturalisation; to date we have only mapped 30 records on a 10 km square basis, including garden plants. When found it is usually only in small stands and is not considered to be very invasive. F. SACHALINENSIS (F. SCHMIDT EX MAXIM.) RONSE DECR. F. sachalinensis is a native of Japan, Sakhalin, and possibly also the isolated Ullung-do Island between Korea and Japan. Early collections were made in the 19th century from both areas, but the exact dates and mode of entry to Europe are not always clear. Apart from the collectors own accounts, Bretschneider (1898) gives the most reliable account of expeditions to these areas. The Russian expeditions were sent from St Petersburg, and the British were sponsored by Kew, privately, or by nursery gardeners such as James Veitch & Sons (Veitch 1906). F. sachalinensis was discovered on Sakhalin by Dr H. Weyrich. He was the surgeon on the Vostok under the command of Capt. Lieut. Rimsky-Korsakov, on the Russian Naval Expedition to eastern Asia (1852-55). [Incidentally the composer N. A. Rimsky-Korsakov was the younger brother of the commander of the Vostok; initially his “..heart was set not on music but on a career in the navy in emulation of his brother 22 years older than himself’, (Grove, ed. Sadie 1980)]. During the survey of the west coast of Sakhalin, in early September 1853, Weyrich landed at Noto-sama, where he collected F. sachalinensis from damp river banks. The Russian Naval Expedition returned to St Petersburg in 1855 and Weyrich’s plants went to the St. Petersburg Botanic Gardens (Bretschneider 1898); his F. sachalinensis was described and named by F. Schmidt in Maximowicz’s Primitiae florae amurensis (1859). The type description records that the fruits were immature, though our examination of the type specimen (LE) contradicts this. Further collections of F. sachalinensis (BM; K) were made from Sakhalin by P. von Glehn in 1861, who, with F. Schmidt, made the first extensive survey of the flora of Sakhalin, during 1860- 1862. They arrived back at St Petersburg in early 1863 (Schmidt 1868). Material of Japanese origin was collected by C. J. Maximovicz on his second expedition to eastern Asia 1859-64 (1861 Hakodate, Japan BM; K). He arrived back at St Petersburg in July 1864 with an enormous collection of living plants, seeds and herbarium specimens. HISTORY OF JAPANESE KNOTWEED 99 Regel (1864) reported that F. sachalinensis was growing in the Botanic Garden at St Petersburg by 1864 and that it was able to overwinter there. At this date he can only have been referring to plants of Sakhalin origin. Regel (1874) further suggested that the St Petersburg F. sachalinensis originated from seed collected by Schmidt, and was then distributed widely to European botanic gardens. Certainly the herbarium specimen from the Glehn and Schmidt expedition (K) has good seed. So, whilst it is certain that the first F. sachalinensis at St Petersburg was indeed from Sakhalin, it is not entirely clear whether it was from the collection of Weyrich or Schmidt or both. Certainly by the time of Mr Bull’s importing it to Britain (see below), both Japanese and Sakhalin plants could well have been available in St Petersburg. Edouard André (1893) claims to have been the first person to have brought F. sachalinensis to the notice of the horticultural world. André relates (André 1870) that when in the Garden of Acclimatisation in Moscow in 1869 he was struck by various plants with big foliage recently introduced from the most distant parts of the Russian empire. One such plant that particularly took his interest was P. sachalinense, which the head gardener, Monsieur Desmurs, informed him had come from Sakhalin Island (André 1893). Concerning the introduction to Britain, Hooker (1881) recalls that it had been in cultivation at Kew for about 20 years and that it had probably come from the collections of Wilford or Oldham. This is indeed possible, as there is an 1859 Wilford specimen of F. sachalinensis (labelled as P. cuspidatum) from Hakodadi (Japan) at K. The Kew “Plants Inwards” book for 1859-67 records two sendings from Wilford in early 1860, but neither mentions P. cuspidatum or P. sachalinense by name. The same ledger records the sendings of Oldham, also from Japan, in 1862-3. These are not generally itemised, but seem to be primarily seed and spores; the April 1862 package being dismissed as *..mostly common and well known species scarcely any germinated”. We have not located any Oldham herbarium specimens of F. sachalinensis, but he cannot be ruled out as a potential source of seed of this taxon. Whatever the complexities and confusions of the early history of the introduction of live plants or seed from Sakhalin and Japan to the West, whether via Russia or Kew, we do have good documentary evidence of when this species was first available to the gardening public in Britain. The 1869-70 catalogue (No. 48) of William Bull of Chelsea records P. sachalinense under “New plants offered for the first time”. It was offered at the then considerable price of 7/6d per plant, along with the assurance that “....coming from the Amoor country will be perfectly hardy”. [At this period the “Amoor country” (Amur) included Sakhalin Island; F. sachalinensis is not known to occur on mainland Amur]. The 1871 (No 59) catalogue notes that it “..is perfectly hardy”, which demonstrates that they had overwintered it in Britain. That it was available in commercial quantities to unleash on an unwary public by 1869 suggests that it had been in cultivation for a year or more by then. Charles Maries, when in Japan in 1878-9, collected further material from the central mountains of Japan, whilst employed by James Veitch & Son (Veitch 1906). We have seen a herbarium specimen dated 1880 at Kew from this collection. If live material was brought back, it would have added further to the genetic diversity of this taxon in Britain. F. sachalinensis was also recommended, particularly on the continent, as a forage plant for cattle (André 1893). In Britain it was planted for this purpose on “oozy hillsides in Yorkshire” (Davies 1896), though there is no record of any such plantings surviving there. It was even introduced to Bengal for this purpose, and as a riverbank stabiliser, in the late 19th century, where it was known as Knot-grass or Machute (Christy 1893). We have recently heard that it still survives along rivercourses in Bengal (W. Griffin, pers. comm.). The Yokohama Nursery Company was in 1907 offering 100 lb of cleaned F. sachalinensis seed for $100, though we have not heard of any plantings on this scale. F. sachalinensis also has certain pharmaceutical properties, and is used commercially in Germany for the manufacture of a fungicide (H. H. Heine, pers. comm.). Early plantings in the British Isles are known from Mr Garnett’s Penketh (near Warrington) 1892, Eastnor Castle (Herefordshire) 1886, Cosmo Melvill’s garden at Brook House, Prestwich (Lancs.) 1901 and from a horticultural specimen of 1874; all Grindon (MANCH). In the 1890s Surgeon-General Cook grew both sexes of F. sachalinensis in his garden near Lydney (Gloucs.) (K). F. sachalinensis plants in Europe possess much more genetic diversity than F. japonica var. japonica (Hollingsworth 1998). This could be the result of multiple introductions from widely separated areas or the introduction of seed rather than rhizomes, or a combination of the two. 100 J.P BAILEY: AND Az P_..CONG@ELY FALLOPIA x BOHEMICA (CHRTEK & CHRTKOVA) J. P. BAILEY F. x bohemica is the name given to hybrids between F. sachalinensis and F. japonica var. Japonica or F. japonica var. compacta. It was first described in 1983 in a Czechoslovakian wildlife magazine (Chrtek & Chrtkova 1983). It was not until 1986 that, at Leicester University, we became aware of this; we had been independently describing the hybrids on morphological and cytological grounds since the early 1980s (Bailey & Conolly 1985). A.P.C. had, however, earlier been suspicious of certain specimens from Par (Cornwall) and Marston (near Oxford), the determinations of which had over the years alternated between F. sachalinensis and F. japonica. These suspicions were heightened by plants grown from seed sent by European botanic gardens, which morphologically did not match the taxon they were sent as, and which were subsequently found to be hybrids. Concentration on plants previously thought to be male-fertile F. japonica var. Japonica, and which were also subsequently found to be hybrids, highlighted the morphological features of the hybrid to us. Bailey er al. (1996) summarise the current knowledge of the hybrid’s history and distribution in the British Isles. None of the locations listed there predates the South Wylam, Co. Durham (v.c. 66) plant which has been established since at least 1954 (Sowerby, B. BM). For the earliest British record, 100 years prior to its description, we are again indebted to the Grindon Collection at MANCH. Here there are two collections (dated October 1872 and September 1876) from Manchester Botanic Garden, on a single sheet labelled P. sieboldii, both of which we have determined as F. x bohemica. A 1911 male-fertile specimen of F. x bohemica from Charles Bailey’s garden “Haymesgarth” at Cleeve Hill near Cheltenham (v.c. 33) is especially interesting as 1t came as a root from Smith’s Nursery, Worcester (see var. compacta section). This is direct evidence that a nursery garden was supplying this taxon at an early date. The initial hybridisation could also have occurred there, and such nursery establishments are likely origins of the hybrid in general. The plants certainly had plenty of opportunity for hybridisation, as the parental taxa had been growing together there since at least 1880 (Table 2). If Smith’s were sending out F. x bohemica widely, one would expect that the same male-fertile clone could be well-established in widely separated geographical locations; this is being tested by molecular work currently underway at Leicester. Hybridisation must have occurred on a number of occasions, whether in nursery, botanic or private gardens or even spontaneously from naturalised plants. This is borne out by the fact that there are at least five genotypes of F. x bohemica known from Britain, 4x, 6x and 8x male-fertile plants and 4x and 6x male-sterile plants (Bailey et al. 1996). THE “COOK MAMMOTH” ENIGMA Catalogues of V. N. Gauntlett & Co. at Redruth from the early 20th century proudly announce: “Polygonum ‘cooki’ - new mammoth species, introduced by Surgeon-General Henry Cook, from North America some years ago, who writes us that this variety has formed an enormous clump in his garden in Gloucestershire, of quite tropical appearance, with culms 16 feet high and 5 inches in circumference. It is perfectly hardy and, like the species, spreads rapidly, making a splendid plant for the wild garden or woodland drives, etc. We have secured the whole stock of this plant. 7/6d.” (Gauntlett’s Hardy Plants Worth Growing, Cat. 88 p. 61). This is the earliest record that we have located. Catalogue 88 is the earliest from the Japanese Nurseries, Redruth which lists this taxon, and it predates the move of the company to Chiddingfold, Surrey in 1906/7 (Willson 1988). Catalogue 88 cannot be earlier than 1901 since it refers to His Majesty’s Government. At Chiddingfold they advertised their “P. Cookii” up to the late 1920s/early 30s. From 1909 onwards, a photograph of “P. Cookii” is added. This photograph is taken from some distance and is not very clear, but depicts either F. sachalinensis or F. x bohemica. Were this plant actually to be F. x bohemica we have the fascinating possibility that the catalogue entry constitutes valid publication of an earlier name for F. x bohemica! Perry’s of Enfield were also selling P. “Cookii” between 1935 and 1937. Henry Cook, in A Gloucestershire Wild Garden (1903), describes his garden “Priors Mesne”’, near Lydney. Here he grew four large “Polygonums” in several places, including both male and female Polygonum sachalinense on either side of a pond, with P. cuspidatum/sieboldii close by. Thus there was potential here for on-site hybridisation and production of F. x bohemica. Cook also notes that Kew had told him that they did not have the male P. sachalinense at the time of writing HISTORY OF JAPANESE KNOTWEED 101 (1899): he swiftly rectified this by sending them a specimen the same year. This was followed in 1901 by a specimen (K) that we have now determined as F. x bohemica. Whilst we have established that Cook had both ?. sachalinensis and F. x bohemica in his garden prior to 1904, when Gauntlett’s would have collected their mammoth Polygonum, there is unfortunately nothing to link either taxon to the Polygonum being distributed by Gauntlett’s as “Cookii”. Thus it is not known whether the plant taken by Gauntlett’s from Cook’s garden was F. x bohemica or just a large clone of F. sachalinensis. In any case the herbarium specimen (K) is good evidence of F. x bohemica being a valued garden plant. Whether it originated in his garden, spontaneously or from deliberate planting of seed, or even in North America may never be known. PROMOTERS AND PERPETRATORS Having once established the probable date and place of arrival of these taxa in the British Isles, we are then set the task of explaining the current widespread distributions of F. japonica var. japonica, F. sachalinensis and F. x bohemica. For this we must turn our attention to the enthusiastic gardeners of the time, the commercial nurseries that pandered to their insatiable demand for novelty and the botanic gardens which were involved in widespread exchange with the former. THE ROLE OF COMMERCIAL AND BOTANIC GARDENS When looking at the dispersal of a garden plant an obvious place to start is the gardening literature. When available, herbarium specimens are of course direct evidence. The Nursery garden catalogues themselves are good primary sources of data. Although some catalogues were quite substantial publications (e.g. Gauntlett’s, Veitch’s), they are quintessentially ephemeral items, and interest would generally be restricted to the current issue. Hence it is not unusual to come across undated and unnumbered catalogues. Another difficulty is, of course, that few people had the foresight to save them. The Royal Botanic Gardens at Kew and Edinburgh and the Royal Horticultural Society's Lindley Library have substantial collections, but few other than the most famous nurseries are represented in any quantity, and complete runs appear virtually unknown. This is hardly surprising when considering the vast numbers of nurseries in business in the Victorian period. Even scarcer are catalogues from the 1850s and 60s - the period of the initial dispersal. Apart from those of Von Siebold & Co., which were fortuitously republished in a contemporary journal, we have seen no catalogues dated earlier than 1869. However, in spite of the fragmentary nature of the resources available, we have gained evidence of a substantial nationwide trade in these plants (Table 2). Nurseries offering them are well distributed throughout the British Isles - not that this was a prerequisite for effective dispersal, since a highly efficient postal service existed during the period under consideration. This list also provides a historic record of where and when F. japonica and F. sachalinensis were being grown together, giving the potential for hybridization. The prices generally follow a similar pattern, a high price for novelty followed by a sharp drop, as illustrated by F. sachalinensis in the Bull catalogues and Cook’s mysterious Polygonum offered for over 30 years by Gauntlett’s. Another point of interest is the scale of the trade in these plants: Sieboid, for instance, giving a price per hundred F. japonica plants and the Yokohama nursery offering F. sachalinensis seed in 100 Ib lots. Gauntlett’s and others were offering F. sachalinensis by the dozen - clearly not aimed at the ordinary suburban garden. Another point is the sheer length of time that these plants were retained on sales lists in spite of their true nature being known. Though we have not been searching actively for latest dates, Table 2 shows that F. japonica was on sale for at least 87 years! Indeed, the mottled variant of F. japonica, “P. spectabile”, was still in Beth Chatto’s list in the 1980s. GARDEN NOMENCLATURE Due to the mixed usage of horticultural and scientific names and the idiosyncratic application of some epithets, a certain amount of botanical licence has been employed in the interpretation of these taxa in the Nursery catalogues. Although Siebold was one of the authorities for P. cuspidatum, he used the eponymous P. sieboldii Reinw. in his catalogues, and it is under this combination that all the earliest references occur. As the compacta epithet was not published until 1880, another name was needed to differentiate the two varieties of F. japonica in the earlier catalogues. One way of doing this was by keeping sieboldii for the normal plant and using J.P. BAILEY AND A. P. CONOLLY 102 udas Jou ansoyeye7 J. OOT$ pods uedee pouvs[o qiool LO6I ‘eUeYOYOK “pyy] ‘O_ ‘Arasinyy eueYyoyo XK Sposa] qn[D jury ApieH Spoon 3 Yor P6 yore po Yous pg 6CO6T Hd “[[PISIDS ‘f POOM suspley Pleyullyl = yore pg z : L681 Foysoy sO) ‘OD ® core 961 2 6L8] UsoMjoq yard pg/] YIed P6 yord pG nq ‘poyepun JOOX7 SUOS 2 Laqoy “Yo3 A, SOHOSINN [IH Asteq ; qoes -/T eee 7/1 Yor? P6 (9'ON) 1681 KIMON ‘Luts yoes -/] yore po yore p6 18-0881 T9JSQOIOM, preyory “YIUs sede (1H o8pq ‘pod -AaTfoAy'D porddng + (OO6T od) J+ lossy ‘OD 29 Ystpurrs ; : yore A] AST squeyd OO] 9S8I uople’] ‘A'd WOA “OD 2 pfogets OS8I Vaiuodpl 7 YUM Moy Ode poyddng : / yoes HC Aoos/swueyd 97 87s uopre’] ‘A'd UOA “OD 2% Plogels AJOSINN BOTSIOATY - yoro -/T Yyord Pb yord pr 8681 - younyoysuyD doLINVYA] ‘preyoLig DIAINOUAAL ‘q SE P[AJUF ‘ISNOH UOJ[PpAJ ‘sapmog “W"q 0} Biopduos payddng yard pg/| Yoed Pb yoed -/¢ Yoed Pb L/DE61 ppyugq soury ‘ALIdg Sjueyd ysreul, se post] : yore 9/7 93 -/T A yoro 9/1 - P6 6L81 SUNOO], Waqoy oyied 7 yors p6/T - Yoee P6 [681 ysinquipy SUOS 29 SPULOYL, “USAYIoJ soueyy, uodn voiuodol -7 WM ysinquipy Ody poryddng + (PSST) +4 UO SSUTY] UOS 29 \L “‘UOsyors - F yors Jdor yore yd¢] 8L8I ynjay IPIWTYIS 2 oseeH SoLiasinN asoueder SS I : “/T (SO€61) TOL ON — PloysurppryD ‘NA ‘Hepunesy SOLIOSINN' osourder ovo O// Yoed P6 - yore -/T (~-ZO6I 2) 88 ‘ON yNIpsy ‘NA ‘nopuney YMOISIOpUN PUL SIDAOD IO] TquIINS - yord pg Yoed PS yord pg LO6I Jaysayg SOLIDSINN S,UOSyoIq SOLIDSINNY YuRg ATaWIoD - + - ~ Joquinu/ajep ON ysinquipy ‘0D X= Jaseiy ‘ wWeysuTuuND udzop wnoiuodpl ‘uks - - Jod -/¢ - L061 usapjioqVy souler “Iayood SOLIDSINN P[OYMO'T - - Yoed P6 yord P6 Jaquinu/ajep ON Ad[MBID SuOS 2» ‘f ‘TeayO : - - yore -/T L681 dUOISPIL|| ‘Ody ‘preAung P9/L E az 8P “ON 0L-6981 vosfoy wey Td . P9/T at “/1 6681 WO 29 ‘souref “Urey Io g - yord p9/] yora -/] -/] 8681 yo X uOSs 2 solUeL ‘asnoyyoeg - yord p9/] - yore -/T C681 yO X uoS 27 sows ‘asnoyyorg sqUOUMOD, «,, Y009,, ~~ sIsuauypyovs — winjovdwod pomuodpl JOQUETN Ayyeoo7 AJOS.INN INSOTLIVS JO yep "TS CHAMLONY ASHNVdVE£ dO NOILVNINASSIC NI GHA TOANI SNHCaVD ASHSANN AO LSTT ¢ ATEaAVL HISTORY OF JAPANESE KNOTWEED 103 cuspidatum for the dwarf, as was done in Smith’s catalogue 1880/81. The name P. reynoutria was also applied to a dwarf variant earlier this century. The Giant Knotweed did not provide any such problems, as it was always listed under its correct name (even though spelt in a number of creative ways). There is considerable evidence from the literature and herbarium specimens that plants were sent out under incorrect names. C. Wolley-Dod noted that Standish’s nursery (Ascot) were sending out F. japonica var. japonica under the name P. japonicum. The Grindon collection at Manchester Museum (MANCH) has an impressive collection of early Japanese Knotweed specimens, with interesting comments and press-cuttings attached. One such specimen implies that Smith’s of Worcester may have been sending out var. compacta labelled as P. alpinum. Another specimen cultivated at the gardens of the Royal Botanic Society London in 1867 is labelled P. sibericum. These names are not synonyms. A sheet with specimens of F. japonica var. japonica and F. sachalinensis (1878) (MANCH) has an 1894 letter attached from a seed company to Leo Grindon requesting “the specific name of the enclosed Polygonum”. Such errors are perhaps inevitable during a period of rapid expansion of the exotic taxa available to gardeners. Difficulties of identification of naturalised alien taxa are to be expected, as there is inevitably a time lag between naturalisation and the incorporation of adequate accounts in the local Floras. Even up to the first edition of Clapham et al. (1952), P. polystachyum and F. sachalinensis were not readily separable. THE WILD GARDEN MOVEMENT It was William Robinson (1838-1935), one of the most influential of the late Victorian gardeners, who advocated replacing the regimented “carpet-bedding” of High Victorian practice (cf. Carter 1984; Thacker 1994) by a more informal “lay-out” and a more natural setting enhanced by the “new” exotic species then becoming available . He promoted his ideas through articles in The Gardeners’ Chronicle and other journals, culminating with the publication in 1870 of the highly influential book The Wild Garden. This book was subtitled “Our Groves and Shrubberies made beautiful by the Naturalisation of hardy exotic plants”. He described Polygonum sieboldii enthusiastically, and recommended P. cuspidatum under “Plants with large or graceful foliage suitable for naturalization”. The enormous influence of Robinson’s writings led, for example, Surgeon-General Henry Cook, whilst in Bombay with the Indian Medical Service, to dream of making just such a garden of his own when he retired. The resultant Gloucestershire garden, where he grew many of the alien species of Polygonum, is described by Cook (1903) and is an important record of early plantings. In The English Flower Garden (1898), Robinson states that “the great Japan Knotworts (Polygonum) are handsome in rough places in the wild garden’, and that there is “no better plant for semi-wild places, or for association with vigorous things on the turf in the pleasure ground” than Polygonum sachalinense. The influence of Gertrude Jekyll (1843-1932) was just as far-ranging, abroad as well as in England, though less concerned with naturalisation and woodland. However, in A Gardener’s Testament (Jekyll, F. & Taylor 1982) she is quoted as envisaging a woodland-walk flanked by groups of “plants of rather large stature ... bamboos and the great Knotweeds of Japan”. In Home and Garden (1900) “we ought not forget the quick growing ways of the great Japan Knotweeds (Polygonum) growing fast and tall”. At least two of her garden plans include Fallopia japonica var. compacta: the wild garden at Little Aston, Birmingham, and Drayton Wood, Norfolk (Bisgrove 1992). Mrs C. W. Earle, a Surrey enthusiast of the “wild garden” (1897), is ecstatic in her praise of her “favourite Polygonum cuspidatum” - the “‘..handsomest, easiest-grown, hardiest, most useful plant for London gardens”. F' sachalinensis is also recommended, but said to lack the “..beautiful up-standing and yet graceful growth.” of F. japonica. THE ROLE OF PRIVATE GARDENS AND THE KNOTWEED CIRCLES Canon Ellacombe of Bitton (Gloucs.), Miss Gertrude Jekyll in west Surrey, E. A. Bowles and Miss Ellen Willmott in Essex, and the Rev. C. Wolley-Dod in Cheshire formed a “circle” of specialist gardeners who corresponded, visited each other and exchanged plants, ideas and expertise. They all had exceptionally rich gardens: Ellacombe’s one and a half acres contained a “unique gathering of out of the way plants” (see Hill 1919). Apart from British nursery gardens they could have also 104 J.P. BAILEY AND A. P. CONOLLY ordered from continental nursery gardens such as Von Siebold & Co. or Haage & Schmidt, both early suppliers of F. japonica (Table 2). At least some were also in regular receipt of plants from the major botanic gardens including those on the continent. There was regular two-way traffic in new plants between Kew Gardens and these wealthy amateurs - Ellacombe was recorded as supplying Kew with some 20 species for description in Curtis’s Botanical Magazine. He was also in correspondence with Kew directors from J. D. Hooker onwards: the 1881-95 Kew ‘Plants Outward” book recording 49 dispatches of plants to him. An 1882 specimen of F. japonica from Ellacombe’s garden is now in the Grindon collection (MANCH). Leo Grindon was a botanist and field naturalist interested in alien plants; he was an avid collector, and maintained a wide circle of correspondents. In addition to Ellacombe’s, specimens of Japanese Knotweed s./. in his collection come from Charles Bailey at Cleeve Hill (Cheltenham), Mr. Garnett of Penketh (near Warrington), John Wood of Kirkstall (Leeds), J. Cosmo Melvill at Prestwich, Mr Scott at Denzell (near Altrincham) and from Eastnor Castle (Herefordshire). There were also specimens from Manchester Botanic Garden and the Garden of the Royal Botanical Society, London (see Appendix 1). Exchanges within and between these two groups could well have included one or more of the Japanese knotweeds they were known to have grown. Further dissemination beyond these immediate circles could have followed later. CLUSTERS AND ASSOCIATIONS In consideration of the subsequent history of the escape and naturalisation of these Fallopia taxa (Conolly 1977; Bailey et al. 1996) we need to distinguish between F. japonica and the two taxa of greater stature, F.sachalinensis and F. x bohemica. Since F. japonica var. japonica is now of almost ubiquitous occurrence, it is in a sense too late to deduce anything from the absence or presence of this taxon in any particular area. The other two taxa on the other hand have a much less “mature” pattern of distribution, and are much more restricted in occurrence and _ habitat. There are also important differences in the history of their introduction - F. sachalinensis having been recommended as a forage plant. There is also evidence of its being used on large estates for scenic plantings and cover for shoots. It is certainly true that F. sachalinensis is found more often than F. japonica well off the beaten track on old estates, as if it had originally been deliberately planted. The case of the hybrid F. x bohemica is a little more complex; in addition to vegetative spread, it can also arise from seed in locations where F. japonica occurs with male-fertile F. sachalinensis. This is most likely to occur in nursery and botanic gardens and in private gardens extensive enough to accommodate such large plants. The association of F. sachalinensis and F. x bohemica with large estates and nursery gardens will be explored to see to what extent current stands of these two taxa can be related to such estates either as residuals or escapes. We also examine in more detail those areas where there are several adjacent spots on the B.R.C. 10-km recording scheme map, which we refer to as clusters. By far the largest cluster of F. x bohemica in the British Isles is in West Surrey (v.c. 17), where there are more than 35 stands, some dozen of which are associated with F. sachalinensis. They occur in the area bordered by Dorking, Guildford, Godalming and Haslemere - more specifically the “Surrey Hills”. This is the heart of Gertrude Jekyll country, and it is tempting to see a direct link with the fact that her house, ““Munstead Wood’, was in the centre of this area with its numerous large estates and expensive country houses. Many of these houses were designed by Lutyens, with Jekyll designing the gardens. Many of the present stands are close to such properties, and the abundance of stands of both taxa in and around Holmbury St. Mary, close to a group of Jekyll gardens, is surely significant. But proof of a direct recommendation by Jekyll and/ or of supply by, for example, Gauntlett’s at nearby Chiddingfold is more difficult to establish; and in any case there were many other nursery gardens in the vicinity. Gauntlett’s did, however, offer Polygonum sieboldii, P. sachalinense and the “mammoth” P. “Cookii” for many years. Moreover, stands, possibly relics, of F. x bohemica grow today a short distance downstream from the former Gauntlett site (Mrs J. Smith, pers.comm.). In another cluster of this kind, in West Cornwall, there are a number of associations between well known gardens and the occurrence of Fallopia taxa. Buryas Bridge, Penzance, where the unusual F. japonica var. japonica x F. japonica var. compacta hybrid grows (Bailey & Conolly 1991), is near the Trewidden estate (formerly the property of T. Bolitho) where over 450 species HISTORY OF JAPANESE KNOTWEED 105 of plants were grown at the end of the last century (Pett 1998). A former nursery garden is also nearby. F. x bohemica persists at Lanarth on the Lizard peninsula, where P. D.Williams (1865— 1935), R.H.S. medallist and one of the greatest Cornish gardeners, had his estate. His relatives had nursery gardens close to Redruth, where Gauntlett’s, prior to their 1906 move to Surrey, sold a range of Japanese Knotweed taxa. In South Wales, concentrated in Glamorgan , there is a further such cluster of F. x bohemica or F. sachalinensis sites associated with estates of “landed gentry”. Clearly derived from or residual are those at Black Pill near Clyne Castle (v.c. 41), former estate of the Vivians (Lord Swansea); Tredegar Park (v.c. 35, Lord Tredegar); Cyfarthfa, Merthyr Tydfil (Crawshays estate); Craig-y- nos, (once the home of Adeline Patti); Gabalfa House, Cardiff and Velindre, Whitchurch (now a mental hospital). In Scotland too, there are further such clusters of F. x bohemica sites, for example in the Glasgow area, mainly along the River Kelvin, possibly relating to the Botanic Garden. Further north, on the Morayshire coast (v.c. 95) around Forres is another Fallopia cluster. Here, in the gardens at Kellas House to the east of Forres, both F. sachalinensis and the hybrid occcur, and along the Muckleburn downstream of Dalvey House (west of Forres) F. sachalinensis is abundant. Further downstream, where the Muckleburn approaches the river Findhorn up to a 100 ha of F. x bohemica are found. The three taxa also occur at a number of other sites in the area. Kellas House and Dalvey House are probably responsible for much of these introductions, but Brodie Castle, home in the last century to the Brodie of Brodie - a noted horticulturist - is also a possibility (v.c. 95 information from I. Green, pers.comm.). In England and Wales a high proportion of those stands which occur as solitary (non-clustered) 10-km entries are also associated with local estates and gardens. These include F. sachalinensis at Edwinsford (Carms.); Falcondale (Cards.), by a lake; Nant-y-Frith, Flintshire (abandoned garden); Easney, Ware (Herts.), a pheasant shoot cover; and at Leigh near Tonbridge (Kent) - all of which are on or close to an estate. In the east of England F. sachalinensis has long out-lived the nursery garden of R. Wallace at Kilnfield Gardens, Colchester, that once cultivated it (fide J. Heath) (Table Zs). There are, of course, many other isolated stands of these taxa, in quite other habitats, not associated with estates but naturalised on roadsides, railway tracksides or banks, for which some other explanation is called for. THE FALL FROM GRACE The “hype” of the early Siebold introductions, coupled with its high price and gold medal status, leaves little doubt of the initial desirability of F. japonica to the Victorian gardener. It was undoubtably a spectacular plant, reliable and hardy even in the far north of the country; and of course there was a much more limited range of exotic garden plants available then. John Wood of Kirkstall (1884, p. 208) heaps unqualified praise on F. japonica var. japonica for its “..gracefully arching stems” and declares var. compactum to be a handsome bush. He unaccountably declares F. japonica var. japonica to be “a capital plant for the small town garden”’. He could, however, be accused of having a vested interest, since he was later to go into the nursery business (Table 2), selling these selfsame plants at 6 pence to 9 pence each! By the time of Robinson (1898) qualifications were beginning to creep in. Whereas F. sachalinensis and var. compacta are still both recommended, var. japonica, we are warned “..1s weedy, and in light soils springs up everywhere’. Similarly, Gertrude Jekyll (1899) cautions that “P. compactum must be reduced” and “P. sieboldii planted with caution”. Walters (1887), in his account of the flora of Alexandra Park in Oldham, noted that F. japonica var. japonica “..turns up unexpectedly in nearly every piece of cultivated ground”. Japanese Knotweed was not long content to remain in gardens and, from the beginning of the century, increasing numbers of escapes were reported (Conolly 1977). By the 1930s F. japonica was known in East Cornwall by the picturesque name of “Hancock’s Curse’, and its presence reputedly reduced the price of a house there by £100 (Conolly op. cit.). It is tempting to link this with Hancock’s nursery at Liskeard which is known to have been in business for at least the period 1897-1906 (Kelly 1897, 1902). In spite of its invasive proclivities now being well-known, Japanese Knotweed was still being offered in the mid-1930s by nursery catalogues such as Gauntlett’s of Chiddingfold (Table 2). 106 J.P. BAILEY AND A. P. CONOLLY Today F. japonica var. japonica poses very serious problems for river authorities, amenity areas and nature reserves alike, and a great deal of money is spent annually on surveying and control measures. F’. sachalinensis is not such a problem, but can nonetheless occupy considerable areas (e.g. Amroth v.c. 45); F x bohemica is a very invasive taxon, and its great stature makes its occurrence even more intrusive. These taxa are a particular problem in South Wales; the 1998 survey of 400 km? of the Swansea city and county found that F. japonica var. japonica occupied approximately 99 hectares, spread across a variety of habitats (S. Hathaway, pers. comm.). Its frequent escape in Glamorganshire as early as 1907 was noted by Conolly (1977). In 1981 the Wildlife and Countryside Act proscribed two land plants, Giant Hogweed and Japanese Knotweed, making it an offence to introduce these species to the wild. This has implications in terms of the redevelopment of sites infested with F. japonica since earth movement and tipping are major factors in spreading the plant further. Duty of care obligations can add considerably to the cost of site development, and contractors are becoming increasingly cautious of taking on infected sites because of the costs of excavation and deep tipping of infested soil. Hancock’s Curse is still with us! FUTURE PROSPECTS This paper leaves quite a few unresolved points and, although on first sight these might appear to be insoluble, there are modern techniques that may be able to shed some light on these historic matters. The main unresolved points are: e Has our F. sachalinensis come from Sakhalin or Japan or both and on how many occasions? e How many times has F. x bohemica arisen in Britain? e What are the relative roles of vegetative reproduction and seed production in the current distribution of F. x bohemica ? e Was P. “Cookii” really F. x bohemica? e Where in Japan did the atypical British clone of F. japonica var. japonica originate? The work of Hollingsworth & Bailey (2000) has indicated that it is possible to use DNA techniques both to identify particular clones and to extract information on the native region of origin of such plants. Work at Leicester University is currently underway on a number of these problems, helped by B.S.B.I. funding. POSTSCRIPT Looking back at the original importers of Japanese Knotweed s./., little did Siebold realize what was in store for the West, when he eventually got his precious Japanese plants back to Leiden, and thus in 1850 despatched that fateful package to Kew Gardens. The role of Kew in subsequent events may not be inconsiderable! We admire rather than deplore the efforts of the early explorers from Russia who first collected F. sachalinensis from Sakhalin and Japan: the naval surgeon Weyrich who “profited by every stay of the vessel to botanize” (Bretschneidér 1898, p. 618), and later Maximovicz travelling back from Japan with his enormous living collections. Their specimens were not collected primarily as novelties for gardeners, but for their taxonomic value. Nor should we blame the early English collectors of F. sachalinensis, such as Maries sent by Veitch, and Oldham and Wilford from Kew, sent to gather potential horticultural jewels, who, in spite of appalling conditions and risks, still managed to discover and bring back living new species. For they could not have guessed that their collections might include potential invaders. Indeed, this handsome Giant Knotweed has rarely if ever been a menace here. Nevertheless, with the introduction of male-fertile plants came the potential for crossing with the male-sterile F. japonica to produce our third taxon, the highly invasive F. x bohemica. A warning maybe of a menace to come! HISTORY OF JAPANESE KNOTWEED 107 At this point we recall the debt we owe to the efforts and encouragement of Duggie Kent and his generosity in sharing his superior Knotweed knowledge, without which we would not now be attempting to fit further pieces into the intriguing Japanese Knotweed jigsaw puzzle. ACKNOWLEDGMENTS The authors are indebted to David McClintock for generously sharing his years of Knotweed experience. We thank the staff at Herbaria at the Natural History Museum (BM), Manchester (MANCH), Kew (K) and Edinburgh (E) for assistance and loans. We are particularly grateful to the Royal Botanic Gardens at Kew and Edinburgh and the National Botanic Gardens, Glasnevin for access to their plant accession archives and Victorian nursery garden catalogue collections. We gratefully acknowledge the assitance of The R.H.S. Lindley library, the Natural History Museum botanical library, the Linnean Society’s library and in particular the Society’s librarian, Gina Douglas. We also thank our many botanical collaborators and correspondents, Ann Boucher, Arthur Chater, Lois Child, John Franks, Ian Green, W. Griffin, Shaun Hathaway, J. Heath, the late H. H. Heine, George Hutchinson, Alan Leslie, Rose Murphy, Kevin Pyne, Joyce Smith, Clive Stace and Roy Vickery. Finally, we acknowledge the assistance of the Herbaria at Uppsala (UPS), Geneva (G), Leiden (L) and Leningrad (LE) for loan of herbarium specimens. REFERENCES ANDERSON, A. W. (1951). Dr. F. P. von Siebold. Gardeners’ Chronicle Feb. 17 1951: 52. ANDRE, M. E. (1870). Un mois en Russie. Masson & Sons, Paris. ANDRE, M. E. (1893). La Sacaline. Revue Horticole 64: 326-327. ANON. (1943). Wisley in September. Journal of the Royal Horticultural Society 68:251-253. BAILEY, J. P. (1989). Cytology and breeding behaviour of giant alien Polygonum species. Ph.D. Thesis, University of Leicester. BAILEY, J. P. (1994). The reproductive biology and fertility of Fallopia japonica (Japanese Knotweed) and its hybrids in the British Isles, in WAAL., C. DE, CHILD, L. E., WADE, P. M. & BROCK, J. H., eds. Ecology and management of invasive riparian plants, pp. 141-158. John Wiley and Sons, Chichester. BAILEY, J. P., CHILD, L. E. & CONOLLY, A. P. (1996). A survey of the distribution of Fallopia x bohemica (Chrtek & Chrtkova) J. P.Bailey (Polygonaceae) in the British Isles. Watsonia 21:187—198. BAILEY, J. P. & CONOLLY, A. P. (1985). Chromosome numbers of some alien Reynoutria species in the British Isles. Watsonia 15: 270-271. BAILEY, J. P. & CONOLLY, A. P. (1991). Alien species of Polygonum and Reynoutria in Cornwall 1989-1990. Botanical Cornwall Newsletter 5: 33-46. BAILEY, J. P. & STACE, C. A. (1992). Chromosome number, morphology, pairing, and DNA values of species and hybrids in the genus Fallopia (Polygonaceae). Plant systematics and evolution 180: 29-52. BAILEY, L. H. (1949). Manual of cultivated plants. MacMillan, New York. BEERLING, D. J., BAILEY, J. P. & CONOLLY, A. P. (1994). Biological Flora of the British Isles No. 183 Fallopia japonica. Journal of ecology 82: 959-979. BISGROVE, R. (1992). The gardens of Gertrude Jeykyll. Frances Lincoln, London. BRETSCHNEIDER, E. (1898). History of European botanical discoveries in China. Sampson Low, Marston & Co., London. CARTER, T. (1984). The Victorian garden. Bell & Hyman Ltd., London. CHRISTY, T. (1893). Gardeners’ Chronicle. July 29, 1893: 132. CHRTEK, J. & CHRTKOVA, A. (1983). Reynoutria x bohemica, novy krinzinec z celedi rdesnovitych. Casopis narodniho muzea v Praze r. prir. 152: 120. CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. (1952). Flora of the British Isles. Cambridge University Press, Cambridge. CONOLLY, A. P. (1977). The distribution and history in the British Isles of some alien species of Polygonum and Reynoutria. Watsonia 11: 291-311. [CooK, H.]. (1903). A Gloucestershire wild garden, by “The Curator’. Elliot Stock, London. DAVIES, J. H. (1896). Notes on some Casuals in County Antrim. The Irish naturalist 5: 309-312. EARLE, C. W. (1897). Pot-pourri from a Surrey garden. Smith Elder & Co., London. GROENLAND, J. (1858). Polygonum cuspidatum. La Revue Horticole 7: 630-633. HILL, A. W., ed. (1919). Henry Nicholson Ellacombe 1822-1916. A Memoir. Country Life & George Newnes Ltd, London. HOLLINGSWORTH, M. L. (1998).Clonal growth and hybridisation in some invasive Fallopia spp. Ph.D. Thesis, University of Leicester. 108 J.P. BAILEY AND*A. P- CONOELY HOLLINGSWORTH, M. L. & BAILEY, J. P., (2000). Hybridisation and clonal diversity in some introduced Fallopia species (Polygonaceae). Watsonia 23: 111-121. HOOKER, J. D. (1880). Polygonum compactum. Curtis’s Botanical Magazine 106: 6476. HOOKER, J. D. (1881).Polygonum sachalinense. Curtis’s Botanical Magazine 107: 6540. HOUTTUYN, F. (1777). Natuurlijke Historie 8. De Erven van F. Houttuyn, Amsterdam. JACKSON, B. D. (1895). Index Kewensis. Clarendon Press, Oxford. JEKYLL, F. & TAYLOR, G. C., eds. (1982). A Gardener’s Testament. A selection of articles and notes by Gertrude Jekyll. Antique Collectors’ Club, Woodbridge, Suffolk. JEKYLL, G. (1899). Wood and garden. Longmans, Green & Co., London. JEKYLL, G. (1900). Home and garden. Longmans, Green & Co., London. KELLY’S (1897). Directory of Devonshire & Cornwall. KELLY’S (1902). Directory of Devonshire & Cornwall. KIMURA, Y. & KITAMURA, S. (1977). A guide to Flora Japonica of P. F. Siebold. Johnson Reprint Corp., New York. LINDLEY, J. & PAXTON, J. (1850). Gleanings and original memoranda. No. 185 Polygonum cuspidatum. Paxton’s Flower Garden. 1: 137-138. MAKINO, T. (1901). Polygonum reynoutria (Houtt.) Makino. Botanical magazine Tokyo 15: 84. MAXIMOWICZ, C. J. (1859). Primitiae florae amurensis. Mémoires presentés a L’Académie Impériale des Sciences de St. Pétersbourg 9: 233. MORREN, C. (1849). Plantes Nouvelles; Polygonum cuspidatum. Annales de Gand 5: 461-462. PETT, D. E. (1998). The Parks and gardens of Cornwall. Alison Hodge, Penzance. REGEL, E. (1864). Polygonum sachalinense F. Schmidt. Gartenflora 13: 68. REGEL, E. (1874). Neue oder empfehlenswerthe Zierpflanzen, 19) Polygonum sachalinense. Gartenflora 23: 87-88. ROBINSON, W. (1870). The wild garden. John Murray, London. ROBINSON, W. (1898). The English flower garden, 6th ed. John Murray, London. SADIE, S., ed. (1980). The new Grove dictionary of music and musicians, 16. Macmillan, London. SCHMIDT, F. (1868). Reisen im Amur-lande und auf der Insel Sachalin. Botanische Theil, Mémoires présentés a L’ Académie Impériale des Sciences de St. Pétersbourg 12 (2). SHARMAN, F. (1990). A pioneer in Japan. Garden, journal of the Royal Horticultural Society 115: 350-354. SHIOSAKA, H. & SHIBATA, O. (1993). Morphological changes in Polygonum cuspidatum Sieb. et Zucc. reciprocally transplanted among different altitudes. Japanese journal of ecology 43: 31-37. SIEBOLD, P. F. (1844). Kruidkundige naamlijst van oud en nieuw ingervoerde Japansche en Chineesche Planten. Jaarboek van de Koninklijke Nederlandsche Maatschappij tot Aanmoediging van den Tuinbouw, pp. 23-39. SIEBOLD, P. F. (1848). Extrait du catalogue et du prix-courant des plantes du Japon et des Indes-Orientales et Occidentales Neerlandaises. Jaarboek van de_ Koninklijke Nederlandsche Maatschappij tot Aanmoediging van den Tuinbouw, pp. 38-49. SIEBOLD, P. F. (1856). Catalogue Raisonné prix-courant des plantes et graines du Japon cultivées dans l’établissement de von Siebold & Company. Henry & Cohen, Leiden. SIEBOLD, P. F. & ZUCCARINI, J. G. (1846). Polygonum cuspidatum. Abhandlungen der Mathematisch-Physikalischen Classe der Koniglich Bayerischen Akademie der Wissenschaften 4(2): 208 (Fl. Jap. Fam. Nat., 2:84). STAFLEU, F. A. & COWAN, R. S. (1986). Taxonomic literature, 2nd ed., 6: Regnum Vegetabile 115. Utrecht/Antwerp. SYNGE, P. M., ed. (1956). The R.H.S. supplement to the dictionary of gardening. Oxford University Press, Oxford. THACKER, C. (1994). The genius of gardening. Weidenfield and Nelson, London. THUNBERG, C. P. (1784). Flora Japonica. J. G. Muller, Leipzig. VEITCH, J. H. (1906). Hortus Veitchii. James Veitch, Royal Exotic Nurseries, Chelsea. VEITCH, J. H. (1896). A traveller’s notes. James Veitch, Royal Exotic Nurseries, Chelsea. VRIESE, W. H. DE (1849). Polygonum cuspidatum. Jaarboek van de Koninklijke Nederlandsche Maatschappij tot Aanmoediging van den Tuinbouw, pp. 30-32 + plate. WALTERS, C. (“Sylvio”). (1887). The Park Flora. Journal of the Oldham Microscopical Society and Field Club 1 (4). WILLSON, E. J. (1989). Nurserymen to the World. E. J. Willson, London. WOOD, J. (1884). Hardy perennials & old-fashioned garden flowers. L. Upcott Gill, London. BIBLIOGRAPHY ALLAN, M. (1982). William Robinson 1838—1935. Faber & Faber, London. COATES, A. M. (1969). The quest for plants. Studio Vista, London. DESMOND, R. (1977). A dictionary of British and Irish botanists & horticulturists. Taylor & Francis, London. FLETCHER, H. R. (1969). The story of the Royal Horticultural Society 1804-1968. Oxford University Press, London. KELLY’S Directory of Cornwall (1889, 1906, 1914) KELLY’S Directory of Hampshire (1903, 1911) KELLY’S Directory of Surrey (1882, 1899, 1924, 1938) KELLY’S Directory of Gloucestershire (1885, 1894, 1897, 1914) WHITTLE, E. (1992). The historic gardens of Wales, C.A.D.W. H.M.S.O., London. (Accepted October 1999) HISTORY OF JAPANESE KNOTWEED 109 APPENDIX 1 HERBARIUM SPECIMENS CITED F. JAPONICA VAR. JAPONICA [Polygonum multiflorum «], Herb Thunberg 9705 UPS Polygonum cuspidatum S. & Z. Legit v.S, Herb. Lugd. Batav. No. 908.174 826 L . cuspidatum Sieb. Hort, Kew 1857, Herbarium Hookerianum K . cuspidatum coll. R. Oldham, Japan No. 235 10/1861. Note saying “..up to 8ft high usually one stemmed but sometimes branching near base”. Good seed set 2 sheets - male and female. K . cuspidatum coll. R. Oldham, Japan No.337 with seed K . cuspidatum coll. R. Oldham, Nagasaki Japan No 235 1862 female. K . cuspidatum coll. Maximowicz Yokohama 1862. (Lvs crinkled like compacta.) K . cuspidatum coll. Maximowicz iter Secundum, Yokahama 1862 (exserted stamens). cuspidatum var. humilis rubiflora coll. Maximowicz Yokohama 1862 “cult Yedo” labelled female K P. cuspidatum coll. Maximowicz Nagasaki 1863, male and female (- compacta-like) K [P. sibericum] cult. R.Bot. Soc. London Garden, 1867 (male-sterile) MANCH P. cuspidatum Mr. Ellacombe’s, Bitton, Oct. 9 1882 (abundant good fruits) MANCH pel ist as) ae) el ane) te} F. JAPONICA VAR. COMPACTA P. cuspidatum var. compactum, Mr. John Wood’s, Kirkstall, Leeds, July 16 1881 MANCH P. cuspidatum var. compactum, Mr. Scott’s Denzel, (Altrincham), Aug 31 1883 only 3ft high MANCH [P.alpinum] Herb Charles Bailey, Cult. Cleeve Hill 13 Oct 1911 ex Smith’s nursery MANCH [P.alpinum] Herb Charles Bailey, Cult. Cleeve Hill 30 July 1911 ex Smith’s nursery MANCH P. cuspidatum var. compactum, Melrose Abbey 1915, coll. ILM. Haywood det. Thellung. E P. cuspidatum var. compactum Bowles’ Garden Myddleton House, Bulls Cross. W.T. Stearn 3 September 1954. BM F. SACHALINENSIS Polygonum sachalinense F. Schmidt, Ins. Sachalin. Weyrich 1853 LE [P. cuspidatum] Hakodadi Japan, 1859 coll. C. Wilford. K P. sachalinense Sachalin Island, P. von Glehn 1861 K, BM P. sachalinense Hakodate1861, Maximowicz iter Secundem. Ex Herb horti bot Petropolitani K P. sachalinense coll. Maxim 2nd Japan, Hakodate Japan 1861 BM P. sachalinense F. Sch. “Siberia” Hort. [cf. K] 23/9/1874 8-9ft fine foliage MANCH [P. sachalinense] dated July 2 1878 Kew (F. japonica leaves on same sheet). 7 Sept. 1894 letter from Dickson Brown & Tait attached. MANCH P. sachalinense central mountains of Japan Aug1880, coll Mr. Maries and presented by Messrs Veitch. Labelled male plant Hort Kew K sachalinense Sent from Eastnor Castle, Herefordshire, Sept. 10 1886 MANCH sachalinense Mr. Garnett’s Penketh, (nr Warrington) Sept. 1892 MANCH sachalinense male-fertile specimen labelled from “Surgeon General ‘Cobbs’ (Cook’s) Garden Lydney” Sept 1899 K sachalinense Cult Kew Sept 1899. (In same hand as “Cobbs” (Cook’s) specimens, female) K sachalinense Brook House, Prestwich, (Lancs.) J. Cosmo Melvill cult in hort herbaceo ex Insula Saghalin, Aug. 1901 MANCH We eras F. x BOHEMICA [P. sieboldii] Manchester Botanic Garden, Two specimens on one sheet; Oct. 1872 & Sept. 23 1876 MANCH [P. sachalinense] Sent by Mr Cook, Lydney Gloucester 13.9.1901 K [P. cuspidatum] Charles Bailey, Haymesgarth, Cleeve Hill, near Cheltenham 22/9/1911. (Male- fertile) root from Smith’s nursery MANCH [P. sachalinense] Marston, Oxford Sept 1922 coll. Claridge Druce (male-sterile) MANCH 110 J.P. BAILEY AND A. P. CONOLLY FALLOPIA MULTIFLORA Polygonum multiflorum Herb Thunberg 9706 UPS [Polygonum (Helxine) Japonicum] Mihi. Vid Nat His. VHI Pl. 49 Fig 3, Herb Delessert collection Burman unnumbered G Watsonia 23: 111—121 (2000) tii Hybridisation and clonal diversity in some introduced Fallopia species (Polygonaceae) M. L. HOLLINGSWORTH* Royal Botanic Garden, Inverleith Row, Edinburgh, EH3 5LR and J. P. BAILEY Department of Biology, University of Leicester, Leicester, LE] 7RH ABSTRACT Molecular markers have been used to examine the amounts of genotypic diversity and sample inter- relationships among introduced invasive Fallopia taxa in Britain. The highly invasive Japanese Knotweed (Fallopia japonica var. japonica) appears to be represented by a single clone in Britain. In contrast, the related species Giant Knotweed (F. sachalinensis) and their hybrid (F. x bohemica) show much higher levels of clonal diversity. The difficulties of distinguishing between recurrent origins and hybrid fertility are discussed in relation to the variability of F. x bohemica. Finally, a very preliminary survey of native Asian F. japonica has detected high levels of genetic diversity. This, coupled with some limited morphological and cytological observations, suggests that a taxonomic revision of Fallopia section Reynoutria is needed to clarify the status of British F. japonica. INTRODUCTION Many plant communities contain a large proportion of introduced species. Human activities have vastly accelerated the movement of plants, carrying thousands of species, either deliberately or accidentally, across natural barriers such as oceans and mountain ranges. Many of these introduced species have become naturalised in the wild and some have become invasive weeds. In Britain, alien plants form a significant component of the flora, with a recent account reporting 1391 aliens out of the 2963 species recorded (Stace 1997). Hybridisation is a frequent and important component of plant evolution and is widespread amongst natural populations (Rieseberg & Ellstrand 1993). Occasionally, alien species hybridise with native or other introduced species. In the New Flora of the British Isles, Stace (1997) recognised 770 angiosperm hybrids; of these, 58 involve at least one non-native taxon and 12 are hybrids between two alien species. Hybridisation involving an introduced species, either with a native species or another introduced taxon, can give rise to hybrid combinations that would be unlikely to occur naturally due to a lack of sympatry of native ranges. The evolutionary implications of such post-introduction interactions are well documented and provide some of the clearest and most unequivocal examples of speciation such as the allopolyploid origins of Senecio cambrensis Rosser (Ashton & Abbott 1992; Harris & Ingram 1992) and Spartina anglica C. E. Hubb. (Marchant 1967; Raybould et al. 1991). Introduced plants make useful model systems for the study of evolutionary processes such as hybridisation and speciation. Long term historical factors can be eliminated as confounding variables if the introduction dates are known. We can study evolutionary events as they are happening (or at least very recently after they have happened) rather than attempting to make inferences about distant historical events. In addition, as some introduced plants show weedy tendencies, it is also informative to study the amounts and partitioning of their genetic variation, as this can provide information on the population genetic consequences of colonisation. “Corresponding author: e-mail: M.Hollingsworth @rbge.org.uk 2 M. L. HOLLINGSWORTH AND J. P. BAILEY INVASIVE FALLOPIA SPECIES IN THE BRITISH ISLES FALLOPIA JAPONICA VAR. JAPONICA - JAPANESE KNOTWEED Perhaps the most notorious invasive plant in the British flora is Japanese Knotweed, F. japonica var. japonica. This species is native to Japan, Taiwan and Northern China where it grows along roadsides, river banks and managed pasture, and it is one of the dominant colonists of lava fields (.C.O.L.E. 1997). It is insect-pollinated and visitors to its flowers include flies, honey bees and wasps (Tanaka 1966). F. japonica var. japonica, as recognised in Britain today, was introduced to the British Isles in the 1850s (Bailey & Conolly 2000). Since then it has spread rapidly. The first recorded escape from cultivation was in 1886 (Storrie 1886); by 1996 it had been recorded from 55% of the 2862 10-km squares of the Biological Records Centre mapping system of the British Isles (B.R.C. unpublished data). Its spread has been documented by Conolly (1977) and Bailey & Conolly (2000). In Britain F. japonica var. japonica occurs in a variety of man-made habitats such as canal sides, road verges, railway embankments and cemeteries, as well as stream and river banks. The abundance of this plant in urban areas and the costs associated with its control have given it much bad publicity. Its shoots are able to push up through asphalt and damage pavements, car parks and other public facilities and the height to which the plants grow leads to reduced visibility along roadsides and railways. Along water courses, decaying shoots can cause blockages and dense stands on riverbanks can impede flow in high water, thus increasing the risk of flooding. Since 1981 it has been a criminal offence to knowingly introduce F. japonica var. japonica into the wild. FALLOPIA SACHALINENSIS - GIANT KNOTWEED A closely related species to F. japonica var. japonica is F. sachalinensis (Giant Knotweed). Fallopia sachalinensis occurs in Britain, again stemming from a horticultural introduction (Conolly 1977; Bailey & Conolly 2000). It is native to eastern Asia where it is known from the southern part of Sakhalin Island, the southern Kurile Islands, the Japanese islands of Hokkaido and Honshu, and Korea (Sukopp & Starfinger 1995). It occurs in disturbed habitats from sea level to c. 1050 m and can be found along roadsides, forest edges, cliff tops and river sides, and like F. Japonica var. japonica, it 1s one of the pioneer colonists of the “lava deserts” caused by volcanic eruptions (Sukopp & Starfinger 1995). Fallopia sachalinensis shows less invasive tendencies than F. japonica var. japonica and is still a rather uncommon plant in much of the British Isles (Bailey 1997). The first account of this species growing outside cultivation was published in 1896 (Davies 1896). OTHER INTRODUCED FALLOPIA SPECIES In addition to F. sachalinensis and F. japonica var. japonica, other members of the genus which have been introduced to the British Isles are F. japonica var. compacta and F. baldschuanica. Fallopia japonica var. compacta is a dwarf variety of F. japonica which is only rarely found outside of cultivation and is known from only 30 10-km squares of the Biological Records Centre mapping system (B.R.C. unpublished data). Fallopia baldschuanica (Russian-vine) was introduced from central Asia and is a common garden plant. It frequently occurs as discarded material on waste ground, but is rarely well naturalised (Stace 1997). Although reference will be made to these two taxa in this paper, we primarily concentrate on F. sachalinensis and F. japonica var. Japonica. REPRODUCTIVE BIOLOGY OF FALLOPIA JAPONICA AND F. SACHALINENSIS Both F. japonica and F. sachalinensis are gynodioecious and can occur as either male-sterile (female) or hermaphrodite plants. In Britain, both female and hermaphrodite forms of F. sachalinensis and F. japonica var. compacta have been recorded, but only female plants of F. Japonica var. japonica are known (Bailey 1994). Fallopia sachalinensis and F. japonica var. compacta thus have the capacity to reproduce both sexually and asexually (via vegetative propagation). However, despite the remarkable spread and abundance of F. japonica var. japonica, as no male-fertile plants are known from Britain, it appears that seed dispersal is not responsible for its success (no evidence for apomictic seed production has been recorded). Rhizome fragments HYBRIDISATION AND CLONAL DIVERSITY IN FALLOPIA 113 as small as 0-7 g are capable of regenerating into new plants (Brock & Wade 1992) and the transport and tipping of earth containing plant fragments and water-borne dispersal along river systems are considered to be the major means of dispersal (Conolly 1977). HYBRIDISATION Morphological and cytological studies have provided convincing evidence that hybrids between F. japonica s. 1. and F. sachalinensis occur in the British Isles (Bailey & Conolly 1985). The hybrid between F. japonica and F. sachalinensis - F. x bohemica (Chrtek & Chrtkova) J. Bailey - was first described in 1983 in the Czech Republic (Chrtek & Chrtkova 1983). Two taxonomic crosses have been detected in Britain: plants with 2n = 66 stemming from a cross between F. japonica var. Japonica (2n = 88) and F. sachalinensis (2n = 44); and plants with 2n = 44 stemming from a cross between F. japonica var. compacta (2n = 44) and F. sachalinensis. Very rarely, hybrids have also been reported between F. japonica var. japonica and F. japonica var. compacta; these plants have 2n = 66 (Bailey & Conolly 1991). Interestingly, the majority of seed on British F. japonica var. Japonica appears to have F. baldschuanica as the pollen donor and these progeny are 2n = 54 (Bailey 1988). However, only one such hybrid is known to have become established in the wild in Britain (Bailey 1988). In contrast the hybrid between F. japonica var. japonica and F. sachalinensis is relatively common in Britain (Bailey et al. 1996). This hybrid shows some of the invasive tendencies of F. japonica var. japonica and occurs as both female and hermaphrodite plants. Although morphological and cytological data have provided much useful information about this complex in Britain, many questions regarding the biology of these plants require additional data. For instance, while there is very good evidence that hybrids between some Fallopia taxa are present in the wild, we have no information on how many times such hybrids may have arisen, and indeed to what extent hybrid fertility allows for generations beyond the FI. In addition, little is known about the amounts and partitioning of genetic diversity in the parental taxa. The lack of male fertile plants of F. japonica var. japonica in Britain suggests that this taxon may have an extremely narrow genetic base and, at an extreme, could consist of a single widespread clone. In F. sachalinensis, which is represented in Britain as both hermaphrodite and female plants, the possibility for both sexual and asexual reproduction exists, although the relative importance of either is unknown. In this paper, we summarise our published and unpublished data which have attempted to shed light on some of these issues. Particular attention is paid to the relative importance of sexual versus asexual reproduction, and the dynamics of inter-specific hybridisation, concentrating primarily on F. japonica var. japonica, F. sachalinensis and their hexaploid (2n = 66) hybrid F. x bohemica. LEVELS OF CLONAL DIVERSITY IN THE PARENTAL TAXA FALLOPIA JAPONICA VAR. JAPONICA To assess the levels of clonal diversity of F. japonica var. japonica in the British Isles, we collected 150 samples from localities ranging from Land’s End to Shetland (Hollingsworth & Bailey, submitted). These samples were assayed for genetic variability using the arbitrary DNA fingerprinting technique RAPDs: Randomly Amplified Polymorphic DNA (Williams ef al. 1990). Using ten primers, we amplified a total of 108 reproducible bands. No genetic variation was detected between any of the 150 samples collected from across the British Isles (Hollingsworth & Bailey, submitted). These data, coupled with the absence of male fertile individuals in the British Isles, provide strong evidence to support the hypothesis that the entire British population of F. Japonica var. japonica consists of a single extraordinarily vigorous clone (Hollingsworth & Bailey, submitted). Furthermore, preliminary studies on a few samples from Europe and North America also showed the same genotype (genetic individual), which suggests that the distribution of this genotype extends well beyond the British Isles (Hollingsworth & Bailey, submitted). It seems possible that the enormous social and economic costs associated with the Japanese Knotweed invasion can be attributed to a single, ill-advised introduction. 114 M. L. HOLLINGSWORTH AND J. P. BAILEY FALLOPIA SACHALINENSIS To provide a comparison with the data from F. japonica var. japonica, we carried out a more limited study on the levels of genotypic diversity in F. sachalinensis. A total of 30 plants were collected from twelve localities. Single samples were collected from most populations, but 17 plants were collected from the River Kelvin, Glasgow (Lanark, v.c. 77), and three samples were collected from the River Clywedog, Caerynwch (Merioneth, v.c. 48). All samples were analysed for genetic variation using 10 RAPD primers. In contrast to the data from F. japonica var. Japonica, high levels of clonal diversity were detected. A total of 14 genotypes were recovered, and no genotype was shared between sites (Hollingsworth 1998). Two genotypes were detected from the 17 plants collected from the River Kelvin, with one of these genotypes restricted to a single plant (Hollingsworth et al. 1998). Two genotypes were detected from the three plants sampled along the River Clywedog. In the absence of additional well-sampled populations such as the River Kelvin it is difficult to say much on the partitioning of genetic variability in this species in Britain. However, in contrast to the simple situation of a single widespread clone of F. japonica var. japonica, F. sachalinensis appears to be genetically variable in Britain. Bailey & Conolly (2000) suggest that multiple introductions of F. sachalinensis to Britain may be the source of the greater levels of genetic variability of F. sachalinensis compared to F. japonica var. japonica. However, while M.L.H. agrees that this may be an important factor, she believes that there are presently insufficient data to eliminate sexual reproduction as a source of at least some of the variability. Fallopia sachalinensis has plants producing both functional ovules and pollen, and sets viable seed. The potential for sexual reproduction in this taxon in Britain is clear, and the correlation of this with genetic variability may well be meaningful. Progeny segregating from only a single heterozygous individual can lead to a multitude of genetic combinations. LEVELS OF GENETIC VARIABILITY IN 6x F. x BOHEMICA Genetic diversity can occur among hybrid individuals due to multiple origins (defined here as multiple gamete fusions) or hybrid fertility. Although meiosis is irregular in 6x F. x bohemica (J. P. Bailey, unpublished) some seed set does occur, and the possibility of F2 (and beyond) hybrid plants exists. To investigate the amounts and partitioning of genetic variability in 6x F. x bohemica we have investigated two populations in detail, the River Kelvin, Glasgow (23 plants), and the River Wnion system, Dolgellau (Merioneth, v.c. 48) (36 plants). In addition we also collected a total of 16 samples from 14 other sites across Britain. The River Kelvin site in Glasgow is of interest as, despite the fact that both F. japonica var. Japonica and F. sachalinensis grow close to the hybrid, only female plants of the parental taxa are present. In contrast, F. x bohemica occurs as both hermaphrodite and female plants along the river (Hart et al. 1997). Their co-occurrence may be due to independent colonisation, rather than being indicative of in situ hybridisation. This situation provides an opportunity to examine levels of sexual reproduction within the hybrid population, with reduced probabilities of multiple origins as a confounding variable. Using arbitrary fingerprinting techniques (RAPDs and Inter-SSRs (inter simple sequence repeats)), five genotypes were detected from 23 F. x bohemica plants (Hollingsworth et al. 1998). Of the four female plants sampled, all had the same genotype. The 18 hermaphrodite plants sampled had one of three different genotypes, with a final genotype restricted to a plant which did not flower (and thus could not be sexed). Given that we detected five different genotypes, and there are no male fertile individuals of the parental taxa at this site, it is possible that some of this genetic variability is due to hybrid fertility, rather than recurrent origins of the hybrid (Hollingsworth et al. 1998). Clearly, one cannot rule out the possibility that male fertile individuals of the parental taxa previously occurred in the area (although a thorough search of surrounding localities and checks of herbarium specimens found no evidence that this is the case (Hart et al. 1997)), or that the different hybrid genotypes are independent colonists of the river. However, given that the highest levels of genetic variation were detected in the only male fertile taxon present along the river (see the above sections on F. japonica var. japonica and F. sachalinensis), perhaps the most parsimonious explanation of the data is that at least some of the genetic variation in F’. x bohemica is attributable to local sexual reproduction. PS HYBRIDISATION AND CLONAL DIVERSITY IN FALLOPIA a ae pass oy} ee UOTU AA, TOA y SUOTe BX) mdol?, af a ues pene a jo eee [ aanoly yomuAseeg "2H yomufueed > yyed yoo4— DopamA|yD JOAIY eoiuodel "4 eolwayog x ‘4 zx sisuauleyoes’ 4 eoiuodel "4 eoIWayog X "4 gx eoluayog x *+} €x BoIwUayog x "4 sisuduljeyoes’ 4 oX BOIWaYod xX “4 ex eolwayog x *4 HopamA|D JeAIY | UCU] JeyeEesH USYM O}IS WO1) pa}]oe]]09 Sejdwies jo Jequunu =ux eojuodel ‘sen eoiuodel eidoye4 = eoiuodel ‘+ Koy ne|jabjoqg mes \ | punosb eolwayog x *4 uyed joo4 —_ voneere a | @ eoiuodel "4 ) edIWaYO x “4 ; oe \ eoiuodel *4 ; wayog x *4 eolwayog x “4 eolWayog Xx "4 eoluayog x "4 eaiwayoqX "4" pSiuodef-4 ealweyog x °4 we nejjabjog UOIUAA JAAIY a 116 M. L. HOLLINGSWORTH AND J. P. BAILEY The second site we have examined in detail is a more complicated population along the River Wnion system (Fig. 1). Fallopia japonica var. japonica is common at many sites along this river system but there are only two known localities for F. x bohemica and one known locality for F. sachalinensis. At Caerynwch Hall on the River Clywedog (Fig. 1) female F. japonica var. japonica, hermaphrodite F. sachalinensis and both female and hermaphrodite F. x bohemica grow together along the edge of the river. The production of Fallopia seedlings has been recorded at this site (Bailey et al. 1996). The second area where F. x bohemica is found is 5 km downstream at Dolgellau (Fig. 1). Both female F. japonica var. japonica and female and hermaphrodite hybrids occur here, growing on the banks of the River Wnion and also in the surrounding field and recreational area; F. sachalinensis is not present at this site. Two samples of F. japonica var. Japonica, three samples of F. sachalinensis and 20 samples of F. x bohemica, including six seedlings, were collected from Caerynwch Hall. Three samples of F. japonica var. japonica and 16 samples of F. x bohemica were collected from Dolgellau. All samples were examined for genetic variation using ten RAPD primers. The five samples of F. japonica var. japonica all gave the standard genotype for this taxon in Britain, and are assumed to be ramets of the widespread genet. Two genotypes of F. sachalinensis were detected from the three samples from this site (see above). A total of 14 genotypes were detected from the 30 mature plants of F. x bohemica, with a further six genotypes detected from the six seedlings (Fig. 2a & b). The genetic structure of F. x bohemica along the river system suggests that both asexual and sexual reproduction have played a role in the history of this population. At some locations the same genotype was found in 2—5 ramets, with these samples being either adjacent to each other (i.e. genotype B5 Fig. 2b), or in two cases widely separated and found at both Caerynwch and Dolgellau (i.e. genotype B2 & B3; Figs 2a & 2b). This suggests the occurrence of both local (within site) and widespread (between site) dispersal via vegetative propagation. All six seedlings we examined had unique genotypes; only a single RAPD primer was required to discriminate between these samples, highlighting the sensitivity of the assay. The presence of female F. japonica var. japonica with hermaphrodite F. sachalinensis at Caerynwch Hall, coupled with the high number of genotypes of F. x bohemica, suggest multiple hybridisation events as the source of this variability. However, other factors may be responsible. The presence of a large number of F. x bohemica genotypes found only at Dolgellau, where F. sachalinensis does not occur, implies the need for caution before multiple hybridisation events are invoked as the causal explanation for the genetic variability of F. x bohemica in this river system. The private genotypes of F. x bohemica at Dolgellau could be explained by water-borne dispersal of seeds from Caerynwch Hall, or dispersal (vegetative or seed) from plants (unknown to us) along parts of the river system not sampled in this study. However, if multiple independent introductions to this site from outside of the river system (which seems unlikely given the large number of genotypes detected) are excluded, there is one other explanation for this variability: hybrid fertility. Distinguishing between hybrid fertility and multiple origins as a source of genetic variability in F. x bohemica is difficult. Some insights may be gained from examining the chromosome numbers of the individuals in question. Although only a small sample, of the seven plants of F. x bohemica from the River Wnion system from which cytological data are available, all are 2n = 66. Bailey (1997) suggested that the irregular meiosis of F. x bohemica would tend to lead to subsequent generations being aneuploids. Cytological analysis of seeds collected from isolated plants of F. x bohemica have shown a diverse array of chromosome numbers (Bailey 1990; Bailey et al. 1996). Thus one may suggest that if hybrid fertility was the source of the genetic variability in the F. x bohemica population from the River Wnion system, then many of these plants would be expected to be aneuploids. However, while the presence of aneuploids would presumably be good evidence that the plants were the result of an irregular meiosis, it should be stressed that the converse is not conclusive. Irregular meiosis could still give rise to 2n = 66 individuals, and this number is presumably at least as likely as any other possible chromosome number from 2n = 66 plants. As we have no information on the survival characteristics of aneuploid versus euploid F. x bohemica it is premature to place too strong an interpretation on the sole recovery of 2n = 66 plants. It is possible that there is a disproportionate survival of euploid plants over aneuploids (but see J.P.B. opinion below). From our limited survey of F. x bohemica from more widespread British localities (16 plants from 14 sites), no genotypes were shared between localities (Hollingsworth 1998). The only shared genotype was from Swansea, where, from our sample of three plants, two, growing 20 m apart, gave identical RAPD profiles. HYBRIDISATION AND CLONAL DIVERSITY IN FALLOPIA 117 B2&B11 B 15, B 16, B 17, B 18,B 19 & B 20 River Clywedog 0.1km Caerynwch FIGURE 2a. Distribution of Fallopia genotypes along the River Clywedog, Caerynwch. J Fallopia japonica var. japonica; S n Fallopia sachalinensis, n refers to genotype code; B n Fallopia x bohemica, n refers to genotype code ( = a 7 Recreation ground ! N Dolgellau 0.1km g : — FIGURE 2b. Distribution of Fallopia genotypes along the River Wnion, Dolgellau. J Fallopia japonica var. japonica; B n Fallopia x bohemica, n refers to genotype code 118 M. L. HOLLINGSWORTH AND J. P. BAILEY Although further work is required (with more intensive within-site sampling), 6x F. x bohemica is a genetically variable taxon, and we conclude that sexual reproduction appears (in one way or another) to have been an important determinant of population structure. There remains, however, some question as to precisely how this has occurred, i.e. what is the relative importance of hybrid fertility versus multiple origins as the source of the genetic variability? This is one area where we as authors disagree. M.L.H. argues that the “jury is still out” and that the current data set (Hollingsworth, 1998; Hollingsworth et al. 1998) is inadequate to distinguish between the two phenomena. She does not believe the possibility of hybrid fertility giving rise to new genotypes can yet be excluded. In contrast, J.P.B. believes that the lack of aneuploids among those plants whose chromosome numbers have been counted is sufficient evidence against hybrid fertility contributing to the observed genotypic diversity. Unlike many biosystematic problems, this issue may be easily tractable. Fallopia japonica var. japonica appears to be represented in the British Isles by a single clone. Our RAPD and inter-SSR data have shown that there are numerous markers that distinguish this clone from our samples of F. sachalinensis. To distinguish between hybrid fertility and multiple origins as determinants of the genetic variability in F. x bohemica the following simple experiments need to be carried out. Controlled crosses should be made between F. japonica var. japonica and F. sachalinensis. The parents and large numbers of progeny should be examined for their RAPD and inter-SSR profiles (polyploid segregation tests require extensive progeny arrays). Although RAPDs and inter-SSRs are dominant markers and it is not possible to distinguish between homozygotes and heterozygotes for any given gene in individual samples, examination of the progeny of controlled crosses will allow homozygotes to be identified. Bands that are homozygous in F. japonica var. japonica should be present in all of its Fl hybrids. In contrast, those bands that are heterozygous in F. japonica var. japonica and segregate, will be present in some, but not all progeny. Following this line of reasoning, any wild-collected 6x F. x bohemica which lacks bands that are homozygous in F. japonica var. japonica (present in all of its controlled Fl progeny) may be considered to be of more complicated origin than Fl hybrids. As a final point, however, it is important to note that there is one other potential confounding variable which is not so easily tested for, and this is whether the variability was generated in situ at all. It is not known to what extent material of F. x bohemica was originally distributed as seed for horticultural purposes (see Bailey & Conolly 2000). Although we feel that the evidence presented here for in situ sexual reproduction within Britain is virtually conclusive (as there are so many different genotypes present), it remains possible that some of the variability in this taxon in Britain could stem from horticulturally raised and distributed seed. JAPANESE KNOTWEED IN ASIA Although the results presented here provide insights into the reproductive biology and population structure of these invasive Fallopia taxa in Britain, there is a conspicuous absence of reference to native material. A detailed knowledge of the genotypic and genetic variability of these plants in their native range would be desirable for comparative purposes and may also serve to identify the geographic source of the introduced material. This is of value as F. japonica var. japonica is represented by only a single genotype in Britain and it may be particularly susceptible to an appropriate biological control. Identification of the source population may be a useful step in the search for such controls. However, not only is there a need for molecular work on the native material, there is also a need for more fundamental alpha taxonomic research. Whilst the morphological and cytological variation in British Fallopia taxa has received considerable attention (Bailey 1994; Bailey et al. 1996; Bailey et al. 1995; Bailey & Conolly, 1985; Bailey & Stace, 1992), the variability in the native range has only begun to be appreciated. Plants of Fallopia japonica collected from Japan and China show a high degree of intraspecific variability, in both cytological and morphological characters (Bailey 1999). Material of F. japonica var. japonica found in the British Isles today, although morphologically more similar to Japanese than Chinese material, is not typical of either (Bailey 1999). Japanese F. japonica are predominantly tetraploids (2n = 44), whilst the three Chinese and one Korean plant that have been counted so far are octoploid or decaploid (2n = 88 and 2n = 110). With this in mind, in the following section we refer to native material as F. japonica var. ‘japonica’ to distinguish it from British F. japonica var. Japonica. HYBRIDISATION AND CLONAL DIVERSITY IN FALLOPIA 119 F. japonica var. compacta British Isles F. japonica var. ‘japonica’ Peking F. japonica var. japonica F. japonica var. ‘japonica’ Europe & USA China F. japonica var. ‘japonica’ Tanagawa River, Tokyo, Japan F. japonica var. ‘japonica’ Mt. Fujija,Japan F. japonica var. ‘japonica’ Honshu, Japan F. japonica var. ‘japonica’ Shinjuka, Tokyo, Japan : } - ee F. japonica var. ‘japonica Japan F. japonica var. ‘japonica’ Mt. Tateyama Japan FIGURE 3. Neighbour Joining tree of Fallopia japonica s.1. We carried out RAPD analysis using 10 primers on 14 native samples of F. japonica var. ‘japonica’ and compared them to the British genotype of F. japonica var. japonica and three British plants of F. japonica var. compacta (Table 1) (Hollingsworth 1998). The interrelationships of the samples are shown in Fig. 3. Although most of the genetic distance between samples in this data set is due to differences among individuals rather than groups of individuals (i.e. most of the genetic distance in the figure is on the terminal, rather than internal branches) a few points stand out. The three F. japonica var. compacta samples cluster together, and they together form a group with the introduced genotype of F. japonica var. japonica. The two Chinese samples cluster together, and there is a cluster containing the Japanese samples (although separated by only a very short branch from the other samples). A key point is that there is considerable diversity among the native samples, with much greater genetic distance between Japanese and Chinese samples of F. Japonica var. ‘japonica’ than there is between the two different named varieties that occur in Britain This point is further emphasised by our studies on the chloroplast genome of these taxa. Using either sequence data from the chloroplast trnL intron or PCR RFLPs of the trnK intron, we are able to distinguish the chloroplast genomes of British material of F. japonica var. japonica, F. japonica var. compacta and F. sachalinensis (Hollingsworth et al. 1999). When we compared the trnL_ sequences of British F. japonica var. japonica and F. japonica var. compacta to five accessions of native F. japonica var. ‘japonica’ we did not detect either of the British chloroplast types among these native plants (Hollingsworth 1998). Instead, from the five native samples we detected five additional chloroplast types (all different), although relationships among these were poorly resolved in cladistic analyses, as most differences were autapomorphic (unique to individual samples). However, although there is little taxonomic information in these sequence data, the results corroborate those of morphology, cytology and RAPDs, i.e. that there is considerable diversity among native plants named as F. japonica var. japonica, and this exceeds the diversity between the British introduced varieties F. japonica var. japonica and var. compacta. Furthermore, no material which precisely matches the introduced and highly invasive F. japonica var. japonica is known to us from the native range. 120 M. L. HOLLINGSWORTH AND J. P. BAILEY TABLE 1 COLLECTION DETAILS OF FALLOPIA TAXA Taxon Locality Chromosome No. F. japonica var. ‘japonica’ Peking Botanic Gardens, China 2n = 88 F. japonica var. ‘japonica’ China (no further details known) 2n = 116 F. japonica var. ‘japonica’ Honshu, Japan 2n = 44 F. japonica var. ‘japonica’ Japan (no further details known) F. japonica var. ‘japonica’ Mt. Fujiya, Japan 2n = 44 F. japonica var. ‘japonica’ Mt. Tateyama, Midagahara, Japan 2n = 44 F. japonica var. ‘japonica’ Mt. Tateyama, Midagahara, Japan F. japonica var. ‘japonica’ Mt. Tateyama, Midagahara, Japan F. japonica var. ‘japonica’ Shinjuku, Tokyo, Japan F.. japonica var. ‘japonica’ Shinjuku, Tokyo, Japan F. japonica var. ‘japonica’ Tanagawa River, Tokyo, Japan F. japonica var. ‘japonica’ Tanagawa River, Tokyo, Japan 2n = 44 F. japonica var. japonica Leicester (v.c. 55), England 2 =56 F. japonica var. compacta Brecon (v.c. 42), Wales 2n = 44 F. japonica var. compacta Bracken Hill, Surrey (v.c. 16), England 2n = 44 F. japonica vat. compacta Blarney Castle, Co. Cork (v.c. H4), Ireland 2n = 44 Thus there are different interpretations of what constitutes Japanese Knotweed (F. japonica var. Japonica) in native (Asian) and introduced (Europe and North American) regions. To resolve and clarify these differences, a detailed monograph of section Reynoutria (F. japonica s.l. and F. sachalinensis s.1.) including morphological, cytological and molecular data is required. Widespread collections should to be made from across the native range and the samples compared using both phenotypic and genotypic data. Only when an understanding of material in both introduced and native ranges is acquired will the most appropriate taxonomic designation for these samples become apparent. One safe prediction, however, 1s that there will undoubtedly be a more focused (narrow) consideration of what should be included under the name of F. japonica var. Japonica. ACKNOWLEDGMENTS We are grateful to Pete Hollingsworth for his helpful comments on earlier drafts of this manuscript. We are also grateful to Ann Conolly, Colin Ferris, Richard Gornall and Clive Stace for many useful discussions regarding our Fallopia research, and to John Akeroyd and Daniel Kelly for constructive review. REFERENCES ASHTON, P. A. & ABBOTT, R. J. (1992). Multiple origins and genetic diversity in the newly arisen allopolyploid species, Senecio cambrensis Rosser (Compositae). Heredity 68: 25-32. BAILEY, J. P. (1988). Putative Reynoutria japonica Houtt. x Fallopia baldschuanica (Regel) Holub hybrids discovered in Britain. Watsonia 17: 163-164. BAILEY, J. P. (1990). Breeding behaviour and seed production in alien Giant Knotweeds in the British Isles, in ANON., ed. The biology of invasive plants; a BES industrial ecology group symposium, pp. 110-130. Richards, Moorehead and Laing, Ruthin. BAILEY, J. P. (1994). Reproductive biology and fertility of Fallopia japonica (Japanese Knotweed) and its hybrids in the British Isles., in DE WAAL, L. C., CHILD, L. E., WADE, P. M. & BROCK, J. H., eds. Ecology and management of invasive riverside plants, pp. 141—158. John Wiley & Sons, Chichester. BAILEY, J. P. (1999). The Japanese Knotweed invasion of Europe, the potential for further evolution in non- native regions, in YANO, E., MATSUO, K., SYITYOMI, M. & ANDOwW, D. A., eds. Biological invasions of ecosystems by pests and beneficial organisms, pp. 27-37. Tsukuba: Institute of Agro-Environmental Sciences, Tsukuba, Japan. BAILEY, J. P., CHILD, L. E. & CONOLLY, A. P. (1996). A survey of the distribution of Fallopia x bohemica (Chrtek & Chrkova) J. Bailey. Watsonia 21: 187-198. HYBRIDISATION AND CLONAL DIVERSITY IN FALLOPIA pA BAILEY, J. P., CHILD, L. E. & WADE, M. (1995). Assessment of the genetic variation and spread of British populations of Fallopia japonica and its hybrid Fallopia x bohemica, in PYSEK, P., PRACH, K., REJMANEK, M. & WADE, M.., eds. Plant invasions - general aspects and special problems, pp. 141-150. S.P.B. Academic Publishing, Amsterdam. BAILEY, J. P. & CONOLLY, A. P. (1985). Chromosome numbers of some alien Reynoutria species in the British Isles. Watsonia 15: 270-271. BAILEY, J. P. & CONOLLY, A. P. (2000). Prize-winners to pariahs - A history of Japanese Knotweed s./. (Polygonaceae) in the British Isles. Watsonia 23: 93-110. BAILEY, J. P. & STACE, C. A. (1992). Chromosome number, morphology, pairing and DNA values of species and hybrids in the genus Fallopia (Polygonaceae). Plant systematics and evolution 180: 29-52. Brock, J. H. & WADE, M. (1992). Regeneration of Japanese Knotweed (Fallopia japonica) from rhizomes and stems: observations from greenhouse trials, in ANON., ed. The 9th International Symposium on the biology of weeds, pp. 85—94. (unpublished conference proceedings), CHRTEK, J. & CHRTKOVA, A. (1983). Reynoutria x bohemica, novy krinzinec z celedi rdesnovitych. Casopis ndrodniho musea. Oddil Prirodovedny 152: 120. CONOLLY, A. P. (1977). The distribution and history in the British Isles of some alien species of Polygonum and Reynoutria. Watsonia 11: 291-311. DAVIES, J. H. (1896). Notes on some casuals in Co. Antrim. /rish naturalist 5: 309-312. HARRIS, S. A. & INGRAM, R. (1992). Molecular systematics of the genus Senecio L. 1: Hybridisation in a British polyploid complex. Heredity 69: 1-10. HART, M. L., BAILEY, J. P., HOLLINGSWORTH, P. M. & WATSON, K. J. (1997). Sterile species and fertile hybrids of Japanese Knotweeds along the River Kelvin. Glasgow naturalist 23: 18-22. HOLLINGSWORTH, M. L. (1998). Clonal growth and hybridisation in some invasive Fallopia spp. Ph.D. thesis, University of Leicester. HOLLINGSWORTH, M. L. & BAILEY, J. P. (submitted). Evidence for massive clonal growth in the invasive weed Fallopia japonia (Japanese Knotweed). HOLLINGSWORTH, M. L., BAILEY, J. P., HOLLINGSWORTH, P. M. & FERRIS, C. (1999). Hybridisation and chloroplast DNA variation amongst some invasive Fallopia (Polygonaceae) species. Botanical journal of the Linnean Society 129: 139-154. HOLLINGSWORTH, M. L., HOLLINGSWORTH, P. M., JENKINS, G. I., BAILEY, J. P. & FERRIS, C. (1998). The use of molecular markers to study patterns of genotypic diversity in some invasive alien Fallopia spp. (Polygonaceae). Molecular ecology 17: 1681-1692. 1.C.0.L.E. (1997). All you ever wanted to know about Japanese Knotweed but were afraid to ask. Japanese Knotweed - a strategy for effective control. Loughborough University (unpublished workshop report). MARCHANT, C. J. (1967). Evolution in Spartina (Gramineae). 1. History and morphology of the genus in Britain. Botanical journal of the Linnean Society 60: 1-24. RAYBOULD, A. F., GRAY, A. J., LAWRENCE, M. J. & MARSHALL, D. F. (1991). The evolution of Spartina anglica C. E. Hubbard (Gramineae): origin and genetic variability. Biological journal of the Linnean Society 43: 111-126. RIESEBERG, L. H. & ELLSTRAND, N. C. (1993). What can molecular and morphological markers tell us about plant hybridisation? Critical reviews in plant sciences 12: 213-241. STACE, C. A. (1997). New Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. STORRIE, J. (1886). The Flora of Cardiff. Cardiff. SuKoppP, H. & STARFINGER, U. (1995). Reynoutria sachalinensis in Europe and in the far east: A comparison of the species ecology in its native and adventive distribution range., in PYSEK, P., PRACH, K., REJAMANEK, M. & WADE, M., eds. Plant invasions - general aspects and special problems, pp. 151-159. S.P.B. Academic Publishing, Amsterdam. TANAKA, H. (1966). The insect visitors of Polygonum cuspidatum Sieb. et Zucc. Collecting and breeding 28: 141-143. WILLIAMS, J. G. K., KUBELIK, A. R., LIVAK, K. J., RAFALSKI, J. A. & TINGEY, S. V. (1990). DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic acids research 18: 6531-6535. (Accepted October 1999) 7 “ e ’ a r. iT o - f . 7 u mn i 1 i i ; ‘= ; ? i ig ” 7 : i * os ‘ : re tal Q 2 i ’ eg) we il Watsonia 23: 123-138 (2000) 123 Hybrid origin of the Oxford Ragwort, Senecio squalidus L. R. J. ABBOTT, J. K. JAMES, J. A. IRWIN* and H. P. COMES’ Institute of Environmental & Evolutionary Biology, School of Biology, University of St Andrews, St Andrews, Fife, KY16 9TH ABSTRACT A survey of morphometric and allozyme variation was conducted to test the hypothesis that Oxford ragwort, Senecio squalidus L. (Asteraceae) is a diploid hybrid species which originated in cultivation from hybrid material between S. aethnensis Jan. ex DC. and S. chrysanthemifolius Poiret that was introduced to the Oxford Botanic Garden from Mt Etna, Sicily, in the 17th century. Morphometric analysis of material raised under the same glasshouse conditions showed that S. squalidus closely resembled material derived from hybrid swarms between S. aethnensis and S. chrysanthemifolius on Mt Etna, and produced a phenotype that was normally intermediate in leaf shape and capitulum size to that of its two putative parent species. Allozyme analysis further showed that populations of S. squalidus were very similar genetically to Sicilian material with some populations clustering tightly with hybrid material and with S. chrysanthemifolius. Most notably, S. squalidus and hybrid material were polymorphic and exhibited almost identical frequencies for two alleles (a and b) that distinguish pure populations of the two parent species at the Acp-2 locus. Thus, in combination with previous documentary evidence, the morphometric and allozyme results obtained from the present study provide strong support for the hypothesis that S. squalidus is a new diploid hybrid species which originated allopatrically through stabilization of hybrid material of S. aethnensis and S. chrysanthemifolius introduced to the British Isles from the hybrid zone on Mt Etna, Sicily. Because allozyme diversity within S. sgualidus was found to be much reduced to that present within Sicilian material, it is likely that only a small sample of hybrid material (and allelic diversity) was introduced to the British Isles from Sicily. KEYWORDS: hybrid speciation, morphometrics, allozymes, introduced plants, Asteraceae. INTRODUCTION In a series of articles published between 1955 and 1966, Douglas Kent chronicled the introduction, escape from cultivation, and subsequent spread of the Oxford ragwort, Senecio squalidus L. (Asteraceae), in the British Isles (Kent, 1955, 1956, 1957, 1960, 1963, 1964a,b,c, 1966). The information contained in these reports is of considerable value for two reasons. First, there are very few examples of an introduced invasive species for which such details of introduction and spread have been rigorously documented. Second, a detailed knowledge of the spread of the Oxford ragwort in the British Isles has been of crucial importance to the timing of two notable evolutionary events that followed hybridization of the species with the native Groundsel, Senecio vulgaris L., viz. the origin of the new allopolyploid species, S. cambrensis Rosser, independently in Wales and Scotland, during the 20th century (Ashton & Abbott 1992a; Lowe & Abbott 1996), and the origin of the radiate form of S. vulgaris (var. hibernicus Syme), a stabilised introgressant, during the 19th century (Abbott 1992; Abbott et al. 1992a). Although Kent (1956) provided information on the source of $. squalidus material introduced to the British Isles, no details were available on how this material might have become modified during cultivation before it escaped and became invasive. In this paper, we provide evidence that the material originally introduced to the British Isles was hybrid in nature and that S. squalidus, as recognised in these islands, arose as a derivative of this material most probably during the period of cultivation. Present address: *John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH TInstitut fiir Spezielle Botanik und Botanischer Garten, Johannes Gutenberg-Universitét Mainz, D-55099 Mainz, Germany 124 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES According to Kent (1956) and Druce (1927), S. squalidus (2n = 20) was collected from Mt Etna, Sicily, and introduced to the Oxford Botanic Garden, England, soon after the Garden opened in 1621. The exact date of its introduction is unknown as early records from the Garden were lost in a fire. Present records show that it has been in cultivation at Oxford since at least 1690 (Kent 1956). Linnaeus described the species in 1753, from plants raised by him possibly from seed sent from Oxford (Walker 1833), although evidence for the latter is disputed (Kent 1956). In 1792, the species was reported growing on walls in the Oxford area (Druce 1927) and was described as a garden escape. The subsequent spread of the species away from Oxford was aided initially by exchange of seeds between botanic gardens but, in the late 19th century, following the establishment of the railway system in the British Isles, S$. squalidus began spreading rapidly from Oxford via the railway network. More recently, the species has spread along motorway verges in Britain, and is now ubiquitous on wasteground especially in urban areas. Records show that the species began to spread northwards from Oxford in the late 19th century and reached different parts of the north of England between the early to mid-part of the 20th century (Kent 1964b.c). The species became established in the wild in the Central Belt of Scotland in the mid-1950s (Kent 1955, 1966), and continues to spread northwards into Fife and north of the River Tay at the present time (Abbott 1992, and pers. obs.). The source and nature of S. sgualidus material now established in the British Isles has been questioned in recent times from two standpoints. First, Walters (1963) emphasized the morphological similarity between S. squalidus in the British Isles and S. rupestris Waldst. and Kit. (2n = 20), a closely related species that occurs on mountains in central and southern Europe. This raised the possibility that S$. rupestris was an alternative source of some, if not all, material of S. squalidus established in Britain. Because of this morphological similarity, Alexander (1979) in a taxonomic revision of Mediterranean Senecio, section Senecio, subsumed S. rupestris under S. squalidus, although he considered British plants to be atypical of the taxon. Second, Crisp (1972), in a biosystematic analysis of Senecio material growing on Mt Etna, was unable to recognise the taxon S. squalidus found in Britain, and argued instead that British S. sqgualidus is a product of hybridization between the diploid Sicilian species S$. aethnensis Jan. ex DC. (2n = 20) and S. chrysanthemifolius Poiret (2n = 20), and that there is no evidence of material from elsewhere in Europe having contributed to the British taxon. According to Crisp (1972), both Sicilian species occur on Mt Etna with S. chrysanthemifolius present up to approximately 1000m altitude and S. aethnensis from approximately 1600 to 2600 m altitude. Between these two species at 1300 m +/- 300 m, a series of hybrid swarms are present within which are found certain plants that bear a close morphological resemblance to British S. sgualidus. Crisp (1972) advanced the view that material introduced to the British Isles most likely was collected from these hybrid populations, and following a century of cultivation in the Oxford Botanic Garden, stabilized derivatives of this hybrid material escaped at the end of the 18th Century and have since spread rapidly to many parts of the British Isles. He concluded that “British S. sqgualidus is of hybrid origin...” and “...can be treated as a separate species as it is both geographically isolated from the parent species, and it has evolved .... over many generations to a stage where it is morphologically distinct from either of them, although still polymorphic and in general intermediate between them.” For the purposes of this paper, and in keeping with the views of Crisp (1972), we treat as separate species S. squalidus found in the British Isles and S. rupestris from central and southern Europe . A survey of chloroplast (cp) DNA variation in S. sgualidus, S. aethnensis, S. chrysanthemifolius and S. rupestris (Abbott et al. 1995) showed that all material of S. sqgualidus examined from the British Isles (21 accessions from 20 populations) possessed the same cpDNA haplotype (haplotype B) as found in material of S. chrysanthemifolius surveyed (11 accessions from seven populations in Sicily). Haplotype B also occurred in three of five individuals of S. aethnensis examined, and five of seven individuals surveyed from the hybrid swarms on Mt Etna. Other individuals from these populations possessed haplotype A, which differed by a single restriction site mutation from haplotype B. In contrast, only one accession surveyed of S. rupestris (from Abruzzi, central Italy) possessed haplotype B, while other accessions of this species contained either haplotype A (from the Italian Alps, Romania, N Bulgaria, and introductions to Germany), or haplotype C (from S$ Bulgaria and Greece). Haplotype C differed from haplotype A by a length mutation and three restriction site mutations. The occurrence of haplotype B in all samples of S. squalidus from the British Isles and in the two Sicilian species, S. aethnensis and S. chrysanthemifolius, and in their HYBRID ORIGIN OF SENECIO SQUALIDUS Pes) hybrid swarms, supported the hypothesis that S. squalidus is derived from hybrid material introduced from Sicily. However, because haplotype B is also present in S. rupestris, albeit at low frequency, it can not be ruled out that the British population of S. squalidus might be derived either in part or entirely from introductions of S. rupestris. Here we report the results of a comparison of morphometric and allozyme variation within material of S$. squalidus from the British Isles with that within Senecio material sampled from Mt Etna, Sicily, namely S. aethnensis, S. chrysanthemifolius, and hybrid populations between these two species. These results provide further support for Crisp’s (1972) hypothesis of a hybrid origin of S. squalidus in the British Isles. MATERIALS AND METHODS PLANT MATERIAL Seed was collected separately from approximately 30 plants from each of the populations listed in Table 1 and Fig. 1. This entailed sampling three populations of each of S. aethnensis, S. chrysanthemifolius, and three of their hybrid swarms occurring on Mt Etna, Sicily, plus nine populations of S. squalidus from the British Isles, and a single population of S. vernalis Waldst. and Kit. from Germany. This material was analysed for allozyme variation, while subsamples were used in a survey of morphometric variation. MORPHOMETRIC ANALYSIS Five plants (one offspring per mother plant) of each of S. aethnensis, S. chrysanthemifolius, and also of hybrid material from Mt Etna (two plants from population 7, and three plants from population 8), were raised with five plants from each of two populations of S. squalidus (from Oxford and Edinburgh) (Table 1). The 25 plants were grown from seed to maturity as single individuals in pots of 13 cm diameter containing compost. Pots were fully randomized within a greenhouse. Illumination was provided by natural daylight supplemented by 400-W mercury vapour lights to give a photoperiod of 16 h, while temperature was maintained at 20 °C +/- 2 °C. On the day of full anthesis of the apical capitulum, each plant was harvested and measured for 32 characters. The character set recorded (Table 2) was modified from that used by Irwin and Abbott (1992). Fifteen of the characters were descriptors of the capitulum and another 15 described leaf size and shape. The remaining two characters were plant height and inflorescence length. Details of measurement are given in James (1995). Data were subjected to univariate one-way ANOVA and Duncan’s multiple range tests using the software package SPSS Base (version 7.5, SPSS Inc. FIGURE |. (a) Locations of populations of Senecio squalidus ( @ ) sampled in the British Isles; (b) Locations of populations of S. aethnensis ( & ), S. chrysanthemifolius ( ® ), and S. aethnensis x S. chrysanthemifolius (™ ) sampled from Mt Etna, Sicily. Key to population numbers is given in Table 1. 126 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES TABLE 1. GEOGRAPHIC LOCATIONS, SITE DESCRIPTIONS, COORDINATES, AND ALTITUDE OF SAMPLED POPULATIONS OF SENECIO Species No. S. aethnensis 1 3% S. chrysanthemifolius — 4* S. aet. x S. chrys. Te 8* S. squalidus 14 15 16 17 18 19 S. vernalis Locality Italy, Sicily, Mt Etna, 2 km E Rifugio Sapienza, lava by roadside Italy, Sicily, Mt Etna, 3 km E Rif. Sapienza, volcanic ash Italy, Sicily, Mt Etna, Rif. Citelli, volcanic ash Italy, Sicily, Mt Etna, Bronte, waste ground Italy, Sicily, Mt Etna, 5 km W of Nicolosi, roadside Italy, Sicily, Mt Etna, 4 km SE of Linguaglossa, roadside Italy, Sicily, Mt Etna, 4 km NNE of Ragalna Est, new lava Italy, Sicily, Mt Etna, Paso Cannelli, N of Pedara, iava Italy, Sicily, Mt Etna, Villagio Turistico Mareneve, roadside British Isles, Scotland, Kirkaldy, car park British Isles, England, Oxford, waste ground, Osney street British Isles, Scotland, Edinburgh, Leith, Mill Lane, waste ground British Isles, Ireland, Co. Cork, S.of Kilbrittain British Isles, England, Norwich, waste ground at railway station British Isles, England, Bristol, wasteground, Coronation Rd British Isles, England, Weymouth, waste ground British Isles, Wales, Cardiff, waste ground, Colington Rd British Isles, Scotland, Kirriemuir, waste ground Germany, Schiisselacker Weide, Eppelheim (nr. Heidelberg) *Populations included in morphometric analysis. Collectors: RJA (R. J. Abbott), PAA (P. A. Ashton), JAI (J. A. Lat. Long. Alt. (m) 37-42N 15:02E 1890 3741 N 15-03} Ee 1650 37-47N 15-04E 1750 37-45N 14:50E 760 37-:35N 14:59E 800 37-50 N’ 15:072E 750 37-37 N - 14:59.E) 19s 3740N 15-04E 1500 37-49N 15:04E 1400 56:08 N 03-10 W <50 51-45 N 01-09 W <50 55:58 N 03-10 W <50 51:37 N 08-35 W <50 52:-40N 01-20E <50 51:30 N 02:37 W <50 50:37 N 02-29 W <50 51-30N 03-10 W <50 56-43N 03:00W 0-5; Table 3). Thus, $. chrysanthemifolius possessed small capitula and highly dissected leaves in contrast to S. aethnensis which produced large capitula and almost entire leaves (Fig. 2, Table 2). Senecio squalidus plants with capitula of intermediate size and less HYBRID ORIGIN OF SENECIO SQUALIDUS 129 S. squalidus (11) S. squalidus (12) S. aethnensis S. chrysanthemifolius a + Pe @ O S. aet. x S. chrys. PC1 FIGURE 2. Plots for individuals of S. aethnensis, S. chrysanthemifolius, S. aethnensis x S. chrysanthemifolius (populations 7 and 8), and S. squalidus (populations 11 and 12) on the first two principal components axes. dissected leaves than S. chrysanthemifolius were also clearly separated from both Sicilian taxa along PC2, but remained clustered with certain hybrid plants. PC2 was largely influenced by leaf size and shape characters (Table 3). ALLOZYME ANALYSIS Three of the eight loci investigated (G3pd-/, Pgi-! and Pgm-1) were monomorphic in all 19 populations surveyed. The total number of alleles detected over the five polymorphic loci was 14 (range 2—3 per locus) (Appendix |). Of particular interest was the pattern of variation at the Acp-2 locus. All three populations of S. chrysanthemifolius were monomorphic for the Acp-2b allele (Appendix 1). In contrast, all of the §$. aethnensis and hybrid (S. aethnensis x S. chrysanthemifolius) populations investigated were polymorphic for the Acp-2a and Acp-2b alleles as were five of the nine populations of S$. squalidus. The remaining four populations of S. squalidus were monomorphic for the Acp-2b allele (Appendix 1). When data were pooled over populations (Table 4) it was evident that the frequencies of the two Acp-2 alleles in S. squalidus were very similar to those in the pooled hybrid material (S. aethnensis x S. chrysanthemifolius), with the Acp-2b allele occurring at a higher frequency (approximately 0-8) than in S. aethnensis material (approximatey 0-35). At other polymorphic loci there were no major differences in allele frequencies between the two Sicilian taxa, nor between them and their hybrid populations (Table 4). However, there was a contrast between Sicilian material and S$. squalidus in that the latter was monomorphic at the B-—Est-3 locus, and contained the Aat-3c allele at a much higher frequency. UPGMA analysis of Nei’s genetic distances between populations based on allele frequencies at all eight loci separated populations into two main clusters with S. vernalis positioned as a basal group (Fig. 3). One cluster comprised seven populations of S. squalidus, while the other contained the remaining populations. Within this second cluster, two subgroups were evident, one of which contained three populations of S$. aethnensis and one of the hybrid populations (population 9), while the other consisted of the three populations of S. chrysanthemifolius, two hybrid populations, and two S. squalidus populations. Levels of diversity varied across populations (Tables 5). Lower levels of diversity were present in S. squalidus populations relative to populations of other taxa examined, based on measures of A (mean number of alleles per locus), P (percentage of polymorphic loci) and H, (unbiased expected 130 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES TABLE 3. TOTAL - SAMPLE STANDARDIZED COEFFICIENTS (LOADINGS) AND EIGENVALUES FOR THE FIRST THREE PRINCIPAL COMPONENTS OF INDIVIDUAL VALUES OF QUANTITATIVE TRAITS Principal Components 1 Z 3 Height 0-131 0-241 0-172 Inflorescence length - 0-630 - 0-233 0-423 Capitulum length - 0-793 - 0-008 0-226 Pedicel length - 0-579 - 0-241 0-416 No. of pedicel bracts - 0-784 - 0-357 - 0-064 Capitulum apical width - 0-612 - 0-370 0-415 Capitulum basal width - 0-286 - 0-277 0-432 No. of phyllary bracts - 0-337 - 0-320 0-267 No. of calyculus bracts - 0-590 - 0-163 - 0-070 Calyculus bract length - 0-031 0-170 0-331 No. of florets per capitulum - 0-606 - 0-468 0-349 Ray floret length - 0-636 - 0-338 - 0-376 Ray floret width - 0-675 - 0-403 - 0-375 No. of disc florets - 0-629 - 0-448 0-344 Disc floret length - 0-693 0-290 0-207 Disc floret corolla length - 0-649 0-272 - 0-184 Disc floret corolla width - 0-317 0-202 - 0-410 Longest leaf length - 0-296 - 0-743 - 0-057 Midleaf length - 0-143 - 0-788 - 0-141 Midleaf lobe number 0-709 - 0-607 0-020 Midleaf maximum left width 0-663 - 0-643 0-018 Midleaf maximum right width 0-683 - 0-610 0-115 Midleaf basal length (left side) - 0-366 - 0-389 - 0-169 Midleaf basal length (right side) - 0-457 - 0-414 - 0-124 Midleaf apical angle A - 0-750 - 0-022 0-103 Midleaf apical angle B 0-020 - 0-778 0-104 Midleaf secondary vein angle 0-863 - 0-056 0-322 Square root of midleaf area - 0-494 - 0-575 - 0-398 Midleaf perimeter 0-582 - 0-759 - 0-226 Square root midleaf area/length - 0-557 0-075 - 0-397 Midleaf perimeter/length 0-727 - 0-573 - 0-177 Midleaf dissection 0-796 - 0-559 0-029 Eigenvalue 10-757 6-379 2-386 Variance (%) 33-616 19-935 7-455 Cumulative variance (%) 33-616 937501 61-006 heterozygosity). More diversity was present in S. aethnensis than in S. chrysanthemifolius and their hybrid populations. Only in very few populations was the coefficient of inbreeding (f) high and deviations from Hardy-Weinberg equilibrium significant. Thus in most populations random mating was indicated. The mean level of subdivision between S. squalidus populations (@ = F,,) over all loci (Table 6) was significantly different from zero (as determined by bootstrap estimates of 95% CIs), indicating significant population differentiation within this species in the British Isles. However, examination of the UPGMA tree (Fig. 3) did not indicate any obvious geographical pattern to this structure, and Mantel’s test of the correlation of @ and geographical distance between populations was not significant (r = 0-106, P = 0-280). Although the mean coefficient of inbreeding within S. squalidus populations (f = Fis) was not significant, a high f value was calculated at the Acp-2 locus (Table 6). This might stem from the HYBRID ORIGIN OF SENECIO SQUALIDUS 13:1 Ol. aet 09.axc S. aethnensis Sicily 02. aet S. aethnensis x S. chrysanthemifolius 03. aet 04. chr 4 OR axc Sicily S. chrysanthemifolius 05. chr S. aethnensis x 06. chr S. chrysanthemifolius 08.axc S. squalidus Sicily nets British Isles 17. sq 11. sq 12. sq 14. sq 15. sq British Isles S. squalidus 10. sq 13 e2sq 16. sq 19. ver Germany S. vernalis .30 .20 5/0) .0O Genetic Distance FIGURE 3. UPGMA tree constructed from genetic distances (after Nei, 1972) between Senecio populations. Key to population numbers is given in Table 1. 132 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES TABLE 4. ALLELE FREQUENCIES AT POLYMORPHIC LOCI IN SENECIO TAXA AFTER POOLING DATA OVER POPULATIONS. (N = SAMPLE SIZE) Taxon Locus/allele S. aethnensis S. chrysanth. S. aet. x S. chrys. S. squalidus Acp-2 N (62) (95) (76) (209) a 0-645 ---- 0-204 0-170 b 0-355 1-000 0-796 0-830 Aat-3 N (62) (94) (76) (214) b 0-911 0-925 0-960 0-327 c 0-089 0-075 0-040 0-673 B-Est-3 N (52) (87) (82) (215) a 0-673 0-632 O-390 Pe ie re === b 0-327 0-368 0-409 1-000 Pgi-2 N (61) (88) (84) (211) a OS 0 b 0-893 1-000 1-000 1-000 Pgm-2 N (59) (85) (81) (237) Bn 0023 0 2 eee wee b 0-881 0-906 0-920 0-996 C 0-119 0-071 0-080 0-004 occurrence of substructure within populations. However, previous studies (Abbott, unpublished) have detected a null allele segregating at low frequency at the Acp-2 locus in some S. squalidus populations. In the present study, null homozygotes were not recorded; however, it can not be ruled out that a null allele was present at low frequency in some of these populations and affected the correct identification of genotypes. The presence of a null allele for this dimeric enzyme would lead to inflation of the number of homozygotes recorded due to masking of the null allele in heterozygotes and, in turn, would result in inflation of the coefficient of inbreeding (/). DISCUSSION HYBRID ORIGIN OF SENECIO SQUALIDUS The hypothesis that Senecio squalidus originated from hybrid material between S. aethnensis and S. chrysanthemifolius introduced from Mt Etna, Sicily, to the British Isles in the 17th century (Crisp 1972) is supported by the results of the surveys of morphometric and allozyme variation reported here. Morphometric analysis showed that nine of ten S. squalidus plants raised under glass bore a close morphological resemblance to hybrid plants derived from material sampled from the hybrid zone on Mt Etna. Both types of plants produced phenotypes intermediate between S. aethnensis and S. chrysanthemifolius. Thus, whereas S. aethnensis produced large capitula and entire leaves, while S. chrysanthemofolius had smaller capitula and highly dissected leaves, S. squalidus and the hybrid plants tended to produce capitula of intermediate size and leaves of an intermediate degree of dissection. The only exceptions to this concerned one S. squalidus individual and one hybrid plant which exhibited a phenotype similar to that of S. chrysanthemifolius. The survey of allozyme variation across populations confirmed a close similarity of certain populations of S. sgualidus (populations 17 and 18) to S. aethnensis x S. chrysanthemifolius hybrid material (populations 7 and 8), and also to S. chrysanthemifolius. These materials were placed in a separate sub-cluster from S. aethnensis and one of the hybrid populations (9), in a UPGMA tree constructed using Nei’s genetic distances between populations (Fig. 3). Other populations of S. squalidus (populations 11-16) were positioned in a different cluster to Sicilian material, although based on the genetic distances recorded they were closely related to this HYBRID ORIGIN OF SENECIO SQUALIDUS 133 TABLE 5. ALLOZYME VARIABILITY ESTIMATES AND WEIR & COCKERHAM’S (1984) COEFFICIENT OF INBREEDING (F = FIS) INSAMPLED POPULATIONS OF SENECIO. TESTS OF HARDY-WEINBERG EQUILIBRIUM ACROSS LOCI WITHIN SAMPLES (FISHER’S METHOD) ARE INDICATED. Mean Species Popn. sample size+ At P Hoy Baik) S. aethnensis | 12-4 (1-2) 1-3 (0-2) 25:0 0-129 (0-084) -0-071 23-9 (0-1) 1-6 (0-2) 62:5 0-184 (0-068) 0-058 3 23-8 (0-6) 1-6 (0-2) 62:5. 0-208 (0-067) 0-209 Mean 1-5 50-0 = 0-174 S. chrysanthemifolius 4 41-0 (1-1) 1-5 (0-3) 37:5 0-107 (0-061) -0-071 >) 23°75 (1-1) 1-5 (0-3) 37:5 0-094 (0-065) 0-054 6 24-8 (0-2) 1-5 (0-3) 37:5 0-094 (0-062) 0-030 Mean 1-5 37:5 0-098 S. aet. x S. chrys. 7 37-8 (1-3) 1-5 (0-2) 50-0 0-076 (0-060) -0-319 8 27-6 (0-2) 1-5 (0-2) 50-0 0-138 (0-059) -0-159 9 15-6 (0-2) 1-5 (0-2) 50-0 0-138 (0-077) 0-365* Mean 1-5 50:0 0-117 S. squalidus 10 28-4 (0°6) 1-3 (0-2) 25:0 0-022 (0-015) -0-031 11 24-8 (1-3) 1-3 (0-2) 25-0 0-091 (0-059) 0-045 12 26-4 (1-7) 1-1 (0-1) 12-5. 0-005 (0-005) 0-000 13 22-4 (2-1) 1-1 (0-1) 12-5. 0-049 (0-049) -0-139 14 25-4 (0-5) 1-3 (0-2) 25-0 0-113 (0-075) 0-467** ifs) 25-5; Gl-5) 1-3 (0-2) 25-0 0-081 (0-060) 0-208 16 27-4 (0-3) 1-3 (0-2) 25-0 0-085 (0-063) 0-364* 17 18-8 (0-6) lel (0-1) 12-5 0-019 (0-019) -0-059 18 25-0 (0-7) 1-3 (0-2) 25-0 0-102 (0-069) 0-217 Mean 1-2 20:8 0-063 S. vernalis oy), 27-0 (0-0) 1-3 (0-2) 25-0 0-103 (0-068) 0-241 7 Standard errors in parentheses ++ Hardy-Weinberg equilibrium rejected at * P < 0-05 and **P < 0-01 A: mean no. of alleles per Jocus P: percentage of polymorphic loci H,.: mean expected heterozygosity material. Strong evidence that S. squalidus is of hybrid origin, rather than a derivative of S. chrysanthemifolius, was the presence of both Acp-2a and Acp-2b alleles in S. squalidus. Our results showed that S. chrysanthemifolius is monomorphic for the Acp-2b allele and consequently another taxon must have donated the Acp-2a allele to S. squalidus, which we propose is S. aethnensis. It could be argued, of course, that S. squalidus might be a derivative of S$. aethnensis rather than of hybrid origin, as all the allozyme alleles present in S. squalidus, including both alleles at the Acp-2 locus, were also found in populations of S. aethnensis (Table 4). However, there is good reason to believe that material of S$. aethnensis we sampled from Mt Etna was introgressed. One of us (James 1999) has since found that populations of S. aethnensis that grow at much higher altitudes on the mountain (>2,500 m) are monomorphic for the Acp-2a allele. We propose, therefore, that S. sgualidus is derived from hybrids between S. aethnensis and S. chrysanthemifolius rather than being derived directly from either parent. A hybrid origin of S. squalidus assumes that the hybrid zone on Mt Etna formed following secondary contact between S. aethnensis and S. chrysanthemifolius, and that material regarded as hybrid on Mt Etna is not ancestral to S. aethnensis and S. chrysanthemifolius. If the latter were the case, then S. squalidus would also be considered as directly derived from this ancestral taxon. 134 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES TABLE 6. WEIR & COCKERHAWM’S (1984) ESTIMATES OF WRIGHT’S F-STATISTICS CALCULATED SEPARATELY FOR EACH LOCUS FOR ALL BRITISH POPULATIONS OF SENECIO SQUALIDUS Locus F @ ei Acp-2 0-659 (0-087) 0-239 (0-091) 0-546 (0-073) Aat-3 0-443 (0-160) 0-426 (0-154) 0-028 (0-083) Pgm-2 0-000 (0-000) 0-030 (0-014) -0-031 (0-015) Mean (Std. Dev.) 0-478 (0-140) 0-385 (0-122) 0-219 (0-332) 95% confidence interval 0-000 - 0-655 0-016 - 0-420 -0-016 - 0-546 Means and standard deviations were obtained by jackknifing over loci. Confidence intervals were obtained by bootstrapping over loci. Locus specific standard deviations were obtained by jackknifing over populations. F = Fit (within total), @ = Fst (among populations), f = Fis (within populations) Although this possibility can not be excluded, it is considered to be highly unlikely. In addition to the fixation of different alleles at the Acp-2 locus in S. aethnensis and S. chrysanthemifolius, James (1999) has found that populations of S. aethnensis and S. chrysanthemifolius from the extremes of the altitudinal range on Mt Etna are monomorphic or almost monomorphic for many different RAPD bands and also for cpDNA haplotype. Such nuclear and cytoplasmic differentiation between these two species indicates that they are likely to have originated in isolation and recently come into contact. Despite these molecular differences, $. aethnensis and S. chrysanthemifolius produce fertile Fl and F2 hybrid populations, and are interfertile with S. squalidus (Abbott & Forbes unpublished). In another study (Abbott & Milne 1995) the Acp-2b allele, which occurs at high frequency in S. squalidus and in hybrid material on Mt Etna, was shown to be absent from 16 populations of S. rupestris sampled widely from the geographical distribution of this species. This shows that S. squalidus could not be derived entirely from S$. rupestris material, although does not rule out the possibility that it might be derived in part from this source. However, we propose that the allozyme evidence combined with the absence of any known documentary evidence of an introduction of S. rupestris to the British Isles, provide a strong argument against the possibility of S. rupestris having contributed to the British taxon. GENETIC STRUCTURE OF SENECIO SQUALIDUS The level of allozyme diversity within S. sgualidus, measured in terms of expected heterozygosity, was approximately one third of that recorded in S. aethnensis and approximately half that in the hybrid material from which we propose S. sgualidus is derived (Table 5). This reduced level of genetic diversity in S$. squalidus is probably due to a founder effect caused by the introduction to the British Isles of a small sample of hybrid material (and allelic diversity) from the hybrid zone on Mt Etna. Rieseberg (1997) has reviewed amounts of allozyme variation found in other homoploid hybrid species relative to their parents, and shown them to differ between species. In three Helianthus diploid hybrid species, levels of diversity were lower than expected, indicating a small number of parental individuals involved in their origin. In contrast, in other homoploid species, levels of diversity were approximately the same or slightly greater than those of parents. Despite the low diversity in S. squalidus, significant population differentiation was found for allozyme variation based on F statistics analysis (@ was significantly different from zero). However, there was no obvious geographical pattern to this variation according to the UPGMA tree constructed from genetic distances between populations (Fig. 3), and the non-significant correlation of @ and geographical distance between populations. It is likely that population differences in allozyme variation are the product of founder effects and genetic drift occurring during the colonization of new sites, and that the recent massive range expansion of S. squalidus in the British Isles has prevented a possible association of genetic and geographical distance from arising between populations due to isolation by distance. The mean coefficient of inbreeding (f) in S. squalidus was not significantly different from zero, indicating that random mating is normal in the species - as noted by Abbott & Forbes (1993). HYBRID ORIGIN OF SENECIO SQUALIDUS 135 However, a high f value was evident at the Acp-2 locus (across populations) and also in two populations (estimated across loci). The high f value at the Acp-2 locus might stem from the presence of a null allele at this locus which, if true, most likely occurs at low frequency, given that null homozygotes were not recovered (see Results section). The high f values in two populations could be due to population sub-structure (the Wahlund effect) which would promote biparental inbreeding. Alternatively, some selfing might be occurring. Abbott & Forbes (1993) showed that S. squalidus is strongly self-incompatible; however, the self-incompatibility reaction might be weakened under certain conditions (Hodgkin et al. 1988) allowing selfing to take place. Although S. squalidus was less variable than either of its parents on the basis of allozyme variation, this finding does not tally with its morphological variation. In the field, it exhibits considerable variation in leaf shape, with individuals at one extreme producing highly dissected leaves similar to S. chrysanthemifolius, and plants at the other extreme that bear almost entire leaves (var. subinteger, Druce 1927). However, most plants produce leaves intermediate between these two extremes though tending towards S. chrysanthemifolius. Other characters known to vary are capitulum size and seed type (Druce 1927). It would seem that morphologically the species might be converging towards an intermediate phenotype, but is not yet completely stabilized. ALLOPATRIC ORIGIN OF HOMOPLOID HYBRID SPECIES The morphometric and allozyme evidence reported here, combined with previous documentary evidence (Druce 1927; Kent 1956; Crisp 1972), provide support for the hypothesis that S. squalidus is a new diploid hybrid species which originated allopatrically through stabilization of hybrid material of S. aethnensis and S. chrysanthemifolius introduced to the British Isles from the hybrid zone that occurs between the two species on Mt Etna, Sicily. Previous to this study, only ten plant species had been indicated to be of homoploid hybrid origin based on rigorous molecular analysis (Rieseberg 1997; Allan et al. 1997; Wolfe et al. 1998). Mechanisms of reproductive isolation between homoploid hybrid species and their parents have been reviewed by Rieseberg (1997) and Rieseberg & Carney (1998) and shown to involve postmating and/or premating barriers. Two homoploid hybrid species, Pinus densata Masters and Encelia virginensis A. Nels, are believed to be derived from parents that produce interspecific F, hybrids that are vigorous and fertile (Wang & Szmidt 1994; Allan et al. 1997) and, in the case of E. virginensis, are known to be interfertile with both parents. In these respects, therefore, they resemble S. squalidus, which is also derived from parents which when intercrossed form vigorous fertile hybrid offspring, and whose parents are interfertile with S. squalidus (Abbott & Forbes unpublished). Natural populations of P. densata and E. virginensis are mainly allopatric with their parents and also ecologically isolated from them. However, it has been argued that they are likely to have originated in sympatry with gene flow between hybrid and parent species prevented by ecological isolation, and only later became distributed allopatrically with their parents. Diploid hybrid species of Paeonia which are also allopatric with their parents may have originated in the same way (Sang et al. 1995). In contrast, S$. squalidus is an example of a diploid hybrid species which originated through stabilization of hybrid material dispersed to locations geographically isolated from where hybridization initially took place. Although the first step in this speciation process involved hybridization between sympatric parents, the crucial process of stabilization resulting in the origin of a new hybrid species occurred at an allopatric site where there was no possibility of hybrid material backcrossing to either parent. During this process S. squalidus is likely to have also become ecologically isolated from both parents, as neither parent species appears able to form viable populations in the wild in the British Isles (Abbott, pers. obs.). Thus, the evolution of premating barriers due, for example, to pollinator behaviour and/or ecological isolation (e.g. in Penstemon, Wolfe et al. 1998; Pinus, Wang & Szmidt 1994; and Encelia, Allan et al. 1997), and postmating barriers resulting from chromosomal or genic sterility factors (e.g. in Helianthus, Iris and Stephanomeria, see Rieseberg (1997)), were not essential in the origin of S. squalidus, as isolation was maintained by geographical separation. Because of the possible ease with which homoploid hybrid speciation might occur at allopatric sites, we may need to revise current views on the importance of this mode of plant speciation and the frequency at which it occurs (Rieseberg 1997). That said, in the case of S. squalidus, considerable assistance was required from man to aid the origin of this new species. Material from Mt Etna was cultivated in the Oxford Botanic Garden for approximately 100 years before it escaped to the surrounding area and started to spread beyond Oxford. It will be of interest in the 136 R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES future to determine what genetic changes might have occurred in this material, due to recombination, segregation and mutation during this period, that could have led to the creation of a plant type that is now well adapted to conditions in the British Isles. ACKNOWLEDGMENTS We thank David Forbes for assistance with the experimental work. Special thanks are due to Professor Guiseppe Ronsisvalle (University of Catania) for his assistance and generous hospitality during the collection of material in Sicily. The research was funded by grants to R.J.A. from the NERC (GR3/7313) and The Russel Trust. REFERENCES ABBOTT, R. J. (1992). Plant invasions, interspecific hybridization and the evolution of new plant taxa. Trends in ecology and evolution 7: 401-405. ABBOTT, R. J. & FORBES, D. G. (1993). Outcrossing rate and self-incompatibility in the colonizing species Senecio squalidus. Heredity 71: 155-159. ABBOTT, R. J. & MILNE, R. I. (1995). Origins and evolutionary effects of invasive weeds, in STIRTON, C., ed. Weeds in a changing world, pp. 53-64. British Crop Protection Council, Farnham. ABBOTT, R. J., ASHTON P. A. & FORBES, D. G. (1992a). Introgressive origin of the radiate groundsel, Senecio vulgaris L. var. hibernicus Syme: Aat-3 evidence. Heredity 68: 425-435. ABBOTT, R. J., IRWIN, J. A. & ASHTON, P. A. (1992b). Genetic diversity for esterases in the recently evolved stabilized introgressant, Senecio vulgaris L. var. hibernicus Syme, and its parental taxa S. vulgaris L. var. vulgaris L. and S. squalidus L. Heredity 68: 546-556. ABBOTT, R. J., CURNOW, D. J. & IRWIN, J. A. (1995). Molecular systematics of Senecio squalidus L. and its close diploid relatives, in HIND, D. J. N., JEFFREY, C. & POPE, G. V., eds. Advances in Compositae systematics, pp. 223-237. Royal Botanic Gardens, Kew, U.K. ALEXANDER, J. C. M. (1979). The Mediterranean species of Senecio sections Senecio and Delphinifolius. Notes of the Royal Botanic Garden Edinburgh 37: 387-428. ALLAN, G. J., CLARK, C. & RIESEBERG, L. H. (1997). Distribution of parental markers in Encelia virginensis (Asteraceae: Heliantheae), a diploid species of putative hybrid origin. Plant systematics & evolution 205: 205-221. ASHTON, P. A. (1990). Multiple origins of Senecio cambrensis Rosser and related evolutionary studies in British Senecio. Ph.D. thesis, University of St Andrews. ASHTON, P. A. & ABBOTT, R. J. (1992a). Multiple origins and genetic diversity in the newly arisen allopolyploid species, Senecio cambrensis Rosser (Compositae). Heredity 68: 25-32. ASHTON, P. A. & ABBOTT, R. J. (1992b). Isozyme evidence and the origin of Senecio vulgaris (Compositae). Plant systematics and evolution 179: 167-174. CoMES, H. P. & ABBOTT, R. J. (1998). The relative importance of historical events and gene flow on the population structure of a Mediterranean ragwort, Senecio gallicus (Asteraceae). Evolution 52: 355-367. CRISP, P. (1972). Cytotaxonomic studies in the Section Annui of Senecio. Ph.D. thesis, University of London. DRUCE, G. C. (1927). The Flora of Oxfordshire, 2nd ed. Clarendon Press, Oxford. GLOVER, B. J. & ABBOTT, R. J. (1995). Low genetic diversity in the Scottish endemic Primula scotica Hook. New phytologist 129: 147-153. GOUDET, J. (1995). FSTAT. Vers. 1.2. A computer program to calculate F-statistics. Journal of heredity 86: 485-486. HODGKIN, T., LYON, G. D.& DICKINSON, H. G. (1988). Recognition in flowering plants: a comparison of the Brassica self-incompatibility system and plant pathogen interactions. New phytologist 110: 557-569. IRWIN, J. A. & ABBOTT, R. J. (1992). Morphometric and isozyme evidence for the hybrid origin of a new tetraploid radiate groundsel in York, England. Heredity 69: 431-439. JAMES, J. K. (1995). Isozyme and morphological analysis of Italian Senecio squalidus populations. B.Sc. thesis, University of St. Andrews. JAMES, J. K. (1999). Molecular analysis of the hybrid origin of the Oxford ragwort, Senecio squalidus L., and related studies. Ph.D. thesis, University of St Andrews. KENT, D. H. (1955). Scottish records of Senecio squalidus L. Proceedings of the Botanical Society of the British Isles 1: 312-313. KENT, D. H. (1956). Senecio squalidus L. in the British Isles. 1: Early records (to 1877). Proceedings of the Botanical Society of the British Isles 2: 115-118 KENT, D. H. (1957). Senecio squalidus L. in the British Isles. 3. East Anglia. Transactions of the Norfolk and Norwich Naturalists’ Society 18: 30-31. HYBRID ORIGIN OF SENECIO SQUALIDUS 137 KENT, D. H. (1960). Senecio squalidus L. in the British Isles. 2. The spread from Oxford. Proceedings of the Botanical Society of the British Isles 4: 375-379. KENT, D. H. (1963). Senecio squalidus L. in the British Isles. 7. Wales. Nature in Wales 8: 175-178. KENT, D. H. (1964a). Senecio squalidus L. in the British Isles. 4. Southern England (1940->). Proceedings of the Botanical Society of the British Isles 5: 210-213. KENT, D. H. (1964b). Senecio squalidus L. in the British Isles. 5. The Midlands (1940 ->). Proceedings of the Botanical Society of the British Isles 5: 214-216. KENT, D. H. (1964c). Senecio squalidus L. in the British Isles. 6. Northern England (1940 ->). Proceedings of the Botanical Society of the British Isles 5: 217-219. KENT, D. H. (1966). Senecio squalidus L. in the British Isles. 8. The recent spread in Scotland. Glasgow naturalist 18: 407-408. Lowe, A. H. & ABBOTT, R. J. (1996). Origins of the new allopolyploid species Senecio cambrensis and its relationship to the Canary Islands endemic Senecio teneriffae (Asteraceae). American journal of botany 83: 1365-1372. MANTEL, N. (1967). The detection of disease clustering and generalized regression approach. Cancer research 27: 209-220. NEI, M. (1972). Genetic distance between populations. American naturalist 106: 283-292. RAYMOND, M. & ROUSSET, F. (1995). GENEPOP. Vers. 1.2. A population genetics software for exact tests and ecumenicism. Journal of heredity 86: 248-249. RIESEBERG, L. H. (1997). Hybrid origins of plant species. Annual review of ecology and systematics 28: 359- 389. RIESEBERG, L. H. & CARNEY, S. E. (1998). Tansley Review No. 102. Plant hybridization. New phytologist 140: 500-624. ROHLF, F. J. (1998). NTSYSpc: Numerical taxonomy and multivariate analysis system. Vers. 2.0. Exeter Software, New York, NY. SANG, T., CRAWFORD, D. J. & STUEsSsy, T. F. (1995). Documentation of reticulate evolution in peonies (Paeonia) using internal transcribed spacer sequences of nuclear ribosomal DNA: Implications for biogeography and concerted evolution. Proceedings of the National Academy of Sciences USA 92: 6813-— 6817. SPSS Inc. (1997). SPSS Base 7.5 for Windows. SPSS Inc., Chicago. SWOFFORD, D. L. & SELANDER, R. B. (1989). BIOSYS-1: a computer program for the analysis of allelic variation in population genetics and biochemical systematics. Vers. 1.7. University of Illinois, Urbana. WALKER, R. (1833). Flora of Oxfordshire and its contiguous counties. Slatter, Oxford. WALTERS, S. M. (1963). Senecio rupestris Waldst. & Kit. and Senecio squalidus L. Proceedings of the Botanical Society of the British Isles 5: 382. WANG, X.-R. & SZMIDT, A. E. (1994). Hybridization and chloroplast DNA variation in a Pinus complex from Asia. Evolution 48: 1020-1031. WEEDEN, N. E. & WENDEL, J. F. (1989). Visualisation and interpretation of plant isozymes, in SOLTIS, D. E. & SOLTIS, P. S., eds. [sozymes in plant biology, pp. 5-45. Dioscorides Press, Portland, Oregon. WEIR, B. S. (1990). Intraspecific differentiation, in Hillis, D.M. & Moritz, C., eds. Molecular systematics, pp. 373-410. Sinauer, Sunderland, Massachusetts. WEIR, B. S. & COCKERHAM, C. C. (1984). Estimating F-statistics for the analysis of population structure. Evolution 38: 1358-1370. WoLFE, A. D., XIANG Q.-Y. & KEPHART, S. P. (1998). Diploid hybrid speciation in Penstemon (Scrophulariaceae). Proceedings of the National Academy of Sciences USA 95: 5112-5115. WRIGHT, S. (1951). The genetical structure of populations. Annals of eugenics 15: 323-354. WRIGHT, S. (1978). Evolution and the genetics of populations. Vol. 4. Variability within and among natural populations. University of Chicago Press, Chicago. (Accepted September 1999) R. J. ABBOTT, J. K. JAMES, J. A. IRWIN AND H. P. COMES 138 000: 000: OOO: 000: 000: 000: 000: 000: OOO: VeO: CeO: vOC: cl0- C80: L80° €S0: cri: LOI: 000: =) 000'T OOO: O00°1 O00°t OO0:. O00-T O00°'T O00: OO0'T 996: L96° 96L:° L86° CLE: 168: Vo6: LS8: SOS: 000: I q 000: 000: 000: 000: O00: 000: 000: 000: 000: 000: 000: 000: 000: CcvO0: CCO: cl0: OOO: 000: OOO: B C-usd [LEG 000: 000- 000: 000: 000: 000: 000: 000: 000: 000: 000: OOO: 000: 000: 000: 000° = O00: 000: a) 6SL: 0O0'T O00: O00°1 000°. OO0°. O00°T O00°T O00°'T OOO. OO00-;T OO00°'T OO00°'T OOO'l OO0O°'T OO0O'T C6L: 86° 000: I q 000: 000: 000: 000: 000: 000- 000: 000: 000: 000: 000: 000: 000: 000: OOO: 000: 80C° £90: 000 eB C-18d L99- 000: 000: 000- 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: ) SES: 000°l OO0-T 000-1 O00°1 O00°T O00-1 OOO-l OOO-'T OOO! ett: C18: L8G: O8t: OOS: LEO OOV: 6£C° CLE: q 000- 000: 000: 000: 000: 000: 000: 000: 000: 000: L99- C8I- CIV: 0C9: OOS: COC: 009: LOL: C9: B ¢-sq—-g 000: CLV: 6L0- L68° Ceo: IvL: 6ST: 186: I9OL: vV6: 1c0: VSO: LEO" OCO- Toor CSI: OCI: vOT: 000: ) 000: CCS: C6" cOl: COC: 6ST: ges 610° 6£C° 9S0- 696° 9V6: 696° O86: 6L6° 878: O88: 968: 000: q 000-1 O00: 000: 000: 000: 000: 000: 000- 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: 000: B EADY 000: OC8: OOO'I 6S: 906: 8LV: 000°T OOO 9ZL: OO00°'l 699: 89L: v86° 000°1 O00°T OO0O:1 OPT: Sor: C8C° q QOO'l O8T- OOO: LOV: v60: CCS" O00: 000: VCC: 000: Tes: CLC’ 910: 000: 000: 000: O9L: CVS: C19: B c-doy a ; sfT[P 6l 91 6 L V € C T /Snsoo'y uonendog ‘T o[qu, Ul UOATS st suONR[Ndod 0} Kay ‘pakaains O19aUag JO (6[—]) SuoNeNdod ut 190] stydrowATod ye satouanboay sfeyPy “| xtpueddy Watsonia 23: 139-147 (2000) 139 A new hybrid between a European and a Chinese species of Artemisia (Asteraceae) C. M. JAMES Department of Plant Breeding and Biotechnology, Horticultural Research International, East Malling, Kent, ME19 6BJ Bas. WURZEEL 47 Rostrevor Avenue, Tottenham, London, N15 6LA and C..A: STACE Department of Biology, University of Leicester, Leicester, LE] 7RH ABSTRACT A new hybrid, Artemisia x wurzellii C. M. James & Stace hybr. nov., is described between A. vulgaris and the Chinese A. verlotiorum. Fifteen localities are known in eastern London, and one in northern West Kent, but there is no convincing evidence that the hybrid arose in situ at any of them. Characters distinguishing the three taxa are summarized. Chromosome counts indicate that A. vulgaris is a hypodiploid and A. verlotiorum a hypohexaploid. Artemisia x wurzellii has the expected chromosome number 2n = 34 (hypotetraploid). KEYWORDS: Mugwort, Chinese Mugwort. INTRODUCTION The genus Artemisia L. (Asteraceae) is traditionally divided into three to six subgenera or sections, one of which has more recently been recognised as the separate genus Seriphidium Fourr. (Bremer & Humphries 1993). Various estimates put the number of species at 300-400, distributed in most temperate parts of the world. In the former U.S.S.R. Poljakov (1961) recognised 116 species in Artemisia sensu stricto, with a further 58 in Seriphidium. In China, Ling (1988) listed 186 species in Artemisia sensu stricto. Tutin et al. (1976) recognized 54 native European species in Artemisia sensu lato, of which only five reach the British Isles (four in Artemisia sensu stricto, one in Seriphidium). In addition, 17 species of Artemisia and one of Seriphidium have been reported as aliens in the British Isles (Clement & Foster 1994). Hybrids are evidently rare in Artemisia, few having been reported. Of these, most are sporadic in occurrence and several are dubious in identity. However, Kawano et al. (1995) reported triploid hybrids (2n = 27) between A. capillaris Thunb. ex Murray (2n = 18) and A. japonica Thunb. ex Murray (2n = 36) in Japan, and found evidence for the existence of backcrosses. Of the two species relevant to the present paper, A. verlotiorum Lamotte is native to S.W. China according to Brenan (1950), while A. vulgaris L. is almost circumboreal, being native to both the British Isles and China. Ling (1988) placed the two species in adjacent series of subgenus Artemisia section Artemisia. A. verlotiorum is treated as one of four species in series Codonocephalae (Pamp.) Y. R. Ling, and A. vulgaris as one of 16 species in series Artemisianae Y. R. Ling (but correctly to be known as series Artemisia). Diagnostic characters of the series were not given. Ling (1988) also gave a much wider distribution for A. verlotiorum in southern and south-eastern Asia. 140 C. M. JAMES, B. S. WURZELL AND C. A. STACE Artemisia verlotiorum occurs as a naturalized alien in many parts of Europe as well as in Australia and South America (Ryves 1984), presumably originating as an escape from herb- gardens. Bini Maleci & Bagni Marchi (1983) reported on a morphological-anatomical comparison of A. verlotiorum and A. vulgaris growing near Florence. The early occurrences in Britain were listed by Brenan (1950), who also provided a detailed description and bibliography. The first record was from Hounslow, Middlesex, made by C. G. Green in 1908 (OXF), and in the 1940s it was recorded widely in metropolitan Surrey and Middlesex, many of the new sites being discovered by D. H. Kent (Brenan 1950). By 1962 it was known as far north and west as W. Norfolk and Dorset respectively (Perring & Walters 1962), and today it occurs north to Easterness and Moray in northern Scotland (Vice-county Census Catalogue data in B.S.B.I. Database at Leicester). It is still most common, however, in the Greater London area, where it occurs on all kinds of rough ground, often with A. vulgaris very close by. The putative hybrid between these two species was first found by B.S.W. in August 1987 on an overgrown railway bank by Northumberland Park Station, Tottenham, Middlesex (Wurzell 1988). Between then and 1995 B.S.W. discovered 13 additional localities for the putative hybrid in eastern London, and one other has been reported there (see below). Although these localities occur in three hectads (TQ/3.7, TQ/3.8 and TQ/3.9) and three vice-counties (Surrey, Middlesex, S. Essex), and come within a few hundred metres of W. Kent, the total area of known distribution in London covers only about 14 x 7 km. Elsewhere it has been recorded only in one outlying locality in W. Kent (see below). Living and herbarium material of the putative hybrid was sent by B.S.W. in 1987 to Leicester, where it was studied by C.M.J. and C.A.S. in 1987 and 1988, and since then by C.A.S. This paper presents the evidence for its putative parentage and describes the hybrid as A. x wurzellii. MATERIALS Living plants of the putative hybrid were obtained from the original locality at Northumberland Park Station, Middlesex (TQ/34.91) and from near Walthamstow Marshes, S. Essex (TQ/35.87); of A. verlotiorum from Tottenham, Middlesex (TQ/34.90), and Abersoch, Caerns. (SH/31.27); and of A. vulgaris from Tottenham, Middlesex (TQ/34.90), near Bromsgrove, Worcs. (SO/98.68), and central Leicester, Leics. (SK/59.04). Herbarium material of the two species in LTR was also used for comparative purposes. METHODS Stem sections were cut on a freezing microtome at 25 um thick from a mid-region of the stem, which was about 5 mm wide at that point. Chromosome counts were obtained from root-tip mitoses using standard squash techniques. Pre- treatment was with 0.002M 8-hydroxyquinoline at 4° C for 24h, fixation in ethanol : ethanoic acid (3:1) for 2h, hydrolysis with 5M HCl for 6-10 min at room temperature, and staining with acetic- orcein. RESULTS Brenan (1950) provided a full description of A. verlotiorum and a detailed table contrasting it with A. vulgaris. All these differences are not repeated here, but the ones that were found to be most useful in separating the two species and in recognizing intermediates (hybrids) are covered below (see also Table 1). VEGETATIVE MORPHOLOGY AND ANATOMY The putative hybrids are generally intermediate between the parents in overall appearance, although most of the characters of the leaves are difficult to quantify precisely. This is partly because it is not easy to designate strictly comparable leaves on different plants, and partly because of variation (mainly temporal and developmental, perhaps also environmental). We found it essential to compare only the middle stem-leaves of mature flowering shoots. Such leaves, in ARTEMISIA X WURZELLIT 14] A B C FIGURE |. Middle stem-leaves of mature flowering shoots of (A) Artemisia vulgaris, (B) A. x wurzellii and (C) A. verlotiorum. The petiolar pectinations are delimited with *] both parents and hybrid, conform to a basic pattern in which a single more or less trifid terminal lobe decurrently connects with three pairs of variously incised lateral lobes which, in their turn, lead downwards to a much smaller group of pectinate appendages (here called “petiolar pectinations”) attached to that lowest section of the midrib that serves as the petiole (Fig. 1). The leaves of A. verlotiorum have narrower primary lobes with fewer secondary lobes and serrations than those of A. vulgaris, giving the impression that the former has longer leaf-lobes (although this is often not so). There is, however, overlap, and the putative hybrids overlap with both parents as well. Brenan (1950) found the same: “Segments of upper leaves usually comparatively short, lanceolate to oblong or linear” (A. vulgaris); “Segments of upper leaves conspicuously elongate, linear-lanceolate to linear” (A. verlotiorum) (our italics). Bini Maleci & Bagni Marchi (1983) also expressed the differences in only comparative terms: “Foglie polimorfe; quelle tipiche pinnatofide con lobi larghi, dentati ed auricole sviluppate” [leaves polymorphic, typically with wide, dentate lobes] (A. vulgaris); “Foglie polimorfe; quelle tipiche pinnatofide con lobi stretti ed allungati, non dentati ed auricole poco sviluppate” (A. verlotiorum) (our italics). We found the best discriminators to be the length:breadth ratio of, and the number of serrations on, the main apical segment of a middle stem-leaf of a mature flowering shoot (Table 1). Even so, there is some overlap in ranges, and the putative hybrids are closer to A. vulgaris than to A. verlotiorum in the leaf segment length:breadth ratio. The petiolar pectinations (Fig. 1) also provide a useful field character for detecting the hybrid. In A. verlotiorum they are narrow and few, but in each population of the hybrid so far examined they are broader, shorter, more abruptly acute and more numerous (Table 1); thus they have served well to distinguish these two rhizomatous taxa at a glance. The constancy of this feature in the hybrid might well indicate that we are here dealing with a single clone. Sexually reproducing A. vulgaris, in contrast, presents petiolar pectinations ranging between or even beyond those of the other two taxa in shape and number. Artemisia verlotiorum is more strongly aromatic than A. vulgaris, and Bini Maleci & Bagni Marchi (1983) found that the former had 4—5 times as high a density of glands on the lower leaf- surface (c. 20/mm? compared with 6-7/mm’). However, the very dense hair-covering present in both species, and the difficulty of removing it, render this character of limited practical use in identification. 142 C. M. JAMES, B.S. WURZELL AND C. A. STACE According to the material we have examined, the nature of the stem pith (1.e. the tissue interior to the ring of vascular bundles) differentiates the two species. In both, the pith region comprises a central area of large, thin-walled, lignified cells (white in the fresh state) surrounded by a zone of smaller, even thinner-walled, non-lignified, chloroplasted cells (green in the fresh state). In A. vulgaris the white pith occupies a much greater proportion of the whole than in A. verlotiorum, and in the putative hybrid the situation is almost exactly half-way between the two. We found this to be the best single quantitative character separating the two species and the putative hybrid (Table 1), but it should be remembered that we have not examined a large number of specimens. Brenan (1950) also noted this character, but only in terms such as “relatively broad”. He also claimed that the young stem of A. verlotiorum was more deeply furrowed than that of A. vulgaris, but this was not confirmed by our observations. The two species have an obviously different growth-habit, A. verlotiorum being extensively rhizomatous and A. vulgaris clump-forming (with very short rhizomes). The putative hybrid more closely resembles A. verlotiorum in this respect; both can, in favourable circumstances, become quite invasive. There are fairly reliable differences between the inflorescences of the two species. In A. vulgaris the inflorescence is more branched, with longer, straighter branches, and leafy only at the base, whereas in A. verlotiorum the branches are fewer and usually somewhat recurved, and the inflorescence is conspicuously leafy almost to its apex. The putative hybrid is intermediate in these characters, especially in inflorescence leafiness, although many more measurements would have to be made before ranges could be allotted with confidence to the three taxa. CAPITULUM CHARACTERS The capitula of A. vulgaris are usually obovoid-campanulate (i.e. widest distally), whereas those of A. verlotiorum are usually ovoid-ellipsoid (1.e. not wider distally than elsewhere), as pointed out by Brenan (1950) but using a different terminology. The difference, however, is not marked and is TABLE 1. CHARACTERS DIFFERENTIATING ARTEMISIA VULGARIS, A. VERLOTIORUM AND THEIR HYBRID, A. x WURZELLII A. vulgaris A. x wurzellii A. verlotiorum Length:breadth ratio of apical 5-0-6-1 4-4—7-0 9-5-15 segment of mid-stem leaf No. of serrations on apical segment 6-10 3-5 1-2 of mid stem-leaf 2—8 on each side, lanceolate, acuminate, 5-6 x as long as wide Petiolar pectinations (4)5—6(8) on each side, ovate to 2—3 on each side, lanceolate to White pith as % diam. of total pith Growth habit No. of leaves in top 20 cm of infl. Capitulum shape Length & breadth of capitulum (mm) Width of ‘bisexual’ florets (mm) No. of stamens in ‘bisexual’ flowers Receptacle shape Flowering time Seed-set in Britain Chromosome no. 80-89 tufted 9-17 obovoid-campanulate 3-1-3-7 x 2-1-2:-5 0-25-0-3 2) conical July-September yes 216 (? also 18) obovate or narrowly so, acute, 2-4(5) x as long as wide 54-69 rhizomatous 18-24 ovoid-ellipsoid 1-8—2-4 x 1-6-1:8 0-2-0-3 0 + spherical October no 2n = 34 narrowly so, acuminate, 5—7 x as long as wide 33-38 rhizomatous 25-32 ovoid-ellipsoid 3-5—4.2 x 2-4-3-0 0-4-0-51 5 low-domed October-November no 2mv= 505592 (? also 48, 54) ARTEMISIA X WURZELLIIT 143 difficult to quantify. The putative hybrid is much closer to A. verlotiorum in this respect. Brenan’s measurements for the two species broadly agree with the ranges seen by us, but the capitula of the putative hybrid are smaller than in either species (mean 2-1 x 1-7 mm) (Table 1). Differences in both the peripheral female and inner bisexual florets in the two species were also recorded by Brenan (1950). There are usually some additional abortive bisexual florets right in the centre of the capitulum. Brenan found (2) 5—22 and (0) 2-6 (11) bisexual florets for A. vulgaris and A. verlotiorum respectively, but our figures (5—8 and 4-8) did not confirm these differences; our range for the putative hybrid was 4—7. Our results do, however, confirm a difference in the widths of the bisexual florets of the two species (Table 1); the putative hybrid has narrower florets than in either parent. Like Brenan (1950), we found no differences in the numbers of glands on the outside of the corollas of the two species, although Bini Maleci & Bagni Marchi (1983) (and others earlier) claimed that they were more abundant in A. verlotiorum (the more aromatic species). We did find a slight difference in the shape of the receptacle in the two species. In A. vulgaris it is elongated-conical whereas in A. verlotiorum it is low-domed; the putative hybrid is closer to the latter but often higher-domed (sub-spherical). PHENOLOGY AND FERTILITY Artemisia vulgaris is a late-summer flowerer, reaching anthesis mainly from July to September in southern England. A. verlotiorum flowers distinctly later, from October to November, and in many years is damaged by frost before anthesis. The natural flowering periods of the two species do not overlap, but plants of A. vulgaris that are cut down in summer will often produce late-flowering stems, and, in the absence of frosts, undoubtedly both species could be in flower simultaneously. The putative hybrid, when growing in the same general area as the two species, reaches flowering at an intermediate time, mostly in October (two to four weeks before A. verlotiorum). Its flowering stems are, however, very frost-susceptible and often do not reach anthesis in the wild. Presumably due to its late flowering A. verlotiorum has not been found to produce ripe fruits in this country. No fruits have been found on the putative hybrid. The florets of the putative hybrid in the position of the bisexual florets of the parental species are much reduced, without the conspicuously distally dilated corolla found in its two parents, and anthers are completely lacking; only a reduced ovary, style and stigma in a very reduced corolla are present. Hence the putative hybrids are probably totally male-sterile. CHROMOSOMES Clear chromosome counts (Fig. 2) were made from A. vulgaris from London, Birmingham and Leicester (all 2n = 16), A. verlotiorum from London and Caernarvonshire (2n = 52 and 50 respectively), and the putative hybrid from both London sites (2n = 34) (Table 2). The numbers of chromosomes possessing satellites in these three taxa were two, six and four respectively. TABLE 2. CHROMOSOME NUMBERS OF ARTEMISIA TAXA IN THE PRESENT STUDY Species Locality Chromosome number Artemisia vulgaris Rough ground, Tottenham, London, Middlesex, v.c. 21 2n = 16 By Tardebigge Canal, east of Bromsgrove, Worcs., v.c. 37 2n = 16 On waste ground, central Leicester, Leics., v.c. 55 2n = 16 Artemisia verlotiorum By public toilets, Abersoch, Lleyn, Caerns., v.c. 49 an = 50 Rough ground, Tottenham, London, Middlesex, v.c. 21 Zio Artemisia x wurzellii Railway bank, Tottenham, London, Middlesex, v.c. 21 (locality 14) 2n=34 Rough ground, near Walthamstow Marshes, London, S. Essex, v.c. 18 (locality 11) 2 = 34 144 C. M. JAMES, B.S. WURZELL AND C. A. STACE Cae ee ‘ ee wt Sy wy © e i» C ; FIGURE 2. Chromosomes of: A. Artemisia vulgaris (2n = 16), B. A. x wurzellii (2n = 34) and C. A. verlotiorum (2n = 52), from Tottenham, London, v.c. 21. Magnification x 1755. DISCUSSION AND CONCLUSIONS IDENTITY OF PUTATIVE HYBRIDS Despite the paucity of quantitative data demonstrating the intermediacy of the putative hybrids, there is strong circumstantial evidence that they are indeed hybrids between A. verlotiorum and A. vulgaris (Table 1). Their overall appearance is convincingly intermediate; this is particularly indicated by the width and serration of the leaf-lobes, especially the apical leaf-segment of the middle stem-leaves, and by the leafiness of the inflorescence. The pith anatomy and capitulum shapes are also intermediate between those of the two putative parents, as is the flowering time in London. The putative hybrids are totally male-sterile, no anthers being formed, a condition never observed in either parent. The fact that other characters separating the two species are not intermediate in the putative hybrid (e.g. rhizome growth, capitulum shape and size, petiolar pectinations) does not argue against its hybrid origin, since this situation is common in nature. Hybrids are rarely exactly intermediate in every diagnostic character, due to the varied genetic backgrounds of the observed phenotypes. Nevertheless, probably the most convincing evidence of hybridity comes from the chromosome numbers. In our London material A. verlotiorum had 2n = 52, A. vulgaris had 2n = 16, and the putative hybrids had the expected 2n = 34 (26 + 8). In Asteraceae tribe Anthemideae, and in the genus Artemisia, the diploid number 2n = 18 is by far the commonest reported chromosome count (Bolkhovskikh ef al. 1969, and later plant chromosome number indices). Tetraploids with 2n = 36 and hexaploids with 2n = 54 are also well distributed in Artemisia, but several aneuploid counts, such as 2n = 16, 14 and 34, exist in addition. Hence our plants of A. vulgaris should probably be viewed as aneuploids, 2n = 2x — 2 = 16, where x (the base number) = 9. . ARTEMISIA X WURZELLII 145 The majority of reported counts of A. vulgaris are 2n = 16 (widespread in Europe and in North America, Russia, India and Japan), but there are a few 2n = 18 counts (India and Caucasus) as well. The single previous British count, from Surrey, was 2n = 16 (Morton 1977). Most previous counts of A. verlotiorum are hexaploids with 2n = 54 (from Italy, Germany, Japan and Russia), but there are also two, presumably aneuploid, counts of 2n = 48 from China and Spain. Our counts of 2n = 50 and 52 for A. verlotiorum are new numbers for this species, and neatly complete the reducing aneuploid series 54, 52, 50, 48. It thus appears that both A. vulgaris and A. verlotiorum exist as both euploid and reducing aneuploid cytotypes. On known evidence of the range of chromosome numbers of the two parental species, hybrids could occur with chromosome counts of 2n = 32, 33, 34, 35 or 36. The number of satellited chromosomes that we recorded in the three taxa confirm their suggested ploidy levels. FORMAL DESCRIPTION OF HYBRID The evidence presented above leads us to conclude that the intermediate plants are hybrids between Artemisia vulgaris and A. verlotiorum. They are described below after their discoverer, B. S.W. Artemisia x wurzellii C. M. James & Stace, hybr. nov. Hybrida inter A. vulgarem et A. verlotiorum intermedia sed ab ambabus floribus sterilibus (anantheris) et capitulis minoribus (1-8—2-4 mm longis) differt. Caules medulla viridi et medulla alba magis minusve aequicrassa. Folia caulina media segmentis apicalibus 4-4 — 7 x longioribus quam latis. HOLOTYPUS: Overgrown railway bank, Northumberland Park Station, Tottenham, Haringey, London, Middlesex, map reference TQ/34.91. 2 October 1987, B. S. Wurzell s.n. (LTR). ORIGIN AND DISTRIBUTION OF ARTEMISIA x WURZELLII Since the original discovery in 1987, B.S.W. has found 13 additional sites for A. x wurzellii in eastern and south-eastern London. sumey, v.c. l7: 1 Roadside, near Stockwell Underground Station, Lambeth, TQ/30.76, 1988, still there 1990, not visited since. 2 Roadside among bushes, near Ruskin Park, Camberwell, Southwark, TQ/32.75, 1990, not visited since. 3 Grassland and scrub, off Neate Street, Burgess Park, Walworth, Southwark, TQ/32.77 & TQ/33.77, 1995, still there 1998. 4 Shrubbery weed by road, Dog Kennel Hill, East Dulwich, Southwark, TQ/33.75, 1992, not visited since. 5 Waste ground, off Tooley Street, by London Bridge Station, Southwark, TQ/33.80, 1988, not visited since. 6 In rough grass, near Straker’s Road, Peckham Rye Common, Southwark, TQ/34.75, 1993, not visited since. 7 Shrubbery weed in small park, near Chambers Wharf, Bermondsey, Southwark, TQ/34.79, 1988, not visited since. 8 Disturbed soil, King Stairs Gardens, off Brunel Tunnel Roundabout, Rotherhithe, Southwark, TQ/34.79, 1992, not visited since. 9 By footpath, near Redriff Road, south end of Russia Dock Woodland, Rotherhithe, Southwark, TQ/36.79, 1993, not visited since. 10 In newly landscaped shrubbery, near Salter Road, north end of Russia Dock Woodland, Rotherhithe, Southwark, TQ/36.80, 1993, not visited since. 146 C. M. JAMES, B. S. WURZELL AND C. A. STACE South Essex, v.c. 18: 11 Rough ground, near Coppermill Lane, off Walthamstow Marshes, Waltham Forest, TQ/35.87, 1987 & 1988, not visited since. 12 Stony waste ground, Bow Creek Ecological Park, Canning Town, Newham, TQ/39.81, 1994, still there 1998. Middlesex, v.c. 21: 13. Garden weed in new housing estate, Watermint Quay, Clapton, Hackney, TQ/34.88, 1994, still there 1998. 14 Overgrown railway bank, Northumberland Park Station, Tottenham, Haringey, TQ/34.91, August 1987, still there 1998. In addition, two other records, neither of which we have any reason to doubt, have come to our notice: 15 West Kent, v.c. 16: on rubbish tip, Swanscombe, TQ/60.74, G. Kitchener & M. Keene, 1994, site cleared soon after (Bull. Kent Field Club 40: 35-36 (1995)). 16 Middlesex, v.c. 21: weed in shrubbery, Fair Street, St John’s Estate, Bermondsey, TQ/32.78, J. R. Palmer, 1989, comm. J. E. Smith. The first-discovered colony (no. 14 above) extends for at least 50 m along a railway bank together with common coarse perennials of such habitats, including Armoracia rusticana, Arrhenatherum elatius, Calystegia silvatica, Rumex obtusifolius and Urtica dioica. Artemisia vulgaris occurs close by, but the nearest colony of A. verlotiorum is at least 400 m distant on Tottenham Marshes. Moreover, the two parents do not normally overlap in flowering times, and A. verlotiorum has not been seen to set seed in this country. A. vulgaris might also fail to set seed from flowers opening as late as October. Hence, although the hybrid was at first thought to have arisen spontaneously in sifu, it seems more likely that it arose elsewhere, perhaps not in Britain, and was inadvertently dumped on the railway bank. All of the other localities are on rough or disturbed ground or in recent habitats created by gardening, and it is very likely that the present distribution is the result of soil and waste being transported by man. One need look no further than Fallopia japonica to understand how effective this method of dispersal can be! It is, indeed, possible that all 16 colonies represent a single clone which was introduced, perhaps as a medicinal plant, from southern Europe, where A. verlotiorum is widespread and regularly forms fruits (and where the hybrid should be sought), or even from China. On the other hand, the chromosome number of the hybrid indicates that its A. verlotiorum parent had 2n = 52, a number not previously reported but characteristic of the Tottenham clone. There are counts of only 2n = 48 and 54 (and 16) for this species from southern Europe (Italy and Spain); such plants could not have produced a hybrid with 2n = 34 by crossing with plants of A. vulgaris with 2n = 16. Hence a London origin for the hybrid is certainly possible. ACKNOWLEDGMENTS We dedicate this paper to Douglas H. Kent, who was the leading expert on the flora of Middlesex (especially its introduced species), who considerably extended our knowledge of the distribution of A. verlotiorum in London, and who was a very good friend to C.A.S. and B.S.W. over many years. We thank Miss A. P. Conolly for providing living material of A. verlotiorum from Abersoch. REFERENCES BINI MALEcI, A. & BAGNI MARCHI, A. (1983). Artemisia vulgaris L. ed A. verlotiorum Lamotte: studio di alcuni caratteri morfo-anatomici distintivi delle due specie. Webbia 37: 185-196. BOLKHOVSKIKH, Z., GRIF, V., MATVEJEVA, T. & ZAKHARYEVA, O. (1969). Chromosome numbers of flowering plants. Academy of Sciences of the U.S.S.R., Leningrad. ARTEMISIA X WURZELLII 147 BREMER, K. & HUMPHRIES, C. J. (1993). Generic monograph of the Asteraceae-Anthemideae. Bulletin of the Natural History Museum, Botany 23: 7-177. BRENAN, J. P. M. (1950). Artemisia verlotiorum Lamotte and its occurrence in Britain. Watsonia 1: 209-223. CLEMENT, E. J. & FOSTER, M. C. (1994). Alien plants of the British Isles. Botanical Society of the British Isles, London. KAWANO, S., NAGAI, Y. & HOSHTYA-USHIDA, S. (1995). A study on the natural hybrid swarms of two Artemisia species, A. capillaris and A. japonica (Compositae) in central Honshu, Japan, with special reference to its biological status. Journal of phytogeography and taxonomy 42: 133-153. LING, Y. R. (1988). The Chinese Artemisia Linn. - the classification, distribution and application of Artemisia Linn. in China. Bulletin of Botanical Research, Harbin 8: \-61. MorTON, J. K. (1977). A cytological study of the Compositae (excluding Hieracium and Taraxacum) of the British Isles. Watsonia 11: 211-223. PERRING, F. H. & WALTERS, S. M. (1962). Atlas of the British flora. Thomas Nelson & Sons, London & Edinburgh. POLIAKOV, P. P. (1961). Artemisia, in SHISHKIN, B. K. & BOBROV, E. G., eds. Flora U.R.S.S. 26: 425-631. Editio Academiae Scientiarum U.R.S.S., Moscow & Leningrad. RYVES, T. B. (1984). Artemisia L. species in the British Isles. Watsonia 15: 109-112. TUTIN, T. G., PERSSON, K. & GUTERMANN, W. (1976). Artemisia L., in TUTIN, T. G. et al., eds. Flora Europaea 4: 178-186. Cambridge University Press, Cambridge. WURZELL, B. S. (1988). Wurzell’s weeds. BSBI news 49: 53. (Accepted September 1999) a . Watsonia 23: 149-159 (2000) 149 Historical records of Lactuca serriola L. and L. virosa L. in Britain, with special reference to Cambridgeshire (v.c. 29) P. H. OSWALD 33 Panton Street, Cambridge, CB2 1HL “Most of our botanical publications are taken one from the other; and thus if an eminent botanist has in the course of his researches fallen into a mistake, the error has been propagated.” (Jacson 1797, pp. 229-230). ABSTRACT Thomas Johnson (1633) was the first British botanist to realise that three sorts of large wild lettuces occur in England, though he apparently thought that what is now called forma integrifolia of Lactuca serriola L. was related to L. virosa L. Records of L. virosa and L. serriola in Cambridgeshire (v.c. 29) are reviewed here. The long-held belief that John Ray (1660) recorded both of them in the county is shown to be incorrect: he, in fact, saw two forms of L. serriola. This paper casts doubt on some of the eighteenth and nineteenth century records of L. virosa in Cambridgeshire. The present distribution of the Lactuca taxa considered in the paper is briefly discussed. KEYWORDS: Asteraceae, wild lettuces, distribution, Thomas Johnson, John Ray, Kent, Middlesex. INTRODUCTION Descriptions of Lactuca serriola and L. virosa in most British Floras are unsatisfactory because the existence of two forms of each with differing leaf characters is ignored. The brief accounts in Stace (1991, 1995, 1999) are accurate but lack many of the useful characters. The differences between the four taxa are, however, satisfactorily described in Plant crib 1998 (Rich & Jermy 1998, pp. 292-293), with improvements in the corrigenda and addenda in the reprint of 1999, and will be fully treated in Volume 4 of Peter Sell and Gina Murrell’s Flora of Great Britain and Ireland (Cambridge University Press, Cambridge). Both species are biennials or overwintering annuals, initially with a basal rosette and later with leafy stems, branched, at least above, and usually with numerous capitula. Stunted plants occur, but well-grown plants of L. virosa can exceed 2:5 m and L. serriola can attain 2 m. However, because L. serriola often grows in recently disturbed sites, plants that have reached full size in the absence of competition are frequent, whereas L. virosa grows mainly in post-disturbance grassy vegetation, where it is reduced in size by competition (R. N. Carter, pers. comm.). The main panicle of L. virosa is characteristically rhomboid, with little obvious leafage within it; that of L. serriola, at least in taller plants, is less regular and usually less pointed or even flat-topped, with the lower branches leafy. L. virosa begins to flower about a fortnight earlier than L. serriola, normally in June, and the diameter of the open capitula is often double that of L. serriola (up to 20 mm: pace Clapham 1952 and later editions!); the capitula of both species close by late morning, so these are often missed. Fruit characters are much more reliable than leaf characters: the fresh achenes of L. virosa are strikingly purple or maroon and larger than the olive-grey achenes of L. serriola. Another useful diagnostic character, which is unfortunately ignored in most British Floras, is the smell when the plant is broken or bruised: L. virosa smells like Opium Poppy (Papaver somniferum L.), L. serriola like Garden Lettuce (Lactuca sativa L.). Two foliage forms of L. serriola occur, sometimes intermixed but often in separate populations, the type with deeply runcinate-pinnatifid cauline leaves, the other with undivided or shallowly lobed cauline leaves. There are also two forms of L. virosa, but here it seems that populations are more often mixed and the difference between the two is less striking and has rarely been taken note of: the type has unlobed, sinuate-dentate cauline leaves throughout; a form with the higher stem leaves pinnatisect with wide lobes has been named var. /actucarii (Lamotte) Rouy, though P. D. Sell proposes to reduce the name to forma level in Flora of Great Britain and Ireland. 150 P. H. OSWALD Prince & Carter (1977) reviewed the characters distinguishing L. serriola, L. virosa and the related but smaller L. saligna L. and revised the taxonomic treatment of L. serriola, giving to its unlobed-leaved form the name forma integrifolia (S. F. Gray) S. D. Prince & R. N. Carter. Later, Carter & Prince (1982) discussed the history of the taxonomic treatment of L. serriola and L. virosa. They showed that, although Ray (1690) provided a “treatment of the British taxa ... as good as any published since” and Linnaeus (1756) “laid the foundation for the taxonomy of L. serriola and L. virosa as we know it’, by “the second half of the nineteenth century the existence of an unlobed-leaved form of L. serriola was almost completely overlooked in Britain”, the general practice being “to call all unlobed-leaved plants L. virosa without even according varietal status to those that were really L. serriola”. They concluded that, as a result, “most pre- 1930 L. virosa records could as easily refer to L. serriola as to L. virosa”. In this paper I show that the existence of three of the four taxa of large wild lettuces was clearly understood by Thomas Johnson by 1633 but that John Ray (1627-1705) at first failed to recognise them. As a combined result of this fact and the later confusion described by Carter & Prince (1982), it has been wrongly believed since the time of John Martyn (1727), through that of C. C. Babington (1860), and right up to the present day that L. virosa was first recorded in Cambridgeshire by Ray (1660). THE EARLY SEVENTEENTH CENTURY In his first account of a botanical excursion into Kent, Thomas Johnson (1629) recorded “Lactuca agrestis odore opi” on the Isle of Sheppey on 14 July 1629. This is the first localised British record of L. virosa (Clarke 1900, p. 85), which is “still present here” (Gilmour 1972, p. 56). In a list of 1632 of “the plants growing on Hampstead Heath and in the places visited ... on the way there and back”, he recorded “Lactuca syl. odore viroso folijs dissectis, Lactuca sylv. prior, Trag. sily. Matth.” and also “Lactuca syl. alter odore magis viroso folijs non dissectis, Lactuca agrestis odore opij Lob. Ad. Endivia major & I. Trag. Thesion, Lugd.”. These names may be interpreted thus: “Wild Lettuce with a fetid smell and divided leaves; the first wild Lettuce of Jerome Bock (Tragus 1552, p. 259; Figure 1| in this paper); the wild Lettuce of P. A. Matthioli (Matthiolus 1558, pp. 295-296)” and “The second wild Lettuce, with a more fetid smell and undivided leaves; the field Lettuce with the smell of opium of Stirpium adversaria nova of Pierre Pena and Matthias de l’ Obel (Pena & de Lobel 1571, p. 89); the greater, first Endive of Jerome Bock (Tragus 1552, p. 267); Thesion of Historia generalis plantarum by J. D’Aléchamps, published at Lyons (Lugdunum) (Dalechampius 1587, pp. 564-565; Figure 2 in this paper)’. Clarke (1900, p. 85) listed the former as the first British record of L. serriola, and Gilmour (1972, p. 128) identified the two plants as L. serriola and L. virosa, while Kent (1975, p. 491) accepted the latter identification but seems to have had doubts about the former. Both Clarke and Kent give “Hampstead Heath” as the locality, but, for reasons that will soon become apparent, I believe that the two plants were found on the way there or back, “betweene London and Pancridge [?Old St Pancras] Church’, and were L. serriola forma serriola and forma integrifolia respectively, though Johnson had not yet worked this out. If my theory is correct, these are the first British records for both forms of L. serriola. In the chapter of his Herball entitled “Of Lettuce”, John Gerarde (1597, p. 238) had stated that there were sorts “both wilde and tame”, which he purposed “to laie down”; but in fact he dealt only with the cultivated kinds. In his revision of Gerarde’s work, Johnson (1633) rectified this omission, in Chapter 38 of Book 2 (p. 309), thus: There are three sorts of wilde Lettuce growing wilde here with vs in England, yet I know not any that haue mentioned more than two; yet I thinke all three of them haue beene written of, though two of them be confounded together and made but one (a thing often happening in the history of Plants) and vnlesse I had seene three distinct ones, I should my selfe haue beene of the same opinion. | The first and rarest of these hath long and broad leaues, not cut in, but only snipt about the edges, and those leaues are they that are on the lower part of the stalke almost to the midle thereof: then come leaues from thence to the top, which are deepely diuided with large gashes: the stalke if it grow in good grounds exceeds the height of a man, (for I haue seene it grow in a garden to the height of eight or nine foot) .... The whole plant is full of a clammy milky iuice, which hath a very strong and grieuous smell of Opium [1.e. the latex of Papaver somniferum]. HISTORICAL RECORDS OF LACTUCA SERRIOLA AND VIROSA 151 2 This hath broad leaues only cut about the edges, but not altogether so large as those of the last described: the stalke, which commonly is two cubits or better high, is also smooth, and diuided into many branches, bearing such floures as the last described; and this also hath a milky iuice of the same smell as the last described, from which it differs only in magnitude, and that this hath all the leaues whole, and not some whole and some diuided, as the former. 3 This in stalkes, floures and seedes is like to the last described, but the leaues are much different, for they are all deeply gashed or cut in like as the leaues of Succory, or Dandelion. This also is full of a milky iuice, but hath not altogether so strong a sent of Opium as the two former, though it partake much thereof. The stalke of this is sometimes a little prickly, and so also is the middle rib vpon the backeside of the leafe. Johnson illustrated only the first and last of these plants (p. 309), labelling them respectively “1 Lactuca syl. maior odore Opij. The greater wilde Lettuce smelling of Opium.” and “3 Lactuca sylvestris folijs dissectis. The wilde Lettuce with the diuided Leafe.” The illustrations are poor representations of the taxa; that of the last is a reduced and reversed version of Fuchsius’ (1542, p. 301) full-page woodcut of “Lactuca syluestris”’, similar to that of Tragus (1552) (Figure 1 in this paper) but single-stemmed. In the section of Chapter 38 headed “The Names” (p. 310), Johnson wrote: 1 I take the first of these to be the Lactuca Sylvestris of Dioscorides and the Ancients, and that which the Authours of the Adversaria [Pena & de Lobel 1571, p. 89] gaue vs vnder the title of Lactuca agrestis scariole hortensis folio, Lactuce flore, Opij odore vehementi, soporifero & viroso [field Lettuce with the leaf of garden chicory, the flower of Lettuce and a strong, soporific and stinking / poisonous smell of Opium]. 2 This is the Endiuia of Tragus [1552], pag. 268. and the Thesion of Daleschampius [1587], pag. 564. Bauhine [Bauhinus 1623, p. 123] confounds this with the former. 3 This is the Lactuca Sylvestris prior, of Tragus [1552, p. 259]: the Lactuca Sylvestris of Matthiolus [1558, pp. 295-296], Fuchsius [1542, pp. 298-303], Dodoneus, and others: it is the Seris Domestica of Lobell [Pena & de Lobel 1571, p. 86]. There is no doubt that the first of these plants is Lactuca virosa (var. lactucarii from the description of the upper leaves) and the last L. serriola forma serriola, to which Johnson allots the two synonyms he gave to his first lettuce of 1632. The synonyms Johnson provides for his second sort here are the last two names he gave to the “second wild Lettuce” of 1632, when it seems that he too was still confounding two of them together and making but one, since he applied to one taxon names that he later divided between two.’ Both Tragus (1552) and Dalechampius (1587) (Figure 2 in this paper) show a plant with the undivided leaves of L. serriola forma integrifolia, but it is odd that Johnson suggests that it too has a “smell of Opium’, which is true only of L. virosa. However, it seems that he found it growing with L. serriola forma serriola, as he gives a joint locality, “plentifully betweene London and Pancridge Church, about the ditches and highway side”, whereas he records L. virosa only from John Parkinson’s garden, grown from seeds sent by John Goodyer, who had found it growing in Hampshire (pp. 309-310). It is surprising that he does not mention his locality of 1629 here, but perhaps, having realised that neither of the plants he had found in 1632 was L. virosa, he was uncertain that he had really seen it in Kent either. JOHN RAY’S ACCOUNTS In his Catalogus plantarum circa Cantabrigiam nascentium, Ray (1660, Pp. 82-83) included two kinds of large wild lettuces and a smaller one that he then thought was “not yet described’, i.e. Lactuca Saligna, to which he gave the name “The least cut-leaved wild Lettuce’.” To the first large lettuce, “Lactuca sylvestris costa spinosa C.B.”, he gave the name “Cut-leaved wild Lettuce” and to the second “Endive-leaved wild Lettuce’. He described the former as growing “Ad agrorum margines [By the margins of arable fields]. As by the way side from Ely to Cambridge, within a mile or two of Ely.” The Latin synonyms given for it, which include “sylv. foliis dissectis Ger. emac [i.e. in Johnson’s emaculated (or emended) edition of Gerarde of 1633] and “Seris domestica Lob.” [Pena & de Lobel 1571, p. 86], make it clear that this was L. serriola forma serriola.’ No locality or habitat is given for the second kind, but the first Latin name given is 152 P. H. OSWALD “Lactuca sylvestris 2. Ger. emac. [1.e. Johnson’s second kind of wild Lettuce]. After this there follow five synonyms, the first three including “viroso” and clearly relating to L. virosa; the last two, however, are “Endivia Trag. p. 268.” and “Thesion Dalechampii Lugd. p. 564.”, which Johnson (1633) gave as names for his second kind. At the end of his account of his own second kind, Ray included a rather contorted note in Latin’, which is poorly translated by Ewen & Prime” (1975, p. 77). This note is better rendered thus: Johnson, in his edition of Gerarde, creates two kinds of wild lettuce with the smell of opium, which he thinks are confused in the works of other writers; and he attributes to the first sort earlier authors’ names which we think are more appropriate to his second sort, since we believe that our plant, which is one with undivided leaves and is Gerarde’s second one, is the only one described by botanical writers, rather than, as he suggests, Gerarde’s first. This shows that, at this time, Ray believed there were only two kinds of wild lettuce other than L. saligna. Moreover, he clearly thought that his own second kind fitted Johnson’s (1633) second description rather than his first. On this evidence alone, Martyn (1727), Babington (1860) and others might perhaps be excused for having reached the conclusion that Ray’s second large Lactuca was L. virosa, but Ray’s subsequent works make it certain that it was in fact L. serriola forma integrifolia.® Later writers, including Ewen & Prime (1975), seem not to have found the entry in the appendix of 1685 to Ray’s Cambridge catalogue about his second lettuce, “Lactuca sylvestris 2. Ger. emac.”. This may be translated thus: Delete all the synonyms. In their place add ‘Lactuce sylvestris sive Endivie multis dicte folio laciniato dorso spinoso varietas [A variety of the wild Lettuce, or Endive as it is known to many people, with a laciniate leaf with a spiny underside] J.B.’’ This does not differ from Gaspard Bauhin’s wild Lettuce with a spiny midrib [Bauhinus 1623, p. 123] in any respect other than in the leaves, which in it are not laciniate even on the stem. We now retract what we propounded about this plant in the Catalogue and we assent to Johnson’s opinion.® The last sentence had appeared already in Ray’s Catalogus plantarum Anglie (1670, 1677), in which he included three taxa of large wild lettuces. In this work Ray indicated the plants that grew wild around Cambridge (“Agri Catabrigiensis alumnas”’) by placing “C.” before their names, and by this means he showed that he had recorded only two of the lettuces in Cambridgeshire and not “The greater strong-scented wild Lettuce”. Even earlier, William How (1650) and Christopher Merrett (1666) had listed four wild lettuces including “Lactuca sylvestris altera folio non laciniato, odore minus vehementi / vehemente [The second wild Lettuce with the leaf not laciniate, with a less powerful smell]”. Presumably, by the time he wrote his retractions, Ray had seen Lactuca virosa for the first time, somewhere outside Cambridgeshire, and had realised that his second taxon in the Cambridge catalogue was not it. At any rate, in his later botanical works he distinguished three taxa of large annual wild lettuces’. Thus, in Ray’s (1686) Historia plantarum, he listed “The greater strong- sented wild-Lettuce”, “Cut-leaved wild Lettuce” and “Endive-leaved wild Lettuce”: each is equated with one of Johnson’s three kinds, but they are reordered to put Johnson’s second kind last. About this kind Ray commented: Johnson, the emender of Gerarde, makes this species a variety of the first species [L. virosa]'°, but J. Bauhin of the second [L. serriola forma serriola: see note 7], whom we also support, since it is less tall than the former and has smaller leaves and also a less fetid smell, at least to our nostrils. On the contrary, it does not differ in any character from the second species except in the shape of the leaves, which in this species are not laciniate even on the stalk. It grows in the same places as the preceding [i.e. second] species." The three kinds are listed in the same order in the Synopsis (Ray 1690, p. 41), where the second (i.e. L. serriola forma serriola) is called “Milder-scented, cut-leaved wild Lettuce’; the third, which Ray here calls “Lactuca sylv. folio non laciniato. Sylv. 2. Ger. emac. [Wild Lettuce with the leaf not laciniate;. the second wild one in the emended Gerarde.] Endive-leaved wild Lettuce’, 1s described as growing “with the preceding one, of which it is a variety, but found more rarely”. & HISTORICAL RECORDS OF LACTUCA SERRIOLA AND VIROSA 153 THES SON Dalechamp. Endiuia Tragi. FIGURE 1. Woodcut of “Lactuca siluestris” from FIGURE 2. Woodcut of “Thesion” from J. D’Aléchamps’ Book 1 of Jerome Bock’s Commentaries General History of Plants (Dalechampius 1587), with (Tragus 1552), interpreted by Thomas Johnson uncut leaves which led Thomas Johnson (1633) to (1633) in his edition of Gerarde’s Herball as regard it as a representation of Lactuca serriola forma representing Lactuca serriola forma serriola. integrifolia. It is a redrawn and reversed image of Tragus’ (1552) illustration of “Endiuia”. EIGHTEENTH AND EARLY NINETEENTH CENTURY CAMBRIDGESHIRE RECORDS John Martyn (1727, p. 21) listed “Lactuca sylvestris costa spinosa” and “‘Lactuca sylvestris major odore Opii” in his account of the Cambridgeshire flora, equating the latter with “Lactuca sy!vestris 2. Ger. emac. Cat. Cant. [i.e. Ray 1660]. He was thus the first of a long line of Cambridgeshire botanists to attribute a record of L. virosa to John Ray. In a working volume containing Ray’s Catalogus and his own Methodus bound together, preserved in the Library of the Department of Plant Sciences of the University of Cambridge, he wrote out a list headed “1727 Desiderata’”’, among them the three Lactuca taxa of Ray’s Catalogus (including L. saligna); he later deleted many of the taxa listed here, but none of the three Lactuca taxa, which suggests that he himself never saw any of them in Cambridgeshire. In his Plante Cantabrigienses (essentially a table providing the Linnaean names for the plants recorded in Ray’s Catalogus and its appendices and in John Martyn’s Methodus), John’s son, Thomas Martyn (1763a, p. 18), listed Lactuca virosa, its var. B and L. saligna. The table shows that he too regarded Ray’s “Lactuca sylvestris 2.” as L. virosa, while equating Ray’s “Lactuca sylvestris costa spinosa” (the cut-leaved form of L. serriola) with its var. B.'° Thomas Martyn also annotated an interleaved copy of his father’s Methodus, which is also in the Library of the 154 P. H. OSWALD Department of Plant Sciences at Cambridge. On a blank page opposite “Lactuca sylvestris major odore Opii” he wrote “Near Clare hithe July 27 1759’, but later he added a line linking this to “Lactuca sylvestris costa spinosa” instead. In a handwritten index of localities inserted at the end of the volume, he entered “Lactuca sylvestris costa spinosa” under “Ely” and “Lactuca sylv. maj. odore Opu” under “Clare hithe”; the latter has not been amended, but nevertheless it seems likely that Thomas Martyn never saw L. virosa. It may be significant that neither species is listed in his Herbationes Cantabrigienses (Martyn 1763b), which consists of lists of plants under localities “comprehended in thirteen Botanical Excursions”. Richard Relhan included only two large lettuces in his Flora Cantabrigiensis (1785, 1802, i820) — L. virosa, “Strong-scented wild Lettuce”, and “L. Scariola” (a synonym for L. serriola), “Prickly Lettuce”, the latter evidently forma serriola as it is equated with the second kind in Ray’s Synopsis and with Johnson’s “L. sylvestris, foliis dissectis”. In 1785 the only locality given for the former was “Burwell-pit’ and for the latter “Jn the Isle of Ely’. In the later editions “Jn the Road to Cottenham, by the second Bridge from Histon” and “Ditch near Denny Abbey” were added for L. virosa and “Hazenfield [Haslingfield], in the Lane leading into the Village from Cambridge’, “Between Histon, and Rampton’, “In a Lane leading from Long Stanton towards Swavesey” and “Burwell Pit’ were added for L. serriola. There seems little reason to doubt the localities for L. serriola, but what about those for L. virosa? The fact that Burwell Pit was later given as a locality for L. serriola suggests that the plant recorded there in 1785 may have been its forma integrifolia. The specimen of “Lactuca Scariola” figured in English Botany (Smith & Sowerby 1795, no. 268) was sent to Sowerby from “near Denny Abbey” by the Rev. John Hemsted, so the record from a ditch nearby may well have related to its unlobed-leaved form.'* Similarly, the locality by the road to Cottenham is suspiciously close to that between Histon and Rampton. As implied by Johnson (1633) and specifically noted by Ray (1690), the two forms of L. serriola often grow together, whereas, at least until the last few decades, L. virosa seems rarely to have occurred with L. serriola.’” C. C. BABINGTON AND HIS CONTEMPORARIES Despite Babington’s (1860) mistake in identifying Ray’s (1660) “Lactuca sylvestris 2.” as L. virosa, there is no doubt that he himself knew the latter, at least at Cherry Hinton: the only new locality for this species given in his Flora of Cambridgeshire is “Chalk-pit-close, Hinton”, on his own authority, though the entry “Burwell Pit!; Relh.” signifies that he claimed to have seen a specimen of Relhan’s from Burwell Pit. The Cherry Hinton locality is supported by three specimens in CGE, the earliest collected by J. S. Henslow on 21 July 1825'° and the two others by Babington himself on 10 June 1846 and 3 August 1855. Amendments (from “‘Scariola’ to “vyirosa’) to the label of the earlier of the two specimens and to the record from “Chalk-pit close Hinton” in the manuscript from which Babington prepared his 1860 Flora (bound in three large volumes kept in the Library of the Cambridge Department of Plant Sciences) suggest that Babington had some difficulty in distinguishing the larger Lactuca species. The only other nineteenth century specimen from Cambridgeshire in CGE was collected by James Backhouse (1825-1890) at “Elms n’. Wisbeach” (i.e. Elm, near Wisbech) in 1844."” Babington (1860) gave three new localities for L. serriola.'* He also included a further entry of “Burwell Pit!; Relh.” and a record from “Denny Abbey” based on Hemsted’s “specimen figured in Eng. Bot. [Smith & Sowerby 1795, no. 268]” (as well as repeating Relhan’s record of L. virosa there). Perring et al. (1964) suggested that in Babington’s time L. serriola was “as rare as L. virosa’”.’” The four nineteenth century specimens from Cambridgeshire in CGE are all of forma serriola.” TWENTIETH CENTURY AUTHORS Evans (1939) recorded L. virosa as “Very rare; probably extinct” in Cambridgeshire. After quoting a supposed record of Ray’s “from Ely”*', Relhan’s records and Babington’s (1860) record from “Hinton”, he added: “We have no later record.” He gave a number of localities for L. serriola, “generally close to towns and villages”. HISTORICAL RECORDS OF LACTUCA SERRIOLA AND VIROSA 155 er ine, © co Lee) as | fe) OI tas 2 3 a SS) 6 ih > FIGURE 3. Map of Cambridgeshire (v.c. 29) showing post-1986 tetrad records for Lactuca virosa. Most of the county is in TL (=52) but the northern part is in TF (=53). 156 P. H. OSWALD Perring, Sell & Walters (1964) repeated the error that John Ray recorded L. virosa in Cambridgeshire~ but they treated the species as extinct in the county, giving 1855 as the last certain date for it.” However, Crompton & Whitehouse (1983) recorded that “it was certainly near the gas works during the 1940—1950s, and it is still to be found on wasteland in the Newmarket Road area and along nearby railway lines”. Perring et al. regarded L. serriola as “now a frequent weed of roadsides and disturbed ground throughout the county” and suggested that a “rapid spread” had taken place “between 1930 and 1945”, adding: “In recent years it may have declined again.”°* They did not mention forma integrifolia, but one at least of the authors was well aware of its existence (P. D. Sell, pers. comm.), despite their misleading comment that “the more nearly simple leaves of L. virosa are usually sufficient to distinguish it”. DISCUSSION The confusion occasioned by the failure of most British Floras of the nineteenth century to recognise that L. serriola can have simple leaves, so well described by Carter & Prince (1982), has led to some uncertainties about the past status of this species and L. virosa in Cambridgeshire, as elsewhere in Britain; but, in its case, the belief that L. virosa was known to Ray has until now further clouded the picture. It is impossible to say with certainty when this species was in fact first seen in the county, but it is not improbable that in the nineteenth century it occurred only at Cherry Hinton (TL/48.56) and Elm (probably TF/46.06 or 47.06) and that it then disappeared until the 1940s. At any rate the first certain record for Cambridgeshire was made by the Rev. Leonard Jenyns at Chalkpit Close, Cherry Hinton, on 12 August 1824 (see note 16). Previously more characteristic of “naturally unstable habitats, such as sand-dunes and cliff-ledges” and “frequently found on chalk and oolitic limestone” (Prince & Carter 1977, p. 336), L. virosa is becoming more widespread in Britain in ruderal habitats like those long colonised by L. serriola.'> Since about the 1970s it has increased substantially in Cambridgeshire, especially on the margins of dual carriageways (e.g. TL/3.4, 4.4 and 6.6) and on waste ground in Cambridge (TL/4.5). Figure 3 shows post-1986 records for 16 out of the 30 hectads that include a substantial area in Cambridgeshire and 50 out of some 575 tetrads in the county. L. serriola has increased even more markedly both in southern Britain and in Cambridgeshire, though with an earlier start. Recording for Atlas 2000 shows its presence now in every hectad in and around Cambridgeshire. Since 1986 records have been made in 201 tetrads in Cambridgeshire, but these probably do not represent its complete distribution. What 1s particularly puzzling 1s why its forma integrifolia, although already present at the time of Johnson and Ray, was formerly apparently rarer than the nominate cut-leaved form, when now it is by far the commoner form. By the 1950s Kent & Lousley (1954, p. 176) wrote of it (as L. serriola L. var. dubia (Jord.) Rouy): “This is the aggressive entire-leaved form, possibly of adventive origin, which has spread so rapidly in the London Area in recent years. Plants with runcinate leaves occur occasionally.” In Cambridgeshire, P. D. Sell (pers. comm.) has suggested that forma serriola is native in the Fens and forma integrifolia is a relatively recent immigrant, and Beckett & Bull (1999, p. 201) apparently regard this as proven; though it is true that many localities for the former are in the Fens (including those for all four nineteenth century specimens in CGE”), this theory seems too simplistic, particularly since mixed populations often occur. Prince & Carter (1977, p. 334) suggested that, “‘although there are no absolute barriers to natural cross-pollination” between the two forms, “it is nevertheless an extremely rare occurrence”’. Perhaps there are environmental factors that favour the predominance of one form or the other in particular circumstances. Prince & Carter also wrote (p. 337): “Whereas the pinnatifid-leaved plant is the less common variant in Britain (having a particularly restricted south-eastern distribution), on the Continent it is the more abundant of the two.” My own observations suggest that this generalisation is too sweeping: in much of western France the position is similar to that in south- east England, and forma serriola certainly occurs in Wales. Could a more continental climate favour the predominance of forma serriola and a more oceanic one that of forma integrifolia? This might help to explain the former predominance of forma serriola in the Fens (with a more continental climate than most of Britain), but the present predominance of forma integrifolia in urban and roadside habitats is harder to account for: although the extremes of winter climate are buffered by the warming effect of housing and traffic, these habitats are hotter and drier in summer. HISTORICAL RECORDS OF LACTUCA SERRIOLA AND VIROSA 157 ACKNOWLEDGEMENTS My interest in Lactuca taxa was kindled by reading Prince & Carter (1977), when it suddenly became clear why I| had experienced such difficulty in identifying these plants as a schoolboy in the Thames Valley. In a nice piece of symmetry, their paper appeared in the issue of Watsonia dedicated to the memory of J. Edward Lousley, in which Duggie Kent described Rumex x lousleyi and provided a bibliography of Lousley’s published works. I am very grateful to the late Dr D. E. Coombe for encouraging my interest in Lactuca, to Dr S. M. Walters for supplying the quotation at the head of this paper, to Mrs G. Crompton for help in tracing Cambridgeshire records, to the staff of Cambridge University Library’s Rare Books and Manuscripts Rooms for their courteous service, to Dr D. Briggs. Mr P. D. Sell and Mrs G. Murrell for their kind assistance in Cambridge University Herbarium (CGE), to Mr R. Savage for similar help in the Library of the Department of Plant Sciences, to Dr C. D. Preston for reading a draft of this paper and making helpful comments, and to Mr G. M. S. Easy and Mr D. A. Wells for supplying many of the records for Figure 3. I also thank the Syndics of Cambridge University Library for the use of illustrations from their books for Figures | and 2 and Mrs J. Gaunt for scanning them and cleaning them up for printing. REFERENCES BABINGTON, C. C. (1860). Flora of Cambridgeshire. J. van Voorst, London. BECKETT, G. & BULL, A. (1999). A Flora of Norfolk. Gillian Beckett, Stanhoe, Norfolk. BAUHINUS, C. (1620, reissued 1671). Prodromus theatri botanici. Francofurti [Frankfurt]. BAUHINUS, C. (1623, reissued 1671). Pinax theatri botanici. Basileae [Basel]. BAUHINUS, J., CHERLER, J. H. & CHABRAEUS, D. (1651). Historia plantarum universalis. Tomus I. Ebroduni [ Yverdon]. CARTER, R. N. & PRINCE, S. D. (1982). A history of the taxonomic treatment of unlobed-leaved prickly lettuce, Lactuca serriola L., in Britain. Watsonia 14: 59-62. CLAPHAM, A. R. (1952). Lactuca L., in CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F., Flora of the British Isles, pp. 1125-1127. Cambridge University Press, Cambridge. CLARKE, W. A. (1900). First records of British flowering plants, 2nd ed. West, Newman & Co., London. CROMPTON, G. (1997). Botanizing in Cambridgeshire in the 1820s. Nature in Cambridgeshire 39: 59-73. CROMPTON, G. & WHITEHOUSE, H. L. K. (1983). A checklist of the flora of Cambridgeshire. Published by the authors, Cambridge. [DALECHAMPIUS, J.] (1587). Historia generalis plantarum. [Part 1]. G. Rovillius, Lugduni [Lyons]. EVANS, A. H. (1939). A Flora of Cambridgeshire. Gurney & Jackson, London & Edinburgh. EWEN, A. H. & PRIME, C. T., trans. & ed. (1975). Ray’s Flora of Cambridgeshire (Catalogus plantarum circa Cantabrigiam nascentium). Wheldon & Wesley Ltd, Hitchin. Fucustus, L. (1542). De historia stirpium commentarii insignes. Basileae [Basel]. GERARDE, J. (1597). The herball or generall historie of plantes. J. Norton, London. GiLMouR, J. S. L., ed. (1972). Thomas Johnson. Botanical journeys in Kent & Hampstead. The Hunt Botanical Library, Pittsburgh, Pennsylvania. [How, W.] (1650). Phytologia Britannica. Londini [London]. [JACSON, M. E.] (1797). Botanical dialogues, between Hortensia and her four children, ... designed for the use of schools, by a lady. J. Johnson, London. JOHNSON, T. [1629]. Iter plantarum investigationis ergo susceptum. Privately printed, [?London]. JOHNSON, T. (1632). Enumeratio plantarum in Ericeto Hampstediano locisque vicinis crescentium. T. Cotes, [London]. JOHNSON, T. (1633). Of wilde lettuce, in GERARDE, J., The herball or generall historie of plantes, 2nd ed., pp. 309-310. A. Islip, J. Norton & R. Whitakers, London. KENT, D. H. (1975). The historical Flora of Middlesex. The Ray Society, London. KENT, D. H. & LOUSLEY, J. E., comp. (1954). A hand list of the plants of the London Area. [Part 4]. Supplement to London Naturalist 33. LINNAEUS, C. (1753). Species plantarum. Holmiae [Stockholm]. LINNAEUS, C. (1756). Centuria II plantarum. Upsalae [Uppsala]. [MARTYN, J.] (1727). Methodus plantarum circa Cantabrigiam nascentium. Londini [London]. MARTYN, T. (1763a). Plantae Cantabrigienses. London. MARTYN, T. (1763b). Herbationes Cantabrigienses. London. MATTHIOLUS, P. A. (1558). Commentarii ... in libros sex ... Dioscoridis ... de medica materia. Venetiis [Venice]. 158 P. H. OSWALD MERRETT, C. (1666, reprinted 1667). Pinax rerum naturalium Britannicarum. Londini [London]. PENA, P. & DE LOBEL, M. (1571, reissued 1605). Stirpium adversaria nova. Londini [London]. PERRING, F. H., SELL, P. D. & WALTERS, S. M. (1964). A Flora of Cambridgeshire. Cambridge University Press, Cambridge. PRESTON, C. D. & HILL, M. O. (1997). The geographical relationships of British and Irish vascular plants. Botanical journal of the Linnean Society 124: 1-120. PRINCE, S. D. & CARTER, R. N. (1977). Prickly Lettuce (Lactuca serriola L.) in Britain. Watsonia 11: 331-338. [RAY, J.] (1660). Catalogus plantarum circa Cantabrigiam nascentium. Cantabrigiae [Cambridge] & Londini [London]. RAY, J. (1670). Catalogus plantarum Angliae. Londini [London]. RAY, J. (1677). Catalogus plantarum Angliae. Editio secunda. Londini [London]. RAY, J. (1686). Historia plantarum. Tomus I. Londini [London]. RAY, J. (1690). Synopsis methodica stirpium Britannicarum. Londini [London]. [RAY, J. & DENT, P.] (1685). Appendix ad catalogum plantarum circa Cantabrigiam nascentium: continens addenda & emendanda. Editio secunda, aucta plantis sexaginta. Cantabrigiae [Cambridge]. RELHAN, R. (1785). Flora Cantabrigiensis. Cantabrigiae [Cambridge]. RELHAN, R. (1802). Flora Cantabrigiensis. Editio altera. Cantabrigiae [Cambridge]. RELHAN, R. (1820). Flora Cantabrigiensis. Editio tertia. Cantabrigiae [Cambridge]. RICH, T. C. G. & JERMY, A. C. (1998). Plant crib 1998. Botanical Society of the British Isles, London. SMITH, J. E. & SOWERBY, J. (1795). English Botany. Vol. IV. London. STACE, C. (1991, 2nd ed. 1997). New Flora of the British Isles. Cambridge University Press, Cambridge. STACE, C. (1999). Field Flora of the British Isles. Cambridge University Press, Cambridge. TRAGUS, H. (1552). De stirpium ... usitatis nomenclaturis, propriisque differentiis, ... commentariorum libri tres. [Book 1]. Argentorati [Strasbourg]. (Translation from the original German into Latin by D. Kyber). FOOTNOTES 1 The second of the names that Johnson gave to the “second wild Lettuce” of 1632 was the name he had given to the plant he found on the Isle of Sheppey in 1629 and a shortened version of the long “title” from the Adversaria given here to the first sort (i.e. L. virosa). “Lactuca sylvestris laciniata minima nondum descripta. The least cut-leaved wild Lettuce. This was found on a bank and in a ditch by the side of a kind of drove or lane leading from London road to the river, just at the water near a quarter of a mile beyond the spittle-house end |“‘near the southern end of the present Coe fen” according to Ewen & Prime (1975, p. 78)].”” Ray later (1670, p. 37) realised that this plant had already been described by Gaspard Bauhin (Bauhinus 1620, p. 68) as “Chondrilla viscosa humilis”. 3._L. serriola forma serriola now occurs with forma integrifolia by the roundabout where A142 (Witchford Road) leaves A10 (the Ely bypass) (TL/527.794). 4 “Johnsonus apud Ger. duo facit genera lactuce sylvestris odore opii, que apud alios scriptores existimat confundi; & nomina autorum que nos secunde ejus speciei magis propria putamus attribuit prime; siquidem nostram que foliis est indivisis & 2 Ger. solam a botanographis descriptam arbitramur, non prout ille vult, primam Gerardi.” 5 “Johnson after Gerard makes two kinds of lactuca sylvestris by the odour of the opium (= milky juice) which he considers are confused by other writers; and he attributes the name given in his authors to the first which we consider more appropriate to the second species. We think that our kind, which has undivided leaves (i.e. the second of Gerard) is the only one described by botanical writers and not the first as he wishes. [*This separation is not accepted today.]” 6 Babington (1860, p. 141) further confused matters by attributing Ray’s locality for his first plant (i.e. L. serriola forma serriola) to this, his second plant, which Babington thought was L. virosa. 7 This is based on a sentence in Jean Bauhin’s account of L. serriola (Bauhinus et al. 1651, p. 1003): “Huius Lactucee quedam mihi species obseruata folio non laciniato etiam per caules. [I have observed a certain kind of this Lettuce with the leaf not laciniate even along the stems.]” On the next page Bauhin states that at Lyons he had observed this species “with its earlier, fairly long leaves [in February and March] not laciniate” (“Ibidem obseruaui foliis non laciniatis prioribus longiusculis.”’). 8 “Lactuca sylvestris 2. Ger. emac | Dele synonyma omnia. Horum loco adde Lactuce sylvestris sive Endiviz multis dicte folio laciniato dorso spinoso varietas J.B. Hec non alia in re differt a Lactuca sylv. costa spinosa C.B. quam foliis que ei ne in caule quidem laciniata sunt. Que de hac planta in Catalogo tradidimus jam retractamus & in Johnsoni sententiam concedimus.” (Ray & Dent 1685, pp. 22-23). Unlike Johnson (1633), Ray evidently did not regard “Endivia” and “Thesion Dalechampii” as reliable synonyms for “Lactuca sylvestris 2.”. 9 In fact they are often biennial (see INTRODUCTION). i) 10 ja] 12 19 20 21 22 Mee 24 25 HISTORICAL RECORDS OF LACTUCA SERRIOLA AND VIROSA 159 This is not actually stated by Johnson (1633) but is implied by the order in which he presented the taxa and by his statement that his last taxon “hath not altogether so strong a sent of Opium as the two former”. “Hance speciem Johnsonus Gerardi emaculator prime speciei varietatem facit; J. Bauhinus secunde, cui & nos suffragamur; siquidem illa humilior est, & foliis minoribus, odore etiam mints gravi, nostris saltem naribus: Non alia autem in re a secunda differt quam foliorum figura, que huic ne in caule quidem laciniata sunt. Iisdem cum priore locis provenit.” (Tomus 1, Lib. V, Cap. I: p. 222). “Cum priore, cujus varietas est, sed raritis.”” Nowadays it is by far the commoner form in Britain (see DISCUSSION). Prince & Carter (1977, p. 337) are wrong in saying that Ray (1690) “described them as altogether different plants”. This should be L. serriola forma integrifolia, since Linnaeus classified this taxon as var. B of L. virosa in 1753; however, in 1756, when Linnaeus first described L. serriola, he made forma integrifolia its var. Y (Carter & Prince 1982). Sowerby’s engraving shows a plant with incised lower leaves but dentate upper cauline leaves and so may have been drawn from specimens of both forma serriola and forma integrifolia. Clapham (1952) described L. virosa as a “sub-Mediterranean species, associated in C. Europe with... thermophilous plants’, but L. serriola as “associated with steppe species”. Preston & Hill (1997, pp. 38-39, 102, 104) classified L. virosa as “Suboceanic Southern-temperate” and L. serriola as “Eurosiberian Southern-temperate” but widely naturalised. The earliest record from this locality appears to be that of the Rev. Leonard Jenyns in his journal for 12 August 1824 (Crompton 1997): “Walked this morning with Profess. Henslow to Hinton nr Cambridge. .... In Chalkpit Close — ... The great lettuce — Lactuca virosa, .... Of the Lettuce we could not find more than one specimen though we made the most diligent search.” The sheet bears a printed label “EX HERB. JAMES BACKHOUSE, JUN.” and a later stamp “Ex Herb. S. H. Bickham” (1841-1933), and the specimen was identified as “Lactuca scariola” and “Lactuca Scariola” (amended in pencil in the hand of S. M. Walters to “L. virosa’). Despite mentioning this specimen, Perring et al. (1964) did not include TF/4.0 among the grid squares for L. virosa; nor was this square listed by Crompton & Whitehouse (1983). Recently Peter Sell has determined this specimen as “forma lactucarii (Lamotte) P. D. Sell” (see INTRODUCTION). He has determined Henslow’s specimen from Cherry Hinton as forma virosa, but both of Babington’s as forma lactucarii. “Shuckburgh Castle, Newmarket Heath’, on his own authority, “Grunty Fen’, attributed to William Marshall, and “Ely”, attributed to Henslow. An additional record, “Burrell’s Walk, Cambridge’, is attributed to Relhan. After giving Hemsted’s locality, J. E. Smith wrote: “It grows among rubbish and on waste ground in other parts of that country, but rarely elsewhere, ....” (Smith & Sowerby 1795, p. 268). Leonard Jenyns clearly regarded L. serriola as a rarity; in his journal for 1 August 1827 he wrote: “... today made an expedition to Grunty Fen. .... In my return — observed, & gathered with much satisfaction for the first time in my life, the prickly Lettuce (Lactuca scariola) which grew somewhat sparingly on a high ditch- bank edging the fen very near its entrance from Ely thro’ the lanes.” Jenyns’ ‘Journal of Natural History’, covering the years 1823-1846, is now among the Alfred Newton papers in Cambridge University Library. This entry was not included by Crompton (1997). They are labelled “Streatham Cambridgeshire” (C. M. Lemann, 22.07.1831), “edge of Grunty-fen Cambridgeshire” (J. S. Henslow, 27.08.1833), “Ely, Cambs” (J. S. Henslow, 27.08.1833), perhaps from the same site, and “Roswell Pits, Ely” (J. E. Little, August 1878). The date of “the oldest specimen in CGE” given by Perring ef al. (1964) — “from Ely in 1853” — must be a mistake for 1833; but the Stretham specimen is older. The earliest Cambridgeshire specimen of forma integrifolia in CGE was collected at Barnwell in September 1937 by H. Gilbert Carter and identified by him as “Lactuca Scariola’’. Evans was here maintaining an error of Babington’s; see note 6 above. The “pre-1930” and “not since 1949” records for grid square TL/5.7, in Perring et al. (1964) and in Crompton & Whitehouse (1983) respectively, suggest that these authors were also perpetuating Babington’s incorrect application of Ray’s (1660) locality near Ely to this species. This date was presumably based on Babington’s specimen in CGE labelled “Chalk-pit, Cherry Hinton, Cambridgeshire.” Even if this was ever true, today it is frequent in Cambridgeshire. Crompton & Whitehouse (1983) wrote: “Many new records yet it is reported to be declining.” Not only does forma integrifolia of L. serriola have nearly simple leaves, but var. lactucarii of L. virosa has pinnatisect cauline leaves (see INTRODUCTION). (Accepted December 1999) 7? cil Watsonia 23: 161—165 (2000) 161 Calystegia (Convolvulaceae) hybrids in West Wales R. K. BRUMMITT The Herbarium, Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AE and A. O. CHATER Windover, Penyrangor, Aberystwyth, Ceredigion SY23 1BJ ABSTRACT Intensive studies of populations of Calystegia (Convolvulaceae) in Cardiganshire, v.c. 46, and Merioneth, v.c. 48, have recognised C. sepium subsp. sepium and subsp. roseata as native taxa there while C. silvatica subsp. silvatica and subsp. disjuncta and C. pulchra are introduced and naturalised. Numerous probable hybrids between these taxa are postulated, some as single populations and others apparently having spread locally probably by vegetative means. Even back-crosses between adjacent populations may be postulated. In particular, one putative hybrid between C. pulchra and C. sepium was recognised over an area of several kilometres in Cardiganshire, resembling C. pulchra in general appearance but differing in having a paler pink corolla and less markedly square leaf sinus. KEYWORDS: Calystegia, Convolvulaceae, hybridisation, Wales. INTRODUCTION Hybrids among the native and naturalised species of Calystegia in the British Isles have long been known (Stace 1961, 1975, 1997, Brummitt 1963, 1980, 1998). The whole genus, in which some 25 species world-wide may conveniently be recognised, is taxonomically difficult, and few if any of the species are morphologically clear-cut. They mostly vary considerably over their ranges and merge geographically one into another, and division into species and subspecies is of necessity somewhat arbitrary. In the British Isles, apart from C. soldanella (L.) R.Br. which is not considered in this note, there are one native and two naturalised species. The native C. sepium (L.) R. Br. is represented by the widespread typical subsp. sepium which is white-flowered (rarely pink-flowered, f. colorata (Lange) Dorfl.) and glabrous, and the predominantly western coastal subsp. roseata Brummitt which is pink-flowered and usually hairy at the shoot apices and leaf sinuses. In addition, one further subspecies, subsp. spectabilis Brummitt, has been recorded once as an introduction. The commoner of the two naturalised species, C. silvatica (Kit.) Griseb., is represented here by the typical subsp. si/vatica from the east Mediterranean, with broad, saccate bracteoles rounded or emarginate at the apex, and subsp. disjuncta Brummitt from the west Mediterranean, with narrower, less saccate bracteoles, obtusely pointed at the apex (Brummitt 1996, 1998). Rather less common, though widely distributed, is C. pulchra Brummitt & Heywood, of unknown origin, with large pink flowers, hairy stems, petioles and pedicels, and often winged peduncles. The following notes are based on fieldwork and specimens collected by both of us in two periods of several days in Cardiganshire, v.c. 46, and Merioneth, v.c. 48, in July 1996 and August 1998, and on various collections by A.O.C. in the former county over several years. Nearly all the recorded British taxa occur in this area, and much of the variation we found seemed likely to be the result of hybridisation. Although in the absence of experimental evidence we are unable to confirm most of our interpretations, we feel that it is worth putting our observations on record in the hope that they may be a stimulus to comparable observations being made in other parts of the country so that a more complete picture of the state of the genus in Britain can be built up. Details of the relevant specimens are given. Collections of all taxa made by us are deposited at NMW or K or both. 162 R. K. BRUMMITT AND A. O CHATER FIELD OBSERVATIONS C. sepium subsp. sepium was widespread in the area, and the chief variation noted was that in several coastal localities plants occurred with distinctly hairy stem apices. This is unusual for this taxon, and the possibility that the hairiness is due to introgression from subsp. roseata must be borne in mind. Subsp. roseata occurred in several coastal and estuarine sites, appearing completely native in reedbeds in both the Dyfi and Mawddach estuaries. It also occurred in coastal hedges, and at one site 18 km inland in Salix carr by the disused railway at Cors Caron where it may have been an accidental introduction. A specimen collected in 1961 by P. M. Benoit from the railings of a waste garden at Bron Fegla near Arthog in Merioneth, now at LIV, was identified by R.K.B. as subsp. spectabilis, a taxon widely naturalised in Scandinavia, and is the basis for the inclusion of this subspecies in Stace (1997). The pink flowers and coastal habitat would fit well with that of subsp. roseata, but the large size of the flowers and the broad leaf sinus suggest that it is more likely to have been a garden introduction. Searches for the plant in 1998 by both Benoit and ourselves were unsuccessful, and we assume that it is now extinct at the locality. Among the populations of C. silvatica examined, 30 were considered to be clearly subsp. disjuncta, 23 subsp. silvatica and 5 were intermediate. 13 plants were assumed to be hybrids between C. sepium and C. silvatica. The name C. x lucana (Ten.) G. Don is available for this interspecific cross, but since there is a wide range of morphological forms resulting from the cross, R.K.B. prefers not to apply a binomial (see Brummitt 1998 and further notes below). The putative hybrid Cardiganshire populations varied considerably in morphology of the bracteoles, and in a few cases this was extreme enough to suggest whether it was subsp. si/vatica or subsp. disjuncta that was involved. A preponderance of hybrids among collections made late in the season suggests that they may continue flowering later into the autumn than the parents. C. pulchra appeared fairly uniform in West Wales, with deep pink flowers, notably square leaf sinuses and large rounded bracteoles. However, in the lower Teifi valley, in an area bounded by Cardigan, Penpare and Cenarth, and particularly conspicuous at intervals along the A484 road, we found plants (some of which had been previously recorded as C. pulchra) with paler pink flowers, more rounded leaf sinuses and somewhat irregularly shaped bracteoles. Morphologically they appeared to be intermediate between C. pulchra and C. sepium subsp. sepium, which was frequent in the same area, and we assumed that they were a hybrid between these two taxa. The name C. x scanica Brummitt is available for this interspecific cross, but the plant to which this name was originally applied, naturalised in several places in southern Sweden, is very different, having white flowers, bracteoles less strongly overlapping, and large leaves with a very characteristic shape. The differences between these nothomorphs is at least as great as the difference between the parent species, and again R.K.B. prefers not to refer to “the hybrid” with a binomial but regards these Cardiganshire populations as “a hybrid” presumably between C. pulchra and C. sepium. At one roadside hedge and scrub site near Llandygwydd where both this hybrid (Brummitt 20058 & Chater) and C. sepium subsp. sepium (Brummitt 20057A & Chater) occurred, we found a white- flowered plant similar to C. silvatica subsp. disjuncta, but with very pointed bracteoles and narrowly pointed leaves (Brummitt 20057B & Chater); morphologically this looked very likely to be a back cross between the putative C. pulchra/sepium hybrid and C. sepium subsp. sepium. Near Aber-porth we found a single plant or small population in a roadside hedge (Brummitt 19513, 20054 & Chater) rather like Brummitt 20057B & Chater. The corolla was white or very slightly flushed pink inside, and with pink mid-petaline bands outside like those often found on C. silvatica subsp. disjuncta, with the dull green, matt leaves of C. pulchra and some stem apices glabrous and others hairy. It too may have been a back cross between the C. pulchra/sepium hybrid and C. sepium subsp. sepium, or possibly a hairy form of C. silvatica subsp. disjuncta showing some introgression from C. pulchra. Near Newcastle Emlyn and at Llandysul, 8 and 15 km further up the Teifi valley respectively, plants with white flowers and grotesquely misshapen bracteoles occurred (Brummitt 20064A, 20064B, 20065 & Chater) which appeared to be a white-flowered nothomorph also resulting from hybridisation between C. pulchra and C. sepium. At Llandysul these plants grew in close association with the putative parents. Further north in Cardiganshire, in a roadside hedge at Ffos-y- ffin, was a mixed population of C. pulchra (Brummitt 20066B & Chater) and white-flowered plants. Some of the latter were glabrous and some had pubescent stem apices, while the bracteoles CALYSTEGIA HYBRIDS IN WEST WALES 163 were very variable though not grotesquely enlarged and distorted as in the Llandysul and Newcastle Emlyn plants. They appeared to us to be a mixture of C. sepium subsp. sepium (Brummitt 20067A & Chater) and a white-flowered hybrid between C. pulchra and C. sepium (Brummitt 20067B & Chater). Close to Morfa Mawddach Station, near Arthog in Merioneth, we noted three populations in subcoastal habitats typical of where C. sepium subsp. roseata grows. One was white-flowered and was Clearly introduced C. silvatica subsp. disjuncta (Brummitt 20051 & Chater). The others were pink-flowered, one (20050A) growing among brambles together with the white flowered 20051, and the other 50 m away climbing up Phragmites (20050B). The latter in particular was in typical habitat of subsp. roseata and looked very like it, but our suspicions were raised by the unusually broad bracteoles and flowers rather large for C. sepium. There is a possibility that both the pink- flowered plants were hybrids of C. sepium subsp. roseata with C. silvatica subsp. disjuncta. On the other hand, the leaf sinus is narrowly V-shaped, and rather large bracteoles like these may be found in subsp. roseata more often than in subsp. sepium, and hybridisation is by no means proven. In the genus as a whole, abnormal bracteoles appear often to be an indication of hybridity, and perhaps a quarter of all putative hybrids have the bracteoles of some flowers variously unequal in size or shape, more or less leaflike, not opposite, or puckered or distorted. DISCUSSION The pattern emerging from our fairly intensive survey of the genus in one area of Wales probably reflects the situation over much of northern Europe. Different taxa have been introduced, probably originally as garden plants, and have become established in the wild through their vigorous vegetative reproduction. They have then hybridised with native taxa and other introduced taxa. In some cases these hybrids may be restricted to single populations, but in a few cases they appear to have spread - probably by vegetative means, although little is really known about how the rhizomes are moved from one locality to another - and one clone may be recognised in numerous populations over one area. The putative hybrid between C. pulchra and C. sepium which we saw frequently in different colonies over an area of several kilometres in Cardiganshire is probably as morphologically distinct and almost as widely distributed as the putative hybrid between C. silvatica and C. pulchra which is well established at different localities in eastern Nottinghamshire and adjacent Lincolnshire and which was named as C. howittiorum by Brummitt (1980). There should probably be a limit, however, to the number of such clones which are given formal names, and we decline to describe this Cardiganshire hybrid as a new taxon. The uncertainty about the actual parentage of such plants is an additional reason for not applying formal binomial nomenclature to putative hybrids, at least until much more is known about their origins and reproductive behaviour. SELECTED HERBARIUM SPECIMENS (ALL DET. R. K. BRUMMITT). C. SEPIUM SUBSP. SEPIUM x C. SILVATICA v.c. 46, Cards., among Phragmites at N end of boardwalk just E of A467 bypass, Teifi Marshes Reserve, Cardigan, altitude 4 m (SN/182.458), 22 September 1995, A. O. Chater (NMW). v.c. 46, Cards., among Fallopia japonica on waste ground, NW corner of Glanyrafon Industrial Estate, Llanbadarn Fawr, 7 m (SN/608.803), 7 July 1995, A. O. Chater (NMW). C. sepium subsp. sepium x C. silvatica subsp. silvatica. v.c. 46, Cards., upper part of salt marsh 100 m S of Isolation Flats, Ystwyth estuary, 4 m (SN/580.805), 7 September 1995, A. O. Chater (NMW). v.c. 46, Cards., abundant on shingle beach opposite mouth of Aberystwyth harbour, 3 m (SN/580.803), 11 September 1995, A. O. Chater (LTR, NMW). 22 July 1996, R. K. Brummitt 19505 & A. O. Chater (K). C. sepium subsp. sepium x C. silvatica subsp. disjuncta. v.c. 46, Cards., scrub by road bridge E of Rhoscellan-fawr, 1-5 km N of Llangorwen, 70 m (SN/599.854), 12 September 1995, A. O. Chater (NMW). v.c. 46, Cards., N hedge of A487 at Penffynnon, Penparc, 135 m (SN/217.481), 13 September 1995, A.O. Chater (NMW). 164 R. K. BRUMMITT AND A. O CHATER v.c. 46, Cards., NW hedge of road 500 m WSW of Brynllynan,Y Ferwig, 95 m (SN/185.484), 13 September 1995, A. O. Chater (NMW). v.c. 46, Cards., forming dense growth over rank grass by road and ditch adjacent to refuse tip, 300 m S of Borth church, 4 m (SN/611.894), 25 July 1996, A. O. Chater (NMW); and ibid. 12 August 1998, R. K. Brummitt 20049B & C & A. O. Chater (K, NMW). v.c. 46, Cards., abundant by track in conifer plantation, Long Wood, 1-4 km S of Llangybi, 210 m (SN/610.515), 1 August 1996, A. O. Chater (NMW). C. sepium subsp. sepium x C. silvatica subsp. disjuncta. v.c. 46, Cards., in scrub on sandy soil 300 m from coast, 100 m W of Ty-gwyn, Mwnt, 30 m (SN/197.518), 15 August 1996, A. O. Chater (NMW). v.c. 46, Cards., roadside hedgebank 600 m S of Temple Bar crossroads, 170 m (SN/533.537), 21 August 1996, A. O. Chater (NMW). v.c. 46, Cards., in hedge and rank marsh 100 m NW of Pentre, Llundain-fach, Abermeurig, 75 m (SN/559.567), 8 September 1996, A. O. Chater (NMW). v.c. 46, Cards., scrub at edge of wood 200 m S of Aberaeron church, 35 m (SN/456.626), 2 July 1997, A. O. Chater (NMW). v.c. 46, Cards., hedge on N side of garden, Wallog house, 10 m (SN/590.857), 3 October 1998, A. O. Chater (KK). POSSIBLY C. SEPIUM SUBSP. ROSEATA x C. SILVATICA SUBSP. DISJUNCTA v.c. 48, Merioneth., among brambles 150 m SE of Morfa Mawddach Station, Arthog, 5m (SN/6297.1401), 12 August 1998, R.K.Brummitt 20050A & A.O. Chater (K, NMW). v.c. 48, Merioneth.,On Phragmites 100m SSE of Morfa Mawddach Station, 5m (SN/6290.1405), 12 August 1998, R.K. Brummitt 20050B & A.O. Chater (K, NMW). C. PULCHRA x SEPIUM, PALE PINK NOTHOMORPH v.c. 46, Cards., hedge and scrub on N side of B4570/A484 road junction, 800 m E of Cardigan church, 5 m (SN/189.460),13 August 1998, R. K. Brummitt 20055 & A. O. Chater (K, NMW). v.c. 46, Cards., hedge on S side of A484 road by Llechryd waterworks, 10 m (SN/230.435), 13 August 1998, R. K. Brummitt 20056 & A. O. Chater (K, NMW),. v.c. 46, Cards., scrub on S side of A484 road 800 m SE of Llwynduris, 12 m (SN/244.427), 13 August 1998, R. K. Brummitt 20058 & A. O. Chater (K, NMW). v.c. 46, Cards., S hedge of A487 road 150 m WSW of crossroads in Penparc village, 130 m (SN/211.479), 19 August 1998, A. O. Chater (K, NMW). C. PULCHRA x SEPIUM, WHITE NOTHOMORPHS v.c. 46, Cards., hedgebank on SW side of road T-junction 600 m SW of Aber-porth church, 111 m (SN/250.507), 23 July 1996, R. K. Brummitt 19513 & A. O. Chater (K, NMW). 13 August 1998, R. K. Brummitt 20053 & A. O. Chater (K, NMW). Possibly a back cross between C. pulchra x sepium and C. sepium subsp. sepium. v.c. 46, Cards., scrub by lay-by at bottom of wooded slope on N side of A484 road 800 m SE of Llwynduris, 12 m (SN/244.427), 13 August 1998, R. K. Brummitt 20057B & A. O. Chater (K, NMW). Probably a back cross between C. pulchra x sepium and C. sepium subsp. sepium. v.c. 46, Cards., scrub at NE corner of Rugby Club car park, 500 m E of Newcastle Emlyn bridge, 25 m (SN/314.409), 13 August 1998, R. K. Brummitt 20061 & A. O. Chater (K, NMW). v.c. 46, Cards., rank vegetation and wall between S part of Llandysul churchyard and Teifi river, 60 m (SN/419.406), 13 July 1998, R. K. Brummitt 20064B & A. O. Chater (K, NMW). v.c. 46, Cards., scrub by gardens 150 m N of Llandysul church, 70 m (SN/418.408), 13 August 1998, R. K. Brummitt 20065 & A. O. Chater (K, NMW). v.c. 46, Cards., W hedge of A487 road 350 m SSE of Henfynyw church, Ffos-y-ffin, 115 m (SN/449.609), 13 August 1998, R. K. Brummitt 20067B & A. O. Chater (K, NMW). REFERENCES BRUMMITT, R. K. (1963). A taxonomic revision of the genus Calystegia. Ph. D. thesis, University of Liverpool. BRUMMITT, R. K. (1980). Further new names in the genus Calystegia (Convolvulaceae). Kew Bull. 35: 327— 334. CALYSTEGIA HYBRIDS IN WEST WALES 165 BRUMMITT, R. K. (1996). Two subspecies of Calystegia silvatica (Kit.) Griseb. (Convolvulaceae) in the Mediterranean region. Lagascalia 18: 338-340. BRUMMITT, R. K. (1998). Calystegia. In RICH, T. C. G. & JERMY, A. C., eds. Plant crib 1998. Botanical Society of the British Isles, London. STACE, C. A. (1961). Some studies in Calystegia: compatibility and hybridisation in C. sepium and C. silvatica. Watsonia 5: 88-105. STACE, C. A. (1975). Hybridisation and the flora of the British Isles. Academic Press, London etc. STACE, C. A. (1997). New Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. (Accepted October 1999) ate é Watsonia 23: 167—172 (2000) 167 Ludwigia x kentiana E. J. Clement: a new hybrid aquatic E. J. CLEMENT 54 Anglesey Road, Alverstoke, Gosport, Hampshire, PO12 2EQ ABSTRACT A hybrid Ludwigia L. (Onagraceae) found wild in England is described and illustrated: its morphology suggests a parentage of L. palustris (L.) Elliott x L. repens J. R. Forst., and it probably equates to a plant known to aquaria enthusiasts as L. x mullertii hort. KEYWORDS: hybridisation, vegetative spread, water weed, Onagraceae, Hampshire-purslane. INTRODUCTION In September 1995 U. Sutcliffe & Ms A. Sutcliffe found in a round pond on Putney Heath (GR TQ/237.733), Surrey, v.c. 17, England, a waterweed that was identified as Ludwigia palustris. Subsequently when Mrs E. Norman revisited the locality, material was sent to me for confirmation and I realised that it differed appreciably from our native species. It was noticeably large and vigorous, and was probably equatable to a plant known by aquarists as L. x mullertii hort. (with orthographic variants ranging from L. x mulerttii to L. X muellertii), with a probable parentage of L. palustris (L.) Elliott x L. repens J. R. Forst. A request for further information contained in a preliminary report (Clement 1997) yielded nothing. I have still not found a botanical description of this taxon nor a validation of the name, nor any other matching taxon, hence my decision to publish herein a new and unambiguous epithet. This hybrid has not yet been definitely confirmed from a second locality, but further records are expected. Any vigorous Ludwigia colony should be sampled after permission to collect has been obtained. The lack of preservation of good, herbarium vouchers has hampered this research paper and it is becoming a recurrent problem in alien plant studies in Britain. The conservation movement has discouraged the collection of all natural history material: photographs, alone, being acceptable - hence there is a dramatic decrease in the collection of herbarium material. THE PROBABLE PARENT SPECIES The taxonomy of the two probable parents presents no problems, but their nomenclature calls for some explanation, as considerable confusion exists in the literature. L. palustris (L.) Elliott (published in 1817) Usnardia palustris L.) has, fortunately, never been a serious nomenclatural or taxonomic problem. At least two varieties have been recognised - e.g. Munz (1961) - but none of them seems worthy of higher status. The statement therein (Munz op. cit. p. 215) that in contradistinction to the American forms “the green bands of the hypanthium terminate well below the summit’in the European form seems untrue (see Fig. 1). The word “hypanthium” has, incidentally, different definitions in different books: Stace (1997) says that the hypanthium is absent in this genus, but his use of the term “hypanthium” (of an epigynous flower) apparently refers solely to that part of the extension of the receptacle beyond the summit of the ovary. (His glossary fails to make this fact clear). L. palustris is a widespread native in the northern hemisphere, occurring in W., C. and S. Europe (north as far as New Forest, England), N. Africa, W. Asia, N. and C. America and W. Indies. It is naturalised in southern Africa, S. America, the Pacific area, Hawaii, S.E. Australia and New Zealand. Illustrations may be found in Fig. 1, and in many British and foreign Floras. L. repens J. R. Forst. (Isnardia repens (Sw.) DC.) presents more problems. It was published in Forster (1771) with a query (?) before the specific name - and some authors appear to have 168 E. J/CLEMEN® rejected the epithet as being “not accepted by the author” (and hence an invalid name), but Article 34.1 in Greuter et al. (1994) makes it very clear that this is not so. The description seems adequate to define the taxon, the question mark indicating merely that Forster was unsure whether his plant was a new species or not. I have not searched for a type specimen: the statement by Stafleu & Cowan (1976) “it is not possible to state where the original Forster herbarium (father and son) is preserved” acted as a deterrent. The entry in Jackson (1893) has also mislead many of the unwary: it falsely equates this species to L. palustris as well as misquoting the page number in Forster’s work: it should be p. 6 (not 22). This is especially annoying, since this book contains no index of page numbers even for generic names! The homonym L. repens Sw. (published in 1797) is synonymous, whereas Jussiaea repens L. (1753) is only distantly related (see synonyms below after the key). Finally, the name L. natans Elliott (published in 1821) does indeed refer to this plant, and is unambiguous, but is, alas, a later synonym that should not be used, in spite of the recent choice of it by e.g. Garve & Meijden (1997) and Clement (1997). The species is restricted to S. and S.E. United States, Mexico and the West Indies. It is naturalised in S. Asia and Japan, also recently in Europe e.g. in Spain (Nieto Feliner 1997). Illustrations may be found in, e.g. Britton (1918), Fawcett & Rendle (1926) and Radford et al. (1968). DESCRIPTION OF THE HYBRID To validate the new epithet proposed herein, a formal diagnosis is required: Ludwigia x kentiana E. J. Clement, hybr. nov. Hybrida probabiliter inter Ludwigiam palustrem (L.) Elliott et L. repentem J. R. Forst., ab eis speciebus fructibus sterilibus, petalis circa 0-5 mm longis differt. HOLOTYPUS: England, Surrey (v.c. 17), Putney Heath, Round Pond, 6 October 1997. Mrs E. Norman, s.n. (BM); isotypus in Herb. EJC. A glabrous perennial with stems 20-80 cm, prostrate or weakly ascending, often largely or entirely submerged in water, abundantly rooting at lower nodes. Leaves 20-50 x 5-25 mm, opposite, the blade narrowly rhombic-obovate, gradually narrowed to a long petiole, with the margin entire. Flowers solitary in leaf axils, + sessile. Bracteoles linear, c. | mm. Calyx-tube + cylindrical, 4-sided, uniformly pale green; calyx teeth 4, c. 2 mm, deltoid. Petals 4, minute (c. 0-5 mm) cream, + spathulate, fugacious (usually dropping off as soon as the buds open); stamens 4, 0-8 mm long; style 0-8 mm long with subglobose stigma divided into 4 indistinct lobes. Fruit a capsule, partially developing and then dropping off, the ovules not expanding. The epithet kentiana is for D. H. Kent in appreciation of his great assistance to me for over 30 years during my studies on the adventive flora of Britain. The aquarium trade may be disappointed that I have not upheld the name that they currently use. The epithet that I have chosen is intended to cover all variations of the hybrid, some undoubtedly not worthy of any aquarium tank. The best clone (preferably the original, correct one!) can be more accurately named as L. x kentiana ‘Mullertii’, which may be legitimately abbreviated to L. “Mullertii’ - the simple quotes being the current convention to indicate a cultivar (i.e. equal to cv. Muellertii in older literature). At present, I refrain from allocating the clone described here as true L. ‘Mullertii’. An English name is also called for - Kent’s Hampshire-Purslane seems appropriate. TABLE 1. COMPARISON OF LUDWIGIA PALUSTRIS, L x KENTIANA AND L. REPENS L. palustris L. x kentiana L. repens Leaf blade Widest near its middle Widest in uppermost third Widest close to its apex Bracteoles 0-0-5 mm, free c. 1mm 1—2 mm, shortly adnate to ovary Sepal lobes — About as long as broad Longer than broad Much longer than broad Petals Absent 0-5 mm, cream 3-5 mm, yellow Fruit Subrotund and persistent; +Cylindrical and caducous; — Cylindrical and persistent; with 4 dark green bands of | uniformly pale green uniformly greenish yellow LUDWIGIA x KENTIANA 169 = Imm o = © 2 = © 3) vn FIGURE 1. Ludwigia x kentiana (A-G) and L. palustris (a—g) del. D. J. P. Smith © 1998. A,a. habit of plant; B,b. leaf from middle of stem; C,c. side view of immature fruit with persistent calyx teeth; D,d. transverse section of immature fruit showing ovules; E,e. view of flower from above; F,f. stamen; G,g stigma and style. 170 E. J“ CLEMENT L. x kentiana can be distinguished from its probable parents by the characters shown in Table 1. Several North American Floras e.g. Fernald (1987) claim that L. repens has no petals - the virtually invariable state in herbaria since the petals drop so quickly - but there seems to be no truth in this statement (Godfrey & Wooten 1981). I notice that Radford er al. (1968) claim “petals absent” in their description, but their wiser key says “petals present”. Clement (1997) claimed (1—) 4 petals for the hybrid, but a more careful study of fresh specimens revealed an invariable number of just 4. The illustration by D. J. P. Smith (Fig. 1) clearly shows these differences between L. palustris and L. x kentiana; in addition other minutiae may be noted, though less easily put into words, including the shape of the anthers. The provenance of the plants drawn are, respectively, near Hatchet Pond, New Forest, (v.c. 11, S. Hants) and Putney Heath. D. J. P. Smith also drew my attention to the fact that two of the opposing anthers, in both taxa, are noticeably larger than the other two - two of them may be better described as staminodes, a fact that I have not seen mentioned elsewhere. KEY TO LUDWIGIA SPECIES OCCURRING IN EUROPE The following key, partly based on Raven (1963), is provided to alert readers to the possibility of other species occurring in Britain in the future. All are typically perennial aquatics, although annuals, shrubs, and even a tree occur within the genus. Only L. palustris is native in Europe and it shows a markedly southern distribution. l. Leaves opposite; flowers + sessile; petals 4, 0-4 mim........2......cossncseateneeatee serene eet: «a8 2. I. Leaves alternate; flowers pedicelled; petals 5(—6), 7—30 mmm ....:5. cc.steecncee ee ee + 2. POtals aS Gti assis casas dcactevscccdistheacchcccaduidcasadennssigudelecealuatee een epee. eee eee L. palustris Dis Petals present; DUt LUSACIOUS 2..isscs.vecceeosenessnncectsnennsbonsicncedael copeagetecBicce sete eee eeeeeet ee eee: 2 3: Petals Cy 0:5 dy CHEB IN 5a0u5iedeaasspnwsvnvaaessvauewatinntende chaveeiees oReece cutee caer eee L. x kentiana Zi Petals: 35 titi; VOUOW sccscssseccapcassabssonncswcesuedaseiea doesaes Uegmodouinsndessec eee tea emene tment L. repens 4. Petals white or pale yellow, with a dark yellow claw ..............cccseescecceeeeeeeeeeees L. adscendens 4. Petals bright yellow throughout .....:::scccccssesssslececeseseachiesaseeeseccnasasoarteeteaneae seemeeeeeenen eee meee ese 5 De Calyx-tube finely hairy to glabrous; petals 7—16 mm ...............sss,ccsssa.deteneeeereres L. peploides oj Calyx-tube WitsUte icsccsaccccssecsinsasxcaacueavenxassiadandedusiudaccatineccisvsteeesgd eee ee eee 6 6. Stem_and leaves pubescent; petals 15—24 mom ....2.c.006.....46.0s0sedected tee eee L. grandiflora 6. Stem and leaves = glabrous; petals 7—15 Mh cssccrcesnccssas iced L. hexapetala The four additional species not discussed above are: Ludwigia adscendens (L.) Hara (Jussiaea repens L.) S. and S.E. Asia and tropical Australia. (a casual? in The Netherlands). Illustrations may be found in, e.g. Soerjani (1987). L. peploides (Kunth) Raven Tropical and subtropical N. and S. America, and perhaps native in Australia. Naturalised in S.W. France and New Zealand. A variable plant, sometimes divided into four subspecies. Illustrations may be found in, e.g. Godfrey & Waoten (1981) and Hoch (1993). L. grandiflora (Michx) W. Greuter & Burdet (L. uruguayensis (Camb.) Hara; Jussiaea repens sensu Coste, non L.) Pantropical and subtropical. Naturalised in S. France, Spain, The Netherlands and Belgium. Illustrations may be found in, e.g. Coste (1903) and Godfrey & Wooten (1981). L. hexapetala (Hook. & Arn.) Zardini, Gu & Raven Tropical America. Records exist for S. France and E. Spain, but they are probably referable to the closely allied L. grandiflora. Some authors, such as Hoch (1993), provide an illustration but clearly do not separate these two species even as varieties. Other potential escapees are also grown in aquaria - e.g. L. arcuata Walter x L. repens (Kasselmann 1985). The first parent has flowers on long pedicels 15-35 mm long, and this influence would presumably be evident in any hybrid that it forms. LUDWIGIA X KENTIANA 171 DISTRIBUTION OF THE HYBRID The hybrid L. x kentiana is probably widespread in aquaria in Britain and elsewhere, but no previous record of its escape into a wild location has come to my notice: it may well have arisen in an aquarium in Europe. Indeed, since both parents appear to be invariably self-pollinating in the wild, with the anthers and stigmas in juxtaposition and the consequent 100% set of fruits, the hybrid may have been deliberately made by human manipulation. The parents overlap in their native range only in S. and S.E. United States and the West Indies, but I can trace no mention of a hybrid in the N. American literature. It is possible, however, that one (or more) of the current records for L. repens in Europe are better incorporated within L. x kentiana, even though a different clone may be involved and so not perfectly matching the description given herein. In England, I have seen and confirmed only the Putney Heath record, but the following are likely to be either the same taxon or possibly pure L. repens. The voucher for v.c. 59 in BM possesses only flower buds: it is certainly not pure L. palustris, as labelled. The earliest British record appears to be in 1927 (from v.c. 59). v.c. 11 (S. Hants). Southampton, Sholing, Miller's Pond, GR SU/45.10, 1958-1965, R. P. Bowman. Probably an aquarist’s introduction; pond now infilled. v.c. 14 (E. Sussex). Seaford Head, South Hill Barn, GR SZ/505.980, 31 August 1991, P. D. L. Maurice. In a dew pond. v.c. 16 (W. Kent). Tonbridge, GR TQ/5.4, 1989-1990, S. Melville (MNE). In a garden pond, but not deliberately planted. v.c. 18 (S. Essex). Earls Path Pond, Epping Forest, GR TQ/415.967, 3 August 1976, K. J. Adams & J. O. Mountford. In an acidic pond. v.c. 59 (S. Lancs). Manchester Docks, GR SJ/8.9, 1927, L. Adams.(?LTN). By Failsworth Canal, GR SD/8.0, 14 May 1960, Miss V. Gordon (LIV); ibid., 26 July 1960, J. E. Lousley & Rev. C. E. Shaw (BM). Plentiful for several years. A map showing all Ludwigia records for the British Isles is given by Preston & Croft (1997), but all the above colonies, except Seaford Head, are probably now extinct. Surprisingly, the v.c. 9 (Dorset) record predicted by Clement (1997) to be this alien has since been determined by D. Pearman & myself as pure (and presumably native) L. palustris: it has been seen in 1996-1998 in a newly dug-out old pond at Sutton Holms, Edmonsham (Pearman 1999), DISCUSSION Conservationists may well be alarmed to hear that yet another vigorous aquatic plant is threatening to oust our native plants from almost any pond, particularly if acidic. The history of this hybrid, to date, suggests that it does not jump from site to site as readily as Crassula helmsii (Kirk) Cockayne and also that extreme cold weather appears to exterminate it completely - but a wary eye on its progress is necessary. Indeed, according to Salisbury (1972), even our native L. palustris “probably ...often behaves as an annual and ‘perennates’ perhaps only when the winter is not severe’. The ability of the hybrid Ludwigia to climb up and over other low vegetation in ponds is a little worrying, while the tendency for warmer summers will certainly favour its spread. More than one clone of the hybrid may have occurred in Britain, each colony spreading solely by vegetative propagation, and with a probable variation in hardiness (and hence its potential threat). A glance at Clement & Foster (1994) reveals that this is the first alien Ludwigia to be found in the British Isles, but others seem likely to follow: worldwide, 82 species are currently recognised (Raven 1963) and many of them have weedy tendencies. ACKNOWLEDGMENTS I am grateful to Prof. C. A. Stace for advice and for providing me with the Latin diagnosis. Dr C. D. Preston kindly provided me with information on all Ludwigia alien records in Britain. 172 E.32.CLEMENT Mrs E. Norman and P. Stanley enabled D. J. P. Smith to observe material in peak, fresh condition for his beautiful artwork presented here. Additional help has been freely given by R. P. Bowman, Miss V. Gordon, D. Pearman, Miss P. Miles, M. Mullin, A. Underhill and B. Wurzell. REFERENCES BRITTON, N. L. (1918). Flora of Bermuda. Charles Scribner’s Sons, New York: CLEMENT, E. J. & FOSTER, M. C. (1994). Alien plants of the British Isles. Botanical Society of the British Isles, London. CLEMENT, E. J. (1997). Ludwigia x muellertii Hort. new to Britain. BSBI News 77; 54. CosTE, H. (1903). Flore de la France 2: 83. Paul Klincksieck, Paris. FERNALD, M. L. (1987). Gray’s Manual of Botany 2: 1055. Dioscorides Press, Portland, Oregon. FORSTER, J. R. (1771). Flora Americae septentrionalis. B. White & T. Davies, London. GARVE, E. & MEIJDEN, R. VAN DE (1997). Ludwigia natans Elliott bei Hannover sowie hinweise zur bestimmung adventiver Ludwigia-arten. Floristische Rundbriefe 31(1): 9-12. GODFREY, R. K. & WOOTEN, J. W. (1981). Aguatic and wetland plants of southeastern United States. Dicotyledons. University of Georgia Press, Athens. GREUTER, W. et al. (1994). International code of botanical nomenclature (Tokyo code). Regnum Vegetabile 131. Hoc, P. C. (1993). Ludwigia in HICKMAN, J. C., ed. The Jepson Manual, pp. 800 & 807. University of California Press, Berkeley. JACKSON, B. D. (1893). Index kewensis 2: 123. Clarendon Press, Oxford. KASSELMANN, C. (1985). Viele jahre in Kultur, erst jetzt erkannt: Ludwigia repens Forster x arcuata Walter. Aguarien - und Terrarien - Zeitschrift (DATZ) 38(8): 377-380 (Not seen). Munz, P. A. (1961). Ludwigia L. in LUNDELL, C. L. et al. Flora of Texas 3: 214-220. Texas Research Foundation, Renner. NIETO FELINER, G., ed. (1997). Ludwigia L. in CASTROVIEJO S. et al. Flora iberica 8: 87-90. Real Jardin Botanico, Madrid. PEARMAN, D. (1999). Ludwigia palustris in Dorset. BSBI News 80: 17-18. PRESTON, C. D. & CROFT, J. M. (1997). Aquatic plants in Britain and Ireland, pp.107—109. Harley Books, Colchester. RADFORD, A. E. et al. (1968). Manual of the vascular flora of the Carolinas. University of North Carolina Press, Chapel Hill. RAVEN, P. H. (1963). The Old World species of Ludwigia (including Jussiaea), with a synopsis of the genus (Onagraceae). Reinwardtia 6: 327-427. RAVEN, P. H. (1977). Onagraceae. Flora Malesiana series I, 8(2): 98-113. SALISBURY, E. J. (1972). Ludwigia palustris (L.) Ell. in England with special reference to its dispersal and germination. Watsonia 9: 33-37. SOERJANI, M. ef al., eds. (1987). Weeds of rice in Indonesia. Balai Pustaka, Jakarta. STACE, C. A. (1997). New Flora of the British Isles, 2nd ed., p. 449. Cambridge University Press, Cambridge. STAFLEU, F. A. & COWAN, R. S. (1976). Taxonomic literature, 2nd ed., 1: 857. Bohn, Scheltema & Holkema, Utrecht. (Accepted November 1999) Watsonia 23: 173-177 (2000) 3 A new section and species of Festuca (Poaceae) naturalized in England R. FLETCHER and C. A. STACE Department of Biology, University of Leicester, Leicester, LE1 7RH ABSTRACT An introduced species of Festuca, well naturalised in a limestone quarry in N.W. Yorkshire for at least 70 years, is identified as F. gautieri (Hack.) K. Richt. subsp. scoparia (A. Kern. & Hack.) Kerguélen. The same taxon occurred on a roadside in S. Devon. A description and chromosome counts are given, and the origin of the plants is discussed. KEYWORDS: Fescue, grass. INTRODUCTION British species of Festuca L. are usually placed in three sections: Bovinae Hack. (F. pratensis Huds., F. arundinacea Schreb. and F. gigantea (L.) Vill.), Montanae Hack. (F. altissima All.) and Festuca (Ovinae Hack.; the F. rubra L. and F. ovina L. aggregates). Species of section Eskia Willk. (Variae Hack.) resemble those of sect. Festuca in being relatively small plants with fine leaves, but as natives they do not approach Britain closer than the Jura, Alps and Pyrenees. Herbarium material (well past flowering) of an unknown Festuca was sent from N.W. Yorkshire to C.A.S. in 1993 by Mr R. M. Payne, and living material as well as fully flowering herbarium material of the same plant by Mrs D. Millward in 1994. This was studied at Leicester in 1994 and 1995 by R.F. and C.A.S., and identified as F. gautieri (Hack.) K. Richt. subsp. scoparia (A. Kern. & Hack.) Kerguélen, a member of section Eskia from the Pyrenees. An introduced Festuca discovered in S. Devon in 1990 by Mr R. Takagi-Arigho had been confirmed by Dr T. A. Cope as F. gautieri, and herbarium material (well before flowering) was sent to Leicester in 1994. Living material was sent to C.A.S. in 1994 by Mr L. J. Margetts. This was also identified as F. gautieri subsp. scoparia. The Yorkshire site was also visited by R.F. MATERIAL As well as the living and herbarium material from the two English localities, herbarium material of F. gautieri subsp. gautieri and subsp. scoparia and of related species in BM, K and LTR was used for comparative purposes. METHODS Leaf-sections were cut on a freezing microtome either from fresh leaves or from dried leaves resuscitated by boiling in water for 5 min. Chromosome counts were obtained from root-tip mitoses using standard squash techniques. Pre- treatment was with saturated a-bromonaphthalene at 4°C for 24 hr, fixation in ethanol: ethanoic acid (3 : 1) for 30 min, hydrolysis with 5M HCl for 10 min at room temperature, and staining with aceto-orcein. Giemsa C-banding techniques were those used by Bailey & Stace (1992), and DAPI staining techniques were those used by Bailey et al. (1993). 174 R. FLETCHER AND:.C.%>. STACE RESULTS CYTOLOGY Good squashes were obtained of the S. Devon and N.W. Yorks plants and the number 2n = 14 found in both. Mitotic C-banding of the chromosomes with Giemsa staining, and staining with DAPI, showed most of the chromosomes with one or two narrow interstitial bands. This contrasts strongly with species of section Festuca (F. ovina agg. and F. rubra agg.), which have thick terminal bands following both of these techniques (Bailey & Stace 1992, Bailey et al. 1993). These bands are considered to represent AT-rich heterochromatin (Bailey et al. 1993). IDENTIFICATION Although superficially resembling a segregate of F. ovina agg., the then unknown Yorkshire Festuca was quickly recognized as a member of section Eskia rather than of section Festuca by the following two characters: 1.Lemmas scarious for most part, with only a narrow central part green or purplish-green; 2.Caryopsis not adherent to the lemma or palea. The following additional characters also distinguish the Yorkshire plant from any British taxa of the F. ovina aggregate: 1.Growth habit a spreading, springy cushion, not a firm dense tuft; 2.Leaf-blades pungent (sharply pointed) at the apex. The combination of the sharp leaf-tips and the springy cushions give the growing plant the feel of a live hedgehog; 3.Ligules of culm-leaves at least 0-5 mm long; 4.Anthers at least 3 mm long; 5.Ovary and caryopsis pubescent at apex (in the British taxa of section Festuca this character is found only in F. heterophylla Lam., a member of the F. rubra aggregate). Use of the keys in Markgraf-Dannenberg (1980) and Kerguélen & Plonka (1989) unequivocally identified the plant as F. gautieri. The combination of ligule less than 2 mm, panicle less than 7 cm, leaves less than 0-7 mm wide and with 5(—7) (not more) vascular bundles, spikelets often >9 mm, and leaves pungent at the apex distinguish it from other European species of section Eskia. Markgraf-Dannenberg (1980) did not recognise any infraspecific taxa in F. gautieri, but “F. scoparia A. Kerner ex Nyman” was cited in the index as a synonym of F. gautieri. Kerguélen & Plonka (1989) recognized two subspecies under F. gautieri: subsp. gautieri and subsp. scoparia (A. Kern. & Hack.) Kerguélen. The differences between those concern chromosome numbers, habitat preferences, and rather minor quantitative features. Table 1 lists the differentiating characters given by Kerguélen & Plonka for the two subspecies, and our measurements of these characters for herbarium material of the two subspecies from the Pyrenees and for plants from the two English localities. The diploid chromosome number of the English plants clearly identifies them as subsp. scoparia. With regard to the morphological characters said by Kerguélen & Plonka (1989) to distinguish the two subspecies, the majority indicate that the Yorkshire plant is subsp. scoparia. According to our measurements of herbarium material, lemma length (up to 6-2 mm) appears to be the best discriminant (see Discussion). Unfortunately the flowering material of the Devon plant was too immature to provide comparative measurements. HABITAT Kerguélen & Plonka (1989) described subsp. scoparia as “often calcicole in the Pyrenees”, and said that G. Claustres claimed that it is a strict calcicole in Dép. Ariége (central French Pyrenees). On the other hand they state that subsp. gautieri occurs on siliceous soils. Both grow in short turf and scree in the Pyrenees and eastern Spanish mountain ranges south to the Sierra Nevada and Morocco (Kiipfer 1974). FESTUCA GAUTIERI 175 TABLE 1. CHARACTERS OF THE TWO SUBSPECIES OF FESTUCA GAUTIERI ACCORDING TO KERGUELEN & PLONKA (1989) AND THE PRESENT AUTHORS Subsp. Subsp. Subsp. Subsp. Subsp. Subsp. gautieri fide —_ scoparia fide gautieri scoparia scoparia — scoparia Kerguélen& Kerguélen & (Pyrenees) (Pyrenees) ex ex Devon Plonka (1989) Plonka fide Fletcher fide Fletcher Yorkshire Ligule of culm leaves 0-5-1 Ce 0-5—1-2 0-4-1 0-7-1-2 0-6-1-1 (mm) Panicle length (cm) 4-5-7 7-9 3-7-4-7 5-2-6-7 5-6-7:3 - Spikelet length to tip (9-5) 10-12 up to 9 (10) 9-5—12-5 8-6-10-6 9-5—11-5 - of 4th lemma (mm) Upper glume length 4-]-4-5 4-5-5 3-54-6 4-1-5-6 3-4-4:3 - (mm) Lemma length (mm) 6-7-7 5-2-6-2 5-5-7-1 4-9-6-2 4-3-6-2 - Anther length (mm) 2-9-3-5 (2-1) 2-2-3 (4) 1-7-3-3 3-4-3-6 2-9-4-] - Chromosome number P= 29 2n= 14 - - 2n = 14 2n= 14 In Yorkshire the plant grows in Seata Quarry, Aysgarth, N.W. Yorks., colonizing a flat limestone slab at the edge of the quarry, which must to some degree mimic the Pyrenean habit of subsp. scoparia. There is a single patch measuring about 4-5 x 1-5 m. In Devon the plant grew on a road embankment near Exeter, S. Devon. ORIGIN AND HISTORY IN ENGLAND The S. Devon plant was “fairly obviously an introduction” (Margetts 1993), and presumably a recent one. We have not been able to ascertain whether it still exists there. Although F. gautieri makes an excellent and decorative rock-garden plant it is not mentioned in The European Garden Flora (Walters et al. 1984); however, its relative (also with pungent leaf apices and from the Pyrenees), F. eskia DC., is and it is possible that F. gautieri could be supplied under that name. In addition, F. gautieri is now listed in The Plant Finder (Philip, 1999) as being available from several nurseries. The N.W. Yorks. plant occurs in Seata Quarry with two other notable aliens, Erinus alpinus L. and Hypericum nummularium L., which also are native in the Pyrenees. The long exserted anthers of F. gautieri indicate that it is an outbreeder, since most such festucoids are self-incompatible (Barker & Stace 1982). Hence the lack of seed-set in the Yorkshire plant is what would be expected from a single clone introduction. If that is the case, a patch now measuring 4-5 m across will be far from a recent arrival. Mrs D. Millward has informed us that the quarry ceased to be worked around the early 1920s, and that as far as local knowledge goes the “aliens” (Erinus and Hypericum) were there then. It is possible that they were introduced deliberately by the Backhouses, who were planting “The Rockery” (now a listed building despite being a rock-garden!) not far away in the middle of the 19th century. NOMENCLATURE This species was first described as F. varia Haenke subsp. scoparia A. Kern. & Hack. in 1881 at a time when only two species of section Eskia were recognized in Europe (Hackel 1882). Hackel (1882) recognized three varieties of subsp. scoparia: var. genuina, the new var. gautieri Hack., and var. lutea Hack. The last (from calcareous areas in the Romanian Carpathians) was completely ignored by Markgraf-Dannenberg (1980), but probably represents a distinct species. At the species level F. scoparia (A. Kern. & Hack.) Nyman was created also in 1882, but this name is pre-dated by F. scoparia Hook. f. (1844). Festuca gautieri (Hack.) K. Richt. was created in 1890, and is therefore the correct name at species level for the two taxa together; Kerguélen (1983) combined scoparia as a subspecies of F. gautieri. 176 R. FLETCHER AND CC. A. STACE DESCRIPTION OF ENGLISH PLANTS F. gautieri (Hack.) K. Richt. subsp. scoparia (A. Kern. & Hack.) Kerguélen Stems spreading, + stoloniferous and forming dense springy cushions or mats up to 30 cm high; rhizomes 0; innovations all intravaginal. Leaf-blades up to 9 cm, 0-4—0-6 mm diameter, glabrous abaxially, often curved, sharply and stiffly pointed at apex, setaceous, distinctly 5—7-angled, with a discrete abaxial sclerenchyma girder in each angle, with 5 (—7) vascular bundles, with 2 grooves; ligule 0-6—1-2 mm; leaf-sheaths glabrous, with free overlapping edges in distal half, with distinct rounded auricles. Culms up to 45 cm, erect, glabrous; panicles 5—7-5 cm, with rather few (up to 15) spikelets, rather compact, with scabridulous branches and pedicels. Spikelets 5-9—11-5 mm (to tip of 4th lemma), with 3-6 florets, glabrous; lower glume 2-9-3-5 mm, with | vein, hyaline for most part; upper glume 3-4—4-3 mm, with 3 veins, hyaline for most part; lemma 4-3—5-5 mm (excl. awn), with 5 veins, rounded on back, hyaline for most part, with awn 0-25—0-6 mm; palea shorter than lemma, 2-keeled and 2-veined, bifid, with scabrid apex and ciliate keels; lodicules 2, bifid; stamens 3, with blackish-purple, long-exserted anthers 2-9-4-1 mm; ovary obovoid, pubescent in apical part. 2n = 14. DISCUSSION The distinction between the two subspecies of F. gautieri 1s not easy to make, and probably it would not merit subspecific rank if the differences in chromosome number and habitat preferences did not exist. The main differences noted by Hackel (1882) were the thinner, green (not glaucous- green), 3-ribbed (not 1-ribbed) leaves with 5 (not 7) veins, a more angular (not more rounded) abaxial surface and discrete bundles (not a continuous band) of abaxial sclerenchyma, and spikelets 9-10 mm (not 10-12 mm) in subsp. scoparia compared with subsp. gautieri. These discriminants are partly corroborated (e.g. spikelet length, leaf angularity in section) but partly contradicted (e.g. leaf colour and sclerenchyma distribution) by Kerguélen & Plonka (1989), and our observations do not all coincide with either (Table 1). It seems clear that all three sets of measurements are based on an insufficient number of specimens, and firm conclusions cannot yet be drawn as to the best diagnostic characters that can be used to separate the two subspecies. However, there is agreement between Kerguélen & Plonka and ourselves that subsp. scoparia has a longer panicle and shorter spikelets, upper glumes and lemmas than subsp. gautieri, although the actual measurements do not agree. Our measurements of the Yorkshire plant agree with our measurements of Spanish herbarium material of subsp. scoparia more than with those of subsp. gautieri with respect to panicle and lemma length, but not with respect to spikelet and upper glume length. The different habitat of the Yorkshire plant might at least partly account for this anomaly. The Yorkshire plant almost certainly originated from a deliberate planting at least 75 and probably well over 100 years ago. With the increased use of foreign grass-seed mixtures on roadsides, etc., and the introduction of ever more exotic and esoteric garden ornamentals, it seems likely that further records of F. gautieri will be made in the wild in Britain. ACKNOWLEDGMENTS We are grateful to R. M. Payne, D. Millward, L. J. Margetts and R. Takagi-Arigho for sending us material and supplying us with information, and to Dr J. P. Bailey for help with the cytological work. REFERENCES BAILEY, J. P., BENNETT, S. T., BENNETT, M. D. & STACE, C. A. (1993). Genomic in situ hybridization identifies parental chromosomes in the wild grass hybrid x Festulpia hubbardii. Heredity 71: 413-420. BAILEY, J. P. & STACE, C. A. (1992). Chromosome banding and pairing behaviour in Festuca and Vulpia (Poaceae, Pooideae). Plant systematics and evolution 182: 21—28. BARKER, C. M. & STACE, C. A. (1982). Hybridization in the genera Festuca and Vulpia: the production of artificial F; plants. Nordic journal of botany 2: 435-444. FESTUCA GAUTIERI La HACKEL, E. (1882). Monographia Festucarum europaearum. Theodor Fischer, Kassel & Berlin. KERGUELEN, M. (1983). Les Graminées de France au travers de ‘Flora Europaea’ et de la ‘Flore’ du C.N.R:S. Lejeunia, n. s., 110: 1-79. KERGUELEN, M. & PLONKA, F. (1989). Les Festuca de la flore de France. Société Botanique du Centre-Ouest, Dignac. KUPFER, P. (1974). Recherches sur les liens de parenté entre la flore orophile des Alpes et celle des Pyrénées. Boissiera 23: 11-322. MARGETTS, L. J. (1993). 85th report on botany. Vascular plants. Report and transactions of the Devonshire Association for the Advancement of Science, Literature and the Arts 125: 227-235. MARKGRAF-DANNENBERG, I. (1980). Festuca, in TUTIN, T. G. ef al., eds. Flora Europaea 5: 125-153. Cambridge University Press, Cambridge. PHILIP, C., ed. (1999). The RHS Plant finder 1999-2000. Royal Horticultural Society/Dorling Kindersley, London. WALTERS, S. M. et al., eds. (1984). The European garden flora 2. Cambridge University Press, Cambridge. (Accepted April 1999) fee : Febe. an Watsonia 23: 179-196 (2000) 179 Assessing the status of Stratiotes aloides L. (Water-soldier) in Co. Fermanagh, Northern Ireland (v.c. H33) R. S. FORBES Institute of Continuing Education, The Queen’s University of Belfast, Belfast, Northern Ireland, BT7 INN ABSTRACT A case is argued for considering Stratiotes aloides L. (Water-soldier) (Hydrocharitaceae) of native status in Co. Fermanagh (v.c. H33), and possibly elsewhere in Ireland. The case is circumstantial in the absence of fossil evidence from the current interglacial, and consists of early 19th century discovery, long persistence in natural water bodies, rare and scattered occurrence, and lack of the “boom and bust” dynamics of many introduced aquatic species. No new stations have been found for over 100 years which strongly suggests that either S$. aloides is not much used in Irish water gardens, or it does not readily escape into the wild. The species complement of two vegetation communities involving S. aloides in Upper Lough Erne, Co. Fermanagh is strikingly similar to communities described in the Norfolk Broads, where the species has long been considered native. Conservation arguments for sites frequently revolve around the occurrence of rare indigenous species, making the question of status more than a purely academic matter. KEYWORDS: Hydrocharitaceae, aquatic macrophyte, native status, garden history. INTRODUCTION Botanists for too long have uncritically accepted notions of indigenous or introduced plant status largely derived from early observations and old studies made in southern England, and have continued to apply them to situations elsewhere without a great deal of reconsideration (Watson 1883; Bentham & Hooker 1900; Hyde & Wade 1957; Clapham et al. 1962; Webb 1985). Progress in botanical knowledge in Ireland has always lagged behind that in Great Britain, originally by up to a century or more, but it has tended to tramp the same path. However, it needs to be said that what is true and self-evident for the south of England does not necessarily apply in the different biological, historical and human circumstances operating in Ireland, or in other parts of these islands. Wealth, leisure and education combine in promoting both natural history and garden pursuits, and in timing and intensity these endeavours developed differently in the two major islands. In particular, the passion for gardening and the fashions in it did not in Ireland slavishly follow the English scene, but had a regional, climatically-governed identity, frequently constrained by limited resources rather than by any lack of imagination (Nelson & Brady 1979; Lamb & Bowe neo). The status of Stratiotes aloides L. (Water-soldier) is a case in point, a species whose native distribution throughout its whole European range is problematic, since it has been grown as an aquatic ornamental for nearly three centuries and its native occurrence has been obscured by numerous escapes from cultivation (Cook & Urmi-K6nig 1983; Preston & Croft 1997). S. aloides is described by Stace (1997) as native, its distribution in the British Isles summarised (abbreviations expanded here) as, “now very local in E Anglia, N and S Lincolnshire, SE Yorkshire and Denbighshire, formerly locally frequent in central Great Britain, introduced in scattered places in Great Britain, north to central Scotland.” He makes no mention whatsoever of S. aloides occurring in Ireland. The species is described as being a dioecious perennial, and Stace adds the significant remark, “only female plants occur in Britain” (but see below). Perring & Walters (1976) in the Atlas of the British Flora are less dogmatic with regard to the status of the plant, their map distinguishing between “probably native” and “probable or certain introductions”. On the basis of the 17th century records, Preston & Croft (1997) regard it as likely that S. aloides is native in eastern England “and perhaps also in Cheshire” (v.c. 58) (but see below). 180 R. S. FORBES With regard to S. aloides in Ireland, Perring & Walters (1976) indicate five widely scattered stations of “probable or certain introduction” status: three around Belfast Lough, Co. Antrim, one on Upper Lough Erne, Co. Fermanagh, and one in Co. Cork. Webb et al. (1996) state that the plant is naturalised in Upper Lough Erne and near Cork. The Census Catalogue of the Flora of Ireland (Scannell & Synnott, 1987), lists $. aloides as an introduction in three vice-counties: West Cork, Fermanagh and Antrim. R. D. Meikle, on the other hand, takes a totally different view of the status of S. aloides in his unpublished (1957) typescript Flora of Co. Fermanagh, based on collaborative field work with E. N.Carrothers, R. C. Davidson & J. McK. Moon. In this Meikle states his opinion that, “Stratiotes aloides has as good a claim to be considered indigenous in Fermanagh as Hydrocharis morsus- ranae (Frogbit) or any other aquatic”’. The present paper considers the status of S. aloides in Fermanagh in the light of definitions and criteria for such deliberation which were carefully delineated by Webb (1985), and supplemented by Preston (1986). Basic to Webb’s definition of native or alien plant status is the consideration of human involvement (i.e., Neolithic man onwards). Fossil evidence from the period before Neolithic farming activity is the only concrete proof of native status that Webb accepts; all other evidence is circumstantial and, as such, for each species it must be subjected to careful, objective scrutiny and the balance of probabilities weighed. I will begin by reviewing the distribution and history of the species in relation to human activities in Europe and the British Isles, in order to determine what patterns emerge, how they might apply in Ireland, and how the status of S. aloides in Fermanagh appears in the light of such investigation. FOSSIL EVIDENCE S. aloides can be recognised to species level both from macro-fossils and from fossil pollen (Andrew 1970). Dickson (1970) describes how the species can be distinguished from the extinct S. intermedius Hartz by its tough resistant leaf spines, which have a main pointed cell and subsidiary basal cells. Both species are illustrated by Katz et al. (1965). The species is also recognised from its fossil seeds which are often abundant, and the history of the genus has been traced back by means of these seeds into the Tertiary Era for some 48 million years, through as many as 15 fossil species (Chandler 1923; Cook & Urmi-Konig 1983). In the South of England, fossil pollen has been discovered in early Pleistocene interglacial sands and clays of the Pastonian (West 1961). This interglacial pre-dates both the Anglian (the first East Anglian glacial stage) and the preceding Cold Beestonian stage. S$. aloides also occurs as fossil pollen in the shelly sands of the Corton Beds, the name given to warm interstadial deposits within the Anglian glacial in East Anglia, which are of Middle Pleistocene age (Godwin 1975). In the Late Pleistocene it is found as seed in the Hoxnian interglacial, and again as macroscopic remains in the Ipswichian interglacial at Wretton and Selsey. Fossil pollen from the Ipswichian has also been recorded from Bobbitshole and Selsay (Godwin 1975). Other aquatics associated with S. aloides in the Selsay fossil record included Hydrocharis morsus-ranae, Sagittaria sagittifolia (Arrowhead) and Butomus umbellatus (Flowering-rush). Godwin obliquely poses an unanswered question about S. aloides’ means of dispersal when he comments that, “It is of interest that whilst the fossil record provides no evidence whatever of this thermophilous plant in the glacial stages, it occurs nevertheless in every interglacial (my emphasis) from the Pastonian onwards” (Godwin 1975). Despite the implication contained in this quote, no specific reference is given here to a fossil presence of S. aloides in the early part of the current interglacial, the Flandrian, and I am not aware of any such record. The above plentiful fossil occurrences, read in conjunction with early first records, help support the native status of S. aloides in at least south-eastern England, yet fail to prove it conclusively in Webb’s terms. Fossil material has been noted in only one Irish deposit so far, at Shortalstown, Co. Wexford, where a single grain of Stratiotes pollen, thought to be of Ipswichian (Late Weichselian) age, was discovered by Colhoun & Mitchell (1971). Although this appears to be the only Irish fossil of Stratiotes, the reason for this may well be that no one has been looking for it; in addition, to date no palynological studies of Holocene lake sediments have ever been made in Fermanagh or Cavan, the region of S. aloides’ main past and current distribution (Dr Valerie Hall, pers. comm., 1998). STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 181 GEOGRAPHICAL DISTRIBUTION AND HISTORICAL EVIDENCE EUROPE The map produced by Cook & Urmi-K6nig (1983) shows S. aloides occurring in the British Isles and in Europe south to Spain, Italy, Bulgaria and European Turkey, north to Finland, and east to Siberia (Fig. 1). Throughout the range Cook & Urmi-KGnig discovered it difficult to determine its status, but they reviewed the scientific opinion in relation to the fossil occurrences, floristic literature (Sometimes documenting definite introductions), and herbarium records. They showed that the distribution of S. aloides today agrees very closely with the distribution of the 15 fossil species. The single exception was the find of S. tuberculatus E. M. Reid from the Pliocene in Portugal, which alone supports the two unconfirmed literature records of S. aloides in Spain from the 18th and 19th century (Willkomm & Lange 1861). This is a particularly interesting disjunction in the species range, as it is generally accepted that S. aloides is not native to France on grounds of its late first mention in the relevant literature (1810), a series of documented introductions, and the lack of any fossil finds. A similar puzzle exists with respect to Italy (where it is considered native in the eastern part of the Po Plain) and Switzerland (“certainly not native’) (Cook & Urmi-K6nig (1983)). Cook revised his opinion of the native range of S. aloides two years later, asserting that it is indigenous only in Central Europe and is introduced and established in Western Europe (Cook TA : t : L .) L — e t} ' L a ? tte i a 0% 4 ee @ elon 4 343 nai ie @ 42 6 4 C4 00 rode) e@ us L L dae L 2 4 ° e =i L L L a L L L L 4 4 ry L a e eA ¥ @ “oe int: @,. 2en L @ a6 c hg ta a L be : aK L L & L 1S L L L Lada e O N ® L > L L FIGURE 1. Distribution of Stratiotes aloides L. in Europe, with sexes determined from herbarium material (after Cook & Urmi-K6nig 1983), and with the British Isles updated by the author. @ female plant, A male plant, O plant of unknown gender, L record from the literature only. 182 R. S. FORBES 1985). No explanation accompanies this reassessment, but in the light of his earlier Stratiotes review (Cook & Urmi-Konig (1983)) I believe it may involve the distribution of a dragonfly associate (See section on insect associates below). GREAT BRITAIN In the British Isles, on the basis of early records, the species is regarded by most naturalists as indigenous in the east of England (Geldart 1906; Stewart et al. 1994; Preston & Croft 1997) (Fig. 2). It was first reported by Thomas Johnson in the East Riding of Yorkshire in 1626 (Crackles 1990), followed by Lincolnshire in 1633 by the same finder (Gerarde 1633; Gibbons 1975). Old records from the 17th and early 18th centuries must be accepted with caution, as the name Stratiotes was applied by different authors to a number of quite unrelated plants, including Achillea millefolium, Hydrocharis, Hottonia palustris and even some Umbelliferae (Geldart 1906). Geldart also lists the English common names given by Johnson in the second edition of Gerard’s Herball (Gerarde 1633) as, “Water houseleek, Knights’ pondwort, Knights’ water sengreen, Freshwater soldier, Wading pondweed, Soldier’s yarrow and Knights’ water woundwort’. The fact that seven English common names for the plant were in use by 1633 suggests that it was already widely known by local people when the first plant record was published. The native range of Stratiotes has contracted sharply in the last 150 years (Fig. 2). It is now extinct in its previous stronghold, the Cambridgeshire fens (Preston & Croft 1997), and it has largely disappeared from Lincolnshire and Yorkshire also (Stewart ef al. 1994). The greatest concentration of surviving Stratiotes sites regarded as native in England occur in East Norfolk (v. c. 27), where it is currently found in 43 tetrads (nine 10 km squares), in ditches (or dykes) in grazing marshes around the River Waveney. Even in this corner of England, however, there are ten to twelve more-or-less isolated sites where the species is suspected of being introduced. Apart from S.E. England and East Yorkshire, where the plant has a long history, there is justifiable doubt and uncertainty with regard to its status in the rest of Great Britain due to a combination of late first records, documented introductions, colonies in man-made or heavily disturbed habitats, and a lack of local fossil evidence, very much paralleling the situation in France (Watson 1883; Cook & Urmi-K6nig 1983; Stewart et al. 1994; Preston & Croft 1997). Preston & Croft (1997) suggested that S. aloides might be native in Cheshire (v.c. 58). It transpires that this assertion was made on the same basis as that used by Stewart et al. (1994) while working on the Scarce Plants in Britain project, 1.e., that when required to separate native from alien records, unless there are good grounds for a difference, the treatment of the species follows that given by the Atlas of the British Flora (Perring & Walters 1962 and subsequent reprints with minor corrections and additions). “We think that this (1.e., The Atlas) presents a realistic assessment of the status of most species, and also recognise that there is an arbitrary element to the decisions which have to be made and there is no point in presenting alternative opinions if the evidence to support them is weak.” (Stewart et al. 1994). Perring & Walters (1962) accepted the view of De Tabley (1899) who stated in his Flora of Cheshire that S. aloides seemed to him as native in central Cheshire as anywhere in Britain, although he gave no reasons for making this assertion. In the same work De Tabley also cited several known introductions of the species in Cheshire (G. McKay, pers. comm., 1998). The more recent Flora of Cheshire (Newton 1971) does not claim native status for S. aloides in the county, and C. D. Preston (pers. comm., 1998) now agrees that in retrospect, “the case for native status in Cheshire is weak and we should perhaps have shown less deference to the Atlas map.” S. aloides occurs in four 10 km squares in Denbighshire, and Ellis (1983) claimed them native (J. Green, pers. comm., 1998). The first Denbighshire record was made by the Rev. W. E. Jones in 1938 and was included in the 2nd edition of Welsh Flowering Plants by Hyde & Wade (1957), who assumed it to be native. Ellis updated the records from the same area for his Flowering Plants of Wales, and repeated the assumption of native occurrence since he had no reason to question it (Ellis 1983). In the light of the Stratiotes map published in Stewart et al. (1994), Ellis has revised his opinion and has written, “without any firm evidence to the contrary, I would probably class the v.c. 50 populations as introduced.” (G. Ellis, pers. comm., 1998). It is not at all surprising that the older Cheshire and Denbighshire records were simply assumed to be native by the botanists of the time. While describing the range of circumstantial evidence that needs to be assembled and weighed to enable a realistic consideration of a plant’s status, Webb STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 183 FIGURE 2. Distribution of Stratiotes aloides L. in the British Isles. Symbols represent at least one record in a 10 km square. O pre-1975 putative native records, ® putative native records made in or after 1975, + records of introduced population made before 1950, x records of introduced populatons made in or after 1950. Based on Preston & Croft (1997) and updated by the Biological Records Centre as requested by the author. (1985) commented that in assigning native status there is often an element of wishful thinking, sometimes amounting to local patriotism, or an impression that the amount present in an otherwise undisturbed “natural looking” habitat was sufficient evidence in itself to claim a find as being native. Webb also gave telling examples of why this level of assessment is unreliable. IRELAND Co. Antrim (H39) The first Co. Antrim record was made in 1815 by John Templeton, a well-known and respected amateur naturalist. He described the site at Stranmillis on the (then) outskirts of Belfast as, “In the back drain of the Belfast watercourse”. Praeger (1938) and Hackney (1992) both regarded this and 184 R. S. FORBES the other Co. Antrim records as introduced, noting that three out of four stations were extinct by 1938. Hackney also records that the plant has not been seen since 1930 at the last remaining Co. Antrim station, the mill-dam at Woodburn, where it was reported as “occupying two acres” (Fisher 1930). Praeger (1938) commented in the second edition of The Flora of the North-East of Ireland that Stratiotes, “should be looked on as planted in our district; in Ireland it seems fully naturalised in Lough Erne alone” (my emphasis). In Co. Antrim the arrival, spread and departure of S. aloides fits the “boom and bust” behaviour pattern shown by many plant introductions (Simpson 1984; Clement & Foster 1994) and, apart from the early first discovery in a natural water course, there is no evidence with which to argue the contrary. I would prefer to consider Templeton’s Lagan site as of “possibly native” status but, mindful of Webb’s warning of vision clouded by local patriotism and recognising that the evidence really is insufficient, it must remain a (now extinct) “probable introduction”. Co. Fermanagh (H33) S. aloides was first found in Fermanagh in 1805, placing it amongst the very first plants recorded in the county. The discoverer of S. aloides, the Rev. Robert Scott, M.D., was Professor of Botany at Trinity College, Dublin from 1785 until 1808, and is regarded as the father of Fermanagh botany. A native of the county from Scottsborough House, Magheraveely, while primarily a bryologist, Scott evidently had a good eye for flowering plants (especially aquatics), and he was responsible for adding Utricularia intermedia Hayne (Intermediate Bladderwort), to the flora of the British Isles (Praeger, 1949). The Fermanagh records of Dr. Scott were published along with his finds in other parts of Ireland in the first edition of J. T. Mackay’s Catalogue of Rare Plants found in Ireland (1806). Mackay himself visited Fermanagh and the adjoining part of Cavan (H30Q) shortly after Scott, for in the same work he records his own find of S. aloides, “in drains on the roadside near Castle Saunderson, in the vicinity of Lough Erne’. Scott’s original S. aloides site is described as, “ growing in a drain near Crum Castle, on the banks of Lough Erne, county of Fermanagh, in August last.” Crum is currently spelt, “Crom”, and this locality near the Fermanagh-Cavan boundary where the Erne connects southwards with Belturbet, is still the main area of distribution for the plant today. S. aloides has recently been recorded again from the Cavan side of the border; the species was refound in 1996 at two loughs near Belturbet (Round Lough, H15.39, and Anoneen Lough, H22.35) (P. Reilly, pers. comm., 1998). This represents a gap of 128 years since the species was last listed in the same area by David Moore (Moore & More 1866). Robert Northridge and I, as joint B.S.B.I. vice-county recorders for Fermanagh (H33), have assembled a total of 44 S. aloides records in 15 tetrads (Fig. 3). Fourteen tetrads contain post-1975 records. The species was first recorded at the S.E. boundary of the county and subsequent finds extended its known occurrence northwards along Upper Lough Erne rather slowly. It was not found at Belleisle, about eight km. south of Enniskillen, until 1900. In 1990 a major survey of aquatic vegetation, commissioned by the Department of the Environment for Northern Ireland, extended the known distribution as far north as Lough Yoan, within 2 km of the town. In reality, however, the evidence is much too scant to conclude that the species is spreading in Fermanagh. Botanical recording in the county (as elsewhere in Ireland) has been extremely occasional, sporadic and unsystematic until the 1980s, and it has involved only a handful of people, few of whom resided locally. One can say with confidence, however, that S. aloides has not exhibited an explosive spread in the manner characteristic of some definitely introduced aquatic species, such as Elodea canadensis (Canadian Waterweed), or E. nuttallii (Nuttall’s Waterweed). Neither has it declined or disappeared on Upper Lough Erne as it has in its Co. Antrim stations, where its occurrence on man-made or disturbed aquatic habitats always tended to support the suspicion it might be an introduction. The headquarters of S. aloides in Fermanagh remains very much where it was first found almost two centuries ago, the southernmost fen-fringed lakes and inlets of Upper Lough Erne, where it sometimes dominates open water. In mid-August 1994, at the southern of two lakes in the district, both called “Corracoash”, Robert Northridge and I came across a striking example of its growth potential; a dense population covered the lake surface, approximately 3 ha in extent. In 1998 the communicating Mill Lough, approximately twice the size of Corracoash Lough, was so overgrown with S. aloides that it was difficult to row a boat through it (J. Bullock, pers. comm., 1998). STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 185 a / . a a FIGURE 3. Tetrad distribution of Stratiotes aloides L. (Water-soldier) in Co. Fermanagh (H33). Hatched symbols represent pre-1975 records, black symbols post-1975 records. Other stations in Ireland Following the Fermanagh, Cavan and Antrim finds already discussed, Mackay published a further site for S. aloides, “on the banks of the Shannon, near Portumna” (Co. Tipperary, H10) in his Flora Hibernica (1836). Mackay’s listing for this site showed S. aloides occurring on the major river of central Ireland with Sagittaria sagittifolia and Eleocharis acicularis (Needle Spike-rush). The Tipperary S. aloides had gone by 1866 when Moore & More published their Cybele Hibernica. In addition to its earlier northern sites, Moore & More (1866) also list S. aloides occurring, “naturalized (my emphasis) in Ballyphehane Bog” (Cork city; H4), a site which Mackay (1825) previously mentioned in passing, and which was near the Cork Botanic Garden. Significantly, Mackay (1825) does not list S. aloides for this site, although twelve other species from Ballyphehane Bog are attributed to his major Cork contributor, Mr Drummond. The Ballyphehane station is again ignored by Mackay in his Flora Hibernica (1836), which also suggests he knew it was not a native occurrence of the species. Drummond was the curator of the short-lived Cork Botanic Gardens, which was situated on a low limestone hillock in the city. It appears that Drummond simply used the nearby marsh (it was never a bog), as a convenient out-station for growing aquatic plant species for the Botanic Garden (M. Scannell, pers. comm., 1998). This is the only unequivocally introduced site for Stratiotes in Ireland. Colgan & Scully (1898) mention S. aloides as having occurred fleetingly (in 1829) in a canal near Drogheda, Co. Louth (H31). The lack of persistence and artificial nature of the waterway both suggest that a garden origin 1s likely in this case. 186 R. S. FORBES Moore & More (1866) considered S. aloides, “probably not indigenous” in southern Ireland, but suggested it might possibly be native in the north of Ireland. They recognised that the species was a rapidly spreading introduction elsewhere in the British Isles, and compared it with Elodea canadensis, (a plant which made its first appearance in these islands in Co. Down around 1836). As was the case in Co. Antrim, the S. aloides plants in these three widely scattered locations in southern Ireland between 1829 and 1836 appear to have become extinct soon after their discovery, except at Ballyphehane Bog where the species survived until some date after 1900. The apparently short-lived nature of the other two south of Ireland stations is typical of most introductions; the persistence of the species at Ballyphehane may be the result of deliberate management. The fact that there have been so few sightings of S. aloides in Ireland over the 192 years since its first discovery suggests that it is either truly rare, or so hard to detect that it is seldom found. As is the case with many Irish lakes, the waters of Upper Lough Erne are an opaque dark brown colour due to peat staining, so that a grapnel is an essential item of equipment to drag for submerged macrophytes. Even if one is actively looking for S. aloides, it can be difficult to see the characteristic “pineapple-like” leaf rosettes in deep murky lake water, and there could be very many plants invisible at depth. Under conditions of low illumination the leaves do not develop sufficient gas in their air-canals and intercellular spaces to raise the plant rosettes to the surface, and it is quite normal to find it as a bottom-dwelling plant in water between | and 5 m deep (Cook & Urmi-Konig 1983). In more shallow waters, particularly after a prolonged spell of warm weather, very rapid growth can produce the sort of Stratiotes “explosion” that occurred at Corracoash and Mill Loughs. HABITAT The typical habitat of Stratiotes is still or slow-flowing, meso-eutrophic water (Preston & Croft 1997). Upper Lough Erne consists of a multitude of small, shallow, linked lakes and intervening islands - a drowned drumlin landscape - and although there is a channel navigable for motor launches, the great majority of its shores are undisturbed. Drainage ditches are cleared from time to time by local farmers and, although fringing emergents encroach, open water is plentiful, as is rainfall throughout the year. Agriculture is entirely pastoral and low intensity on the surrounding rushy meadows and, while slurry is spread, eutrophication from run-off has not become a widespread problem. In recent years 60% of Fermanagh farmland has been covered by a government-funded Environmentally Sensitive Area farm subsidy scheme. In England S. aloides was formerly found in lakes and slow-moving rivers but now is virtually confined to drainage ditches (Preston & Croft 1997). In the Broads of south-eastern England S. aloides occurs in three habitats, all of them man-made. The Broads are flooded peat excavations, the fen dykes were created to allow boat access into the fens, and the network of ditches in the reclaimed marshes were created for land drainage (R. J. Driscoll, pers. comm., 1998). S. aloides appears to prefer the still water of grazing marsh ditches, particularly where the water is moderately rich in nutrients and calcareous, and it grows where regular cleaning of the ditches or ponds, (often grazed on their margins), suppresses the growth of shading reeds and other tall emergent species. S. aloides is even more intolerant of fluctuating water levels than its relative and frequent associate, Hydrocharis morsus-ranae, and unlike it does not survive on mud to form turions (Cock & Lii6nd 1982). Eutrophication, and habitat disturbance and destruction due to agricultural changes, are readily identified as the major reasons for the rapid decline in distribution and decreasing abundance in the 20th century. In east Norfolk, the main area of surviving populations considered indigenous, habitat loss continues due to a change from pasture to arable agriculture, subsequent pollution and, since the area is low-lying and near the coast, sporadic tidal inundation (A. Bull, pers. comm., 1998). On the other hand, Wheeler & Giller (1982) suggested that S. aloides, in the dyke system of the Catfield and Irstead fens in the Norfolk Broads, showed abundant growth at the fen margins, possibly in response to a degree of agricultural nutrient run-off, whilst the species has become rare in the central areas of these fens during the last 30 years. On the basis of water chemistry they considered it at least possible that this decline might be due to nutrient depletion associated with oligotrophication and acidification. A similar catastrophic decline of the species in. the Netherlands from the 1960s onwards followed increased drainage associated with modern agricultural practices, which diminished the STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 187 influence of ground water. To prevent the low-lying peat soil from drying out, sulphate-enriched alkaline water from the Rhine was admitted, which produced a complex of environmental stresses involving iron limitation, sulphide toxicity, a strong increase of non-rooted species due to internal and external eutrophication, and ammonium toxicity (Smolders ef al. 1996). The habitat requirements and tolerances of S. aloides are thus very specific and narrow (see community ecology section below). REPRODUCTIVE PATTERN Phytogeographically, Stratiotes is regarded as a Continental species by Matthews (1955), a position refined by Preston & Hill (1997), who described it as Eurosiberian Boreo-temperate. When trying to relate a species’ distribution to geography and climate, however, one must take particular account of the fact that different optimum and minimum conditions may occur for life processes such as seed set, germination and vegetative growth, and that biotypic variation may modify these physiological characteristics within the total range of a species. This is certainly the case with regard to growth and sexual reproduction of Stratiotes. Currently, reproduction of S. aloides in the British Isles and in most of Europe is entirely vegetative and is carried on by the production of new open leaf rosettes (or closed turions (Kornatowski 1979)), budded off on monopodial lateral shoots (Cook & Urmi-K6nig 1983). Renman (1989) studied the life histories of two clonal populations of S. aloides in northern Sweden and found that emergent forms of the plant produced more daughter turions than did submerged forms. There is the added complication of the dioecious nature of the plant and the present-day distribution of the sexes. The dioecy of Stratiotes is not absolute. Dioecy is secondary in origin and in S. aloides flowers of both sexes contain whorls of staminodes, which in the female flowers secrete nectar at their base (Daumann 1931). Although the staminodes are normally sterile, rarely the female flowers contain a few apparently fertile stamens on their outer whorl (Geldart 1906; Cook & Urmi-K6nig 1983; Preston & Croft 1997). Hutchinson (1948) went so far as to describe S. aloides as “subdioecious’’. The existence of hermaphrodite plants (even at low frequency) casts a different light on the possible reproductive history of the species in the post-glacial period. Full dioecy is represented when plants (or genets - defined as genetically distinct individuals), are totally male or totally female. According to Richards (1986), who was writing on dioecy in general terms, when sex is phenotypically unstable, hermaphrodites that are capable of selfing may occur. When only some of the flowers on a female genet become hermaphrodite (as in the case of S. aloides), they are termed subgynoecious, and this is considered the result of unstable sex expression in a genetically dioecious plant (Richards 1986). (When all the flowers of a female plant or genet become hermaphrodite, the species is said to show gynodioecy (i.e. female and hermaphrodites occurring together), but this does not apply in the case of Stratiotes). Plants that are fickle in their sex expression are often subject to environmental influence and, in the case of thermophilous species such as S. aloides, temperature is clearly the most likely significant factor - possibly linked with daylength. Sex expression in plants and its physiological control is a fascinating area of study (Meagher 1988), and touching on it ought to remind us that concepts such as “incompatible” and “dioecy”’, are decidedly “leaky”, and that our description of the natural world is frequently a convenient oversimplification of the variation that really exists (Proctor et al. 1996). Cook & Urmi-Konig (1983) reviewed the current distribution of the sexes in S. aloides, and found that while female plants strongly predominate north of a line drawn from Cherbourg to Leningrad (St Petersburg), isolated male plants have been found in Denmark, Sweden and Finland (Fig 1). North of the Cherbourg/Leningrad line, however, they did not find any evidence of S. aloides bearing seed and, although fruits may form, no viable seed has been found in the British Isles either, since the Pleistocene (Crackles 1982). This suggests three non-exclusive possibilities; (a) that for successful seed production, S. aloides requires a warmer summer than that currently found north of the Cherbourg/Leningrad line; (b) that male S. aloides plants are more susceptible to cold and have become extinct in what is now a scattered, relict population; or (c) male flowers are never or only rarely triggered to occur by the longer summer daylength or other environmental factors to the north. 188 R. S. FORBES The fossil assemblage found with S. aloides in the Corton Beds in East Anglia of Middle Pleistocene age indicated to West & Wilson (1968) a strong northerly aspect to the vegetation, with some persistence of Arctic-alpine species which occurred, “together with a number of aquatics, including S. aloides and Potamogeton crispus, with a mainly southern distribution”, (my emphasis). Considering the occurrence of fossil Stratiotes along with Cladium mariscus, Hydrocharis morsus-ranae and Lemna minor, West (1961) suggested that their joint presence indicated a considerable summer warmth in the Cromerian and Pastonian interglacials of the Norfolk coast. Evidence like this supports the notion that the prevailing temperature in the north and west of S. aloides’ current range is too cool for successful sexual reproduction, as even in regions where both sexes occur today (i.e. Sweden, Finland and Denmark), no viable seed appears to be set. Further south, in a warmer climate, seed is successfully produced, e.g. in the Netherlands (Smolders et al. 1993). Palynologists tell us that we are past the climatic optimum in the current interglacial period (the “Littletonian” in Ireland ( Mitchell & Ryan 1997)), so it is arguable that S. aloides flourished, reproduced and spread by seed in an ancient warmer climatic phase, and when cooler conditions developed it declined to its current disjunct relict areas, persisting by vegetative reproduction. To explain the occurrence of single sex populations in the British Isles, we have to postulate differential survival capacities of the sexes. A similar explanation involving differential vigour, ecology and survival between the sexes is used to account for the restricted occurrence of female plants of Petasites hybridus in the British Isles (one site only in Ireland), while males are widespread throughout both islands (Valentine 1946, 1947; Perring & Sell 1968). DISPERSAL ABILITY As Preston & Croft (1997) point out, without seed S. aloides appears to lack a means of long-range dispersal over land. The current lack of fruiting ability may have arisen, however, after small founder populations of seed successfully established in both England and Ireland, some plants perhaps temporarily retaining a degree of fertility through the possession of a subgynoecious habit. “Baker’s Law” (Baker 1955; Stebbins 1957) suggests that dioecious species appear to be at a major disadvantage when colonising islands compared to self-compatible hermaphrodite plants. Bawa (1982), however, has suggested that dioecious taxa may have been disproportionately successful in colonising islands. Analyses of numerous oceanic island floras by Baker & Cox (1984) indicated that dioecious taxa do not do better in long-distance dispersal than self- compatible hermaphrodites, but neither do they fare worse, a result that was unexpected and was not predicted by extensive numerical simulations (Cox 1985). While the distances and the timescales involved in colonisation of oceanic islands are far greater than those associated with continental islands such as the British Isles, the evidence remains that jump dispersal is effective and plants do colonise islands. The fact that in S. aloides dioecy is not absolute but “leaky” (Cox 1985), and that it is capable of producing subgynoecious self-compatible hermaphodites from time to time, overcomes some of the conceptual difficulties we face in imagining jump dispersal of the species to Britain and Ireland. Like two aquatics often associated with Stratiotes, Ceratophyllum demersum (Rigid Hornwort), of which Scannell (1976) found that it flowered but failed to ripen fruit in Ireland, and Hydrocharis morsus-ranae, which flowers erratically and seldom fruits (Cook & Liiénd 1982; Preston & March 1996), S. aloides possesses a successful means of vegetative multiplication. It is possible that a single founder or a single aggressive clone may have survived many years of asexual competition and a slowly changing, cooling environment, giving rise to high levels of sterility, even in species or regions where both sexes are currently sympatric (Richards 1986). Throughout north-west Europe S. aloides appears to fit this pattern closely, and on offshore islands the probability of a restricted genome ts clearly increased. The occurrence of fertile seed-producing plants across the North Sea in the Netherlands, and the small but possibly highly significant seed set which self-compatible subgynoecious plants could produce, provide an ancient seed source for colonisation of the British Isles, most likely through transportation by birds. Jump-dispersal of turions is possible, but diaspores adhering securely tc the dispersal agent externally (i.e. exozoic transport) is somewhat more difficult to envisage thar endozoic transport of ingested seed. STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 189 Fruit of S$. aloides ripens late, (end of September to end of October), at a time when birds are migrating. Fresh seeds of S. aloides are heavily clothed with mucilaginous hairs and are liberated in a gelatinous mass by longitudinal splitting of the fruit (Cook & Urmi-K6nig 1983), which suggests the possibility that they might stick to feet or feathers of birds feeding among fertile fruiting plants. The exotesta which bears the mucilaginous hairs is however shed, shortly after the seed is released from the fruit (Cook 1987), and in any case the mucilage tends to be sticky when dry, or almost so, but non-sticky when wet (Sculthorpe 1967). However, while S. aloides is said to attract a range of herbivores including water snails, beetles, insects and “several vertebrates” (Cook & Urmi-Ko6nig 1983), birds are not specifically mentioned as feeding on the plant. This may simply reflect the fact that, at present, seed is rare on Stratiotes plants anywhere in Europe. Similar dispersal circumstances attend other non- or seldom-fruiting aquatic macrophytes, including several that are commonly associated with Stratiotes in Britain and Ireland and generally considered native. Other native aquatic species of the same community freely set seed, yet appear to remain poorly adapted for long-range jump dispersal as their seeds are totally digested when eaten by birds and fish, e.g. Nymphaea alba (White Water-lily) and Nuphar lutea (Yellow Water- lily) (Preston & Croft 1997). All long-range or jump-dispersal events are rare and unlikely and as such are difficult to envisage (Carlquist 1974; Cook 1985). Birds, particularly waterfowl, are often cited as agents of dispersal. Because of their rarity, natural transport events are almost impossible to observe and, for the same reason, they are not readily open to experiment. In the great majority of cases we do not know how or when the dispersal event(s) happen, but we have indisputable evidence that they do occur. GENETIC DIVERSITY As noted by Webb (1985), this is one area where data for most plants are very rarely available. I doubt that there are any relevant data on S. aloides apart from comments made by Cook & Urmi- Konig (1983). These authors undertook in their revision of S. aloides to rid the accumulated literature on it of contradictory observations and views, by reviewing the species on the basis of their own direct experience of fresh and fixed material. After their study they concluded that they had found no genetic or ecotypic variability worthy of formal taxonomic recognition. Three features they mentioned were: a. minor differences existed with respect to seed sizes between Swiss and Czechoslovakian plants, b. British specimens were somewhat stouter than Continental and Scandinavian ones, yet fell within the overall pattern of variation for the species, c. totally submerged plants with more flaccid, stomata-less leaves up to 100 cm long were more common in the north of the range of the species. The reproductive pattern and current behaviour of S. aloides suggest that a small or very small founder population might well be all that the species succeeded in dispersing across the North Sea and the Irish Sea, so that plants on each island may represent one clone or a small range of clones. Comparative genetic investigation using isozymes is required to answer the questions posed by this clonal hypothesis. COMMUNITY ECOLOGY The Middle Pleistocene fossil assemblage in East Anglian sediments described by West & Wilson (1968) showed S. aloides occurring with Potamogeton crispus (Curled Pondweed), Cladium mariscus (Sword Sedge), Hydrocharis morsus-ranae and Lemna minor (Common Duckweed). All these fossil allies occur today with S. aloides in Upper Lough Erne, Co. Fermanagh, together with numerous other frequently associated aquatics, e.g., Lemna trisulca (Ivy-leaved Duckweed), 190 R. S: FORBES Sagittaria sagittifolia (Arrowhead), and several narrow-leaved Potomogeton species typical of more shallow waters, such as P. pusillus (Lesser Pondweed), and P. pectinatus (Fennel Pondweed). In phytosociology, communities are drawn floristically by species presence, plus a particular set of frequency and abundance values of all the species found in the samples or relevés. A sample may be pigeon-holed into a previously described community even if the name-giving species, or some of the other characteristic “constant species”, are absent. The information available from Upper Lough Erne forms part of the substantial Fermanagh Flora database and consists, not of standardised quadrats or relevé samples, but of site specific species lists which do not incorporate frequency and abundance values. It is possible, however, to glean an idea of species frequency from the overall species statistics of records and tetrads occupied, and I present these estimates of species presence here (Table 1). The National Vegetation Classification survey of aquatic vegetation in Great Britain (Rodwell 1995) shows S. aloides present in two plant associations, prominently as a characteristic constant species in the Hydrocharis morsus-ranae—Stratiotes aloides community (A4) (Table 1), and as a rare companion species in the Potamogeton pectinatus—Myriophyllum spicatum community (A11). The Aydrocharis morsus-ranae—Stratiotes aloides community is equivalent to the continental TABLE 1: FLORISTIC TABLE FOR THE HYDROCHARIS MORSUS-RANAE-STRATIOTES ALOIDES COMMUNITY (A4) OF RODWELL (1995) Norfolk Broads Fermanagh Species Frequency Cover value No. recs. No. of tetrads Hydrocharis morsus-ranae V 1-4 78 19 Stratiotes aloides V 1-7 4- 12 Lemna minor V 1-3 720 50+ L. trisulca V 1-5 487 50+ Callitriche platycarpa V 1-3 13 5 Polygonum amphibium IV 1-3 370 47+ Utricularia vulgaris IV 1-7 98 30+ Ceratophyllum demersum IV 1-7 0 0 Myriophyllum verticillatum Il 1-3 26 6 Nuphar lutea Il 1-2 619 50+ Potamogeton obtusifolius Il 1-6 240 50+ Elodea canadensis Il 1-4 604 50+ Potentilla palustris II 1-2 674 50+ Hottonia palustris I 1-2 0 0 Berula erecta I ] 23 6 Ceratophyllum submersum U 1-6 0 0 Oenanthe aquatica Ul 1-3 126 2a, Sparganium erectum I 1-3 613 50+ Nasturtium officinale I 1-3 432 35+ Sium latifolium I 1-3 Sed, 50+ Alisma plantago-aquatica I 1-3 708 50+ Nymphaea alba i 1-3 185 18 Ranunculus lingua I 1-3 181 35+ Sagittaria sagittifolia I 1-3 87 28 Potamogeton crispus I 1 98 18 Sparganium emersum I 3-4 267 50+ Number of samples 12 Number of species/sample 14 (11-19) Based on data (columns | & 2) from the Norfolk Broads (Wheeler & Giller 1982), compared with data from the Co. Fermanagh Flora database (unpub.). Column 3 lists the total number of records for the species in the database, and column 4 indicates the number of tetrads in around Upper Lough Erne in which it occurs. STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 191 Hydrochario—Stratiotetum (Westhoff & den Held 1969) or the Hydrocharitetum of Oberdorfer (1977). Rodwell admits that he has characterised this community in Great Britain almost entirely on what he refers to as, “the small data set of Wheeler & Giller (1982)” from the Norfolk Broads, and he recognises that more comprehensive sampling is needed to arrive at a satisfactory definition of this kind of assemblage in Britain. Keeping this reservation in mind, examination of the published floristic table for the Hydrocharis morsus-ranae-Stratiotes aloides community demonstrates that in addition to the two eponymous aquatics, there are 24 additional species listed, of which 21 occur in Upper Lough Erne (Table 1). The only plants missing from Upper Lough Erne are the two Hornworts, Ceratophyllum demersum and C. submersum (Soft Hornwort), plus Hottonia palustris (Water Violet). Of these three, C. demersum alone has a prominent role as a “constant species” in the plant community, but the other seven “constant” species are all present. The fact that S. aloides occurs in Fermanagh backed up by a very nearly complete complement of the characteristic species of the Ceratophyllum—Stratiotes nodum (Wheeler & Giller 1982), or the Hydrocharis morsus-ranae—Stratiotes aloides community of Rodwell (1995) - and this despite the depauperate flora of Ireland - must lend very strong circumstantial support to the claim for Water-soldier’s native status in Upper Lough Erne. With regard to the second community, in which Stratiotes plays a much more minor role, the Potamogeton pectinatus—Myriophyllum spicatum community (All) of Rodwell (1995), Upper Lough Erne supports 47 of the 56 characteristic species listed by Rodwell for the three sub- communities he recognises (neglecting the moss Fontinalis antipyretica). The nine missing species include five which are nationally scarce in Great Britain, two that are strictly coastal, and two which have never been recorded anywhere in Ireland. Thus we can say with confidence that Stratiotes occurs in Upper Lough Erne with very much the same vascular plant species complement that accompanies it in Great Britain and, importantly, that in turn these match Continental equivalents. INSECT ASSOCIATES Preston (1986) suggested an additional criterion for assessing native status of plants to the eight discussed by Webb (1985), namely, the relationship of the plant species in question to oligophagous herbivorous insects. Entomological evidence, Preston indicated, needs to be used with the same degree of caution as Webb’s criteria, for introduced plant species may attract insect feeders from related long recognised native species, as is known in the case of Impatiens noli- tangere (Touch-me-not Balsam), which supplies oligophagous insects to the definitely introduced garden escape /. parviflora (Small Balsam) when the two species are adjacent. With regard to S. aloides, the absence of the dragonfly Aeshna viridis Eversmann, which is described as dependent on the macrophyte, led Cook & Urmi-K6nig (1983) to argue that S. aloides might not be native in Great Britain and other areas of Europe from which the insect is missing. These authors report the findings of Miinchberg (1956), who researched the habits and life history of the insect: it is carnivorous and, although the larvae live on S. aloides rosettes for two or three years until they reach maturity, they do not appear to harm the leaves of the host plant at all. Knowing this, it is not at all clear why the dragonfly should be specifically associated with S. aloides. Higler (1977) found over 200 invertebrate species associated with S. aloides in the Dutch broads, but apart from a few Lepidopteran caterpillars, which did insignificant damage, none of the other species appeared to feed on the living plant tissues - they merely used the plant as a habitat surface or shelter. The argument linking Aeshna viridis to the native status of S. aloides is further weakened by the fact that the distribution of the dragonfly only partially overlaps that of S. aloides, and the dragonfly is absent from N.E. Italy and the mouth of the Danube where Cook & Urmi-K6nig believe S. aloides is likely to be native (Cook & Urmi-K6nig 1983). Another dragonfly, Anaciaeshna isosceles (Miiller) (Norfolk Hawker), a protected, endangered species, is restricted in Great Britain to the Broadlands and there heavily dependent on S. aloides, again apparently purely for shelter, although it is also found to a lesser extent on Glyceria maxima (Reed Sweet-grass), Juncus effusus (Soft-rush) and Iris pseudacorus (Yellow Iris) (Leyshon & Moore 1993). This dragonfly was found in the examined Suffolk Broadland dykes only where a dense population of S. aloides occurred. 192 R..S. FORBES EASE AND SOURCE OF KNOWN NATURALIZATION GREAT BRITAIN S. aloides was recommended by Philip Miller as a suitable plant for water gardens in his famous Gardener’s dictionary, first published in 1724 and with many subsequent revisions and abridged editions. Until Loudon’s An encyclopaedia of gardening appeared in 1822, Miller’s publication was the most widely used and influential garden handbook and his ideas became fashionable. William Robinson’s The English flower garden (first published in 1883 and which also ran to many subsequent editions), continued the fashion, although he was not impressed with Stratiotes, “spike of unattractive blossoms”. A recently published manual of water gardening sponsored by the Royal Horticultural Society (Robinson 1997) continues to recommend S. aloides as a hardy submerged aquatic species for garden pool decoration. The map of the British Isles distribution of S. aloides in Preston & Croft (1997) and the revised version published here (Fig. 2) provide a convenient summary of current knowledge of the distribution of the species, the extent to which it has become naturalised in Great Britain, and its decline this century. (The coverage of Preston & Croft’s map gives a very much more complete and detailed picture of the status of S. aloides than the Atlas map (Perring & Walters 1976), having information plotted for 216 squares compared with just 93). It requires only a cursory glance to appreciate that only a small minority (16%) of the post-1969 symbols represent native occurrences, the balance being records of introduced populations. The revised map (Fig. 2) contains four classes of record symbols, 25% representing post-1969 native occurrences. The introductions extend north to Brechin (v.c. 90), south to Portsmouth (v.c. 11), and west to Bideford (v.c. 4) but are, as one might expect, centred on the major conurbations of the English midlands. The other impression quickly gained from Fig. 2 1s the extent to which S. aloides as a species (irrespective of the status of the particular station) is declining in the Britain Isles. Even after revising the status of the Fermanagh and Cavan records, the number of native symbols on the map has dropped from a total of 75 10-km squares pre-1950 to just 23 today, while in areas where it was introduced, the decline has only been from 149 squares pre-1950 to 69 squares after that date. IRELAND In Co. Fermanagh the main garden activities in the late 18th and early 19th centuries were still landscaping with trees, and the less formal, more natural Rococo style of gardening initiated by Alexander Pope, his friend Dean Jonathan Swift and his friend, the Dean of Down, Patrick Delaney. The Rococo style involved vistas of trees grown in a natural manner, rather than the clipped dress of the formal baroque garden, and it had subsidiary vistas in the form of meanders or angles, the whole effect being intricate, whimsical and natural (Lamb & Bowe 1995). It is known that Delaney advised on garden improvements in Co. Fermanagh, certainly at Belleisle near Enniskillen, and perhaps also at Manor Waterhouse (Malins & the Knight of Glin 1976). The Manor Waterhouse estate (H3730), a few km south of Lisnaskea and five km north of the Crom estate, belonged to an Englishman, John Madden (1652-1703). Madden and his son Samuel were interested in plants (Nelson 1981), and gardening (including the use of water features); however, there are no records of S. aloides from Manor Waterhouse and no evidence suggesting that the Maddens knew the species, or that they had any hand in introducing it to the county. The grounds of Crom Castle, where (or close to where) S. aloides was first discovered in 1805, at that particular time were probably extremely neglected from the gardening point of view, since Crom Castle had been destroyed by a major fire in 1764, and was not rebuilt until 1834. The grounds around the castle were then laid out in the new classical Landscape style, complete with false ruins and a picturesque cottage or garden house on nearby islands in Upper Lough Erne. Miller’s Gardener’s dictionary (1724 and subsequent editions) had many Irish subscribers, and three Dublin editions were published, in 1732, 1741 and 1764 (Nelson 1980). Although Miller recommended S. aloides for use in garden water features, | cannot unearth any evidence that the plant was ever fashionable or widely available in Ireland. The Garden History Society has published The availability of hardy plants of the late eighteenth century (Harvey 1988), in which there is no mention of either S. aloides, or its often associated relative, Hydrocharis morsus-ranae, indicating that S. aloides was not available for purchase in the horticulture trade immediately prior to its discovery in Ireland. STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 193 S. aloides was first found associated with or close to landed estates e.g., Crom and Castle Saunderson. Does this not suggest a cultivated origin? Such estates were sited in places that were visually interesting and strategically important and, being large enough, often contained relatively undisturbed habitats that were likely to contain interesting plants. Educated people lived in these surroundings and they used the lakes for sailing and pleasure boating and were likely to notice unusual plants. S. aloides has an unusual life-cycle, rising and sinking as the year progresses, and we can surmise that estate residents would comment on it to each other and to interested visitors. Botanists like Scott and Mackay, while not directly involved in the life of the grand houses, through their knowledge of plants and gardens, had access to the estates, their owners and gardeners, a tradition of hospitality amongst plantspeople which happily continues to the present day. The botanists would thus gain knowledge of local wild plants from the residents of the large country houses, and very many of the older plant records in the Fermanagh Flora database are associated with the names of old estates, e.g., Florencecourt, Belleisle, Crom and Castle Coole (R. H. Northridge pers. comm., 1998). If, contrary to published information (Harvey 1988), sources of S. aloides were widely available and it was used for decorative effect in water gardens throughout Ireland for a period, why, if it readily escapes (as it appears to do in Great Britain), have there been no further reports of it in the wild in Ireland during the last 160 years? CONCLUSIONS The only incontrovertible proof of native status for S$. aloides in Co. Fermanagh, pre-Neolithic macrofossil material, is absent so far - here and elsewhere in the British Isles. Therefore we must weigh up the circumstantial evidence offered here to decide if it is sufficient to allow us to say, as Meikle did in 1957 (unpub. ms.), that S$. aloides is as indigenous to Co. Fermanagh as Hydrocharis morsus-ranae, or any other aquatic plant species. The major points in favour of an indigenous origin for S. aloides in Fermanagh are: 1 Early discovery (in an Irish context at least), by a reputable and very experienced botanist who knew the locality intimately and could therefore have judged the likelihood of garden origin, yet did not argue for it, or comment on it as a possibility. 2 The rare and scattered occurrence of records and the lack of spread of the species when compared to other known aquatic introductions. 3. The total lack of new stations for the species for well over 100 years, despite the recording efforts of a growing body of botanical recorders, suggests either that S. aloides is not widely grown in Ireland, or if it is, it does not escape or spread. 4 §. aloides was not available in the horticulture trade in the late 18th century and no information suggests that it ever became a popular or fashionable garden plant in Ireland. 5 Persistence for almost 200 years in a natural watercourse (i.e. Upper Lough Erne) compared with its relatively rapid disappearance at sites where introduction is either known (ie. Ballyphehane Bog), or possible. 6 The remarkable degree of similarity in the range of associated species when compared to the vegetation described from the Norfolk Broads, the most likely native station in Great Britain. 7 The existence in Upper Lough Erne of accepted native aquatic plant species, some rare, some with disjunct distributions, and some which also currently lack seed dispersal. 8 The habitat in Upper Lough Erne fits the habitat typical for S. aloides everywhere in Europe, namely slow-flowing meso-eutrophic water. 194 R. S. FORBES The main evidence suggesting S. aloides could be an introduction is: 1 Miller’s Dictionary of Gardening (1724 & subsequent editions) recommended S. aloides to gardeners creating water features. Early 19th century occurrences of the species in a few widely scattered stations in Ireland, including canals, flax dams and other artificial waterways, might imply some level of cultivation and subsequent escape. 2 Some of these possibly escaped S. aloides colonies persisted for up to 120 years before dying out and could therefore appear naturalised for a time. 3S. aloides appears to lack an obvious means of long-range “jump” dispersal over land, and it is thus difficult to account for its presence anywhere in the British Isles, and more so on the offshore and more ancient island of Ireland. 4 The restricted distribution of its extant Irish populations, especially its absence from other watercourses in drumlin regions (and even from the lower parts of the River Erne catchment). However, as discussed earlier, the habitat requirements and tolerances of S. aloides are very specific and narrow, and locations that /ook appropriate may not provide suitable sites (e.g., the physiography and water chemistry of Lower Lough Erne differs considerably from the Upper Lough). I believe that, on balance, the circumstantial evidence assembled here strongly supports Meikle’s assertion that S$. aloides has as much right to indigenous status in Ireland as Hydrocharis or any other aquatic species. Like Webb (1985), I feel sure that there are other cases where the current status of a plant species is heavily dependent on received wisdom rather than reasoned debate over assembled facts. Conservation arguments for sites frequently revolve around the occurrence of rare and declining indigenous species, making the question of status much more than a purely academic matter. ACKNOWLEDGMENTS I especially thank Robert Northridge, my partner as joint Recorder for Co. Fermanagh (H33), for over 20 years of enjoyment during field work, and for persistence in the creation of the Fermanagh Flora database which forms the basis of this work. Drs Damian McFerran and Julia Nunn at the Centre for Environmental Data and Recording (CEDaR), Ulster Museum, Belfast assisted with the creation and maintenance of the database. Prof. Christopher Cook provided a copy of his published map of European distribution and kindly permitted me to modify it (Fig. 1). Drs Chris Preston and Jane Croft at BRC provided the revised British Isles distribution (Fig. 2). The Co. Fermanagh distribution (Fig. 3) was drawn using Dr Stuart Ball’s RECORDER and Dr Alan Morton’s DMAPW software. I also thank colleagues who offered information or responded to enquiries: Joan Bullock, R. J. Driscoll, Gwynn Ellis, Dr Charles Nelson, Michael Walpole, Shaun Wolfe- Murphy and B.S.B.I. v.c. Recorders, Maura Scannell (H3), Patrick Reilly (H30), Alec Bull (27), Jean Green (50) and Graeme Kay (58). Dr Valerie Hall (Queen’s University, Belfast) provided helpful information and assistance. Finally, my wife Deirdre read the paper and encouraged its submission. REFERENCES ANDREW, R. (1970). The Cambridge pollen reference collection, in WALKER, D. & WEST, R. G., eds. Studies in the vegetational history of the British Isles. Cambridge University Press, Cambridge. BAKER, H. G. (1955). Self-compatibility and establishment after ‘long-distance’ dispersal. Evolution 9: 347— 349, BAKER, H. G. & Cox, P. A. (1984). Further thoughts on islands and dioecism. Annals of the Missouri Botanical Garden 71: 230-239. BAWA, K. S. (1982). Outcrossing and the incidence of dioecism in island floras. American naturalist 119: 866-871. CARLQUIST, S. (1974). Island biology. Columbia University Press, New York. CHANDLER, M. E. J. (1923). The geological history of the genus Stratiotes: an account of the evolutionary changes which have occurred within the genus during Tertiary and Quaternary times. Quarterly journal of the Geological Society 79:117-138. STATUS OF STRATIOTES ALOIDES IN CO. FERMANAGH 195 CLEMENT, E. J. & FOSTER, M. C. (1994). Alien plants of the British Isles. Botanical Society of the British Isles, London. COLGAN, N. & SCULLY, R. W. (1898). Cybele Hibernica, 2nd. ed. Ponsonby, Dublin. COLHOUN, E. A. & MITCHELL, G. F. (1971). Interglacial Marine formation and Lateglacial Freshwater formation in Shortalstown Townland, Co. Wexford. Proceedings of the Royal Irish Academy 71 (B): 211-245. Cook, C. D. K. (1985). Range extensions of aquatic vascular plant species. Journal of aquatic plant management 23:1-6. Cook, C. D. K. (1987). Dispersal in aquatic and amphibious vascular plants, in CRAWFORD, R. M. M., ed. Plant life in aquatic and amphibious habitats. British Ecological Society Special Publication No. 5, pp. 179-190. Blackwell, Oxford. Cook, C. D. K. & LUOND, R. (1982). A revision of the genus Hydrocharis (Hydrocharitaceae). Aquatic botany 14: 177-204. Cook, C. D. K. & URMI-KONIG, K. (1983). A revision of the genus Stratiotes (Hydrocharitaceae). Aquatic botany 16: 213-249. Cox, P. A. (1985). Islands and dioecism: Insights from the reproductive ecology of Pandanus tectorius in Polynesia, in WHITE, J., ed. Studies on plant demography: a Festschrift for John L. Harper, pp. 359-372. Academic Press, London. CRACKLES, E. (1982). Stratiotes aloides L. in the East Riding of Yorkshire. Naturalist (Leeds) 107: 99-101. CRACKLES, E. (1990). Flora of the East Riding of Yorkshire. Hull University Press, Hull. DAUMANN, E. (1931). Zur Morphologie und Oekologie der Bliite von Stratiotes aloides L. Planta 4: 766-776. DE TABLEY (Lord) (1899). The Flora of Cheshire. Longmans, Green & Co., London. DICKSON, C. A. (1970) The study of plant macrofossils in British Quaternary deposits, in WALKER, D. & WEST, R. G., eds. Studies in the vegetational history of the British Isles. Cambridge University Press, Cambridge. ELLIS, R. G. (1983). Flowering plants of Wales. National Museum of Wales, Cardiff. FISHER, N. (1930). Water-Soldier, Stratiotes aloides, L., at Woodburn. Irish naturalists’ journal 3: 112. GELDART, A. M. (1906). On Stratiotes aloides, L. Transactions of the Norfolk & Norwich Naturalists’ Society 8: 181-200. GERARDE, J. (1633). The Herball or Generall Historie of Plantes, 2nd ed., revised by T. Johnson. London. GIBBONS, E. J. (1975). The Flora of Lincolnshire. Lincolnshire Naturalists’ Union, Lincoln. Gopwin, H. (1975). The history of the British Flora, 2nd ed. Cambridge University Press, Cambridge. HACKNEY, P., ed. (1992). Stewart & Corry’s Flora of the North-East of Ireland, 3rd ed. Institute of Irish Studies, The Queen’s University of Belfast, Belfast. HARVEY, J. H. (1988). The availability of hardy plants of the late Eighteenth Century. The Garden History Society, sine loc. HIGLER, L. W. G. (1977). Macrofauna-cenoses on Stratiotes plants in Dutch broads. Verhandelingen Rijksinstituut voor Natuurbeheer 11: 1-86. HYDE, H. A. & WADE, A. E. (1957). Welsh flowering plants, 2nd ed. National Museum of Wales, Cardiff. KATZ, N. J., KATZ, S. V. & KIPIANI, M. G. (1965). Atlas and keys of fruits and seeds occurring in the Quaternary deposits of the U.S.S.R. Publishing House Nauk, Moscow. KORNATOWSKY, J. (1979). Turions and offsets of Stratiotes aloides L. Acta Hydrobiologica 21: 183-204. LAMB, K. & BOWE, P. (1995). A history of gardening in Ireland, Stationary Office, Dublin. LEYSHON, O. J. & Moore, N. W. (1993). A note on the British Dragonfly Society’s survey of Anaciaeschna isosceles at Castle Marshes, Barnby, Suffolk, 1991-1992. Journal of the British Dragonfly Society 9: 5— zh LOUDON, J. C. (1822). An encyclopaedia of gardening. Longman, Hurst, Rees, Orme and Brown, London MACKAY, J. T. (1806). A systematic catalogue of rare plants found in Ireland. Transactions of the Royal Dublin Society 5: 127-184. MACKAY, J. T. (1825). Catalogue of the indigenous plants of Ireland. Transactions of the Royal Irish Academy, B, 14: 103-198. MACKAY, J. T. (1836). Flora Hibernica. William Curry jun. & Co., Dublin. MALINS, E. & THE KNIGHT OF GLIN (1976). Lost Demesnes: Irish landscape gardening 1660-1845. Barrie & Jenkins, London. MATTHEWS, J. R. (1955). Origin and distribution of the British Flora. Hutchinson, London. MEAGHER, T. R. (1988). Sex determination in plants, in LovETT DousT, J & LOVETT Doust, L., eds. Plant reproductive ecology: patterns and strategies, pp. 125-138. Oxford University Press, Oxford. MILLER, P. (1732). Gardener’s dictionary. London. MITCHELL, G. F. & RYAN, M. (1997). Reading the Irish landscape. Town House, Dublin. Moore, D. & Mork, A. G. (1866). Cybele Hibernica, 1st ed. Hodges, Smith & Co., Dublin. MUNCHBERG, P. (1956). Zur Bindung der Libelle Aeschna viridis Eversm. an die Pflanze Stratiotes aloides L. (Odon.) Nachrichtenblatt der Bayerischen Entomologen 5: 113-118. 196 R. S. FORBES NELSON, E. C. (1980). Works of botanical interest published before 1800 held in Irish libraries. Occasional papers No.1: National Botanic Gardens, Glasnevin, Dublin. NELSON, E. C. (1981). A late 17th century Irish herbarium in the library of Trinity College, Dublin. /rish naturalists’ journal 20: 334-335. NELSON, E. C. & BRADY, A. (1979). Irish gardening and horticulture. Royal Horticultural Society of Ireland, Dublin. NEWTON, A. (1971). Flora of Cheshire. Cheshire Community Council, Chester. OBERDOREER, E. (1977). Siiddeutsche Plfanzengesellschaften. Teil 1. Fischer, Stuttgart. PERRING, F. H. & SELL, P. D. (1968). Critical supplement to the Atlas of the British flora. Nelson, London. PERRING, F. H. & WALTERS, S. M., eds. (1962). Atlas of the British flora. Nelson, for Botanical Society of the British Isles, London. PERRING, F. H. & WALTERs, S. M., eds. (1976). Atlas of the British flora, 2nd ed. EP publishing, Wakefield. PRAEGER, R. L. (1938). The flora of the North-east of Ireland, 2nd ed. Quota Press, Belfast. PRAEGER, R. L. (1949). Some Irish naturalists. Dundalgan Press, Dundalk. PRESTON, C. D. (1986). An additional criterion for assessing native status. Watsonia 16: 83. PRESTON, C. D. & MARCH, M. D. (1996). Hydrocharis morsus-ranae L. (Hydrocharitaceae) fruited in Britain in 1995. Watsonia 21: 206-208. PRESTON, C. D. & CROFT, J. M. (1997). Aquatic plants in Britain and Ireland. Harley, Colchester. PRESTON, C. D. & HILL, M. O. (1997). The geographical relationships of British and Irish vascular plants. Botanical journal of the Linnean Society 124: 1-120. Proctor, M. C. F., YEO, P. F. & LACK, A. J. (1996). The natural history of pollination. Harper Collins, London. RENMAN, G. (1989). Life histories of two clonal populations of Stratiotes aloides L. Hydrobiologia 185: 211- Doe) RICHARDS, A. J. (1986). Plant breeding systems. Allen & Unwin, London. ROBINSON, P. (1997). RHS Water Gardening. Dorling Kindersley, London. ROBINSON, W. (1883). The English flower garden and home grounds. John Murray, London. RODWELL, J. S., ed. (1995). British plant communities 4: Aquatic communities, swamps and tall-herb fens. Cambridge University Press, Cambridge. SCANNELL, M. J. P. (1976). Ceratophyllum demersum L. and fruit performance. /rish naturalists’ journal 18: 348-349. SCANNELL, M. J. P. & SYNNOTT, D. M. (1987). Census catalogue of the Flora of Ireland. Stationary Office, Dublin. SCULTHORPE, C. D. (1967). The biology of aquatic vascular plants. Edward Arnold, London. SIMPSON, D. A. (1984). A short history of the introduction and spread of Elodea Michx in the British Isles. Watsonia 15: 1-9. SMOLDERS, A., VAN DUYNHOVEN, A. H. N. & ROELOTS, J. G. M. (1993). Vruchzetting en zaadproductie van Krabbescheer ( Stratiotes aloides L. ) in Nederland. Gorteria 19: 55-61. SMOLDERS, A. J. P., ROELOFS, J. G. M. & DEN HARTOG, C. (1996). Possible causes for the decline of the water soldier (Stratiotes aloides L.) in the Netherlands. Archiv fur Hydrobiologie 136: 327-342. STACE, C. A. (1997). New flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. STEBBINS, G. L. (1957). Self-fertilization and variability in the higher plants. American naturalist 41: 337- 354. STEWART, A., PEARMAN, D. A. & PRESTON, C. D. (1994). Scarce plants in Britain. Joint Nature Conservation Committee, Peterborough. VALENTINE, D. H. (1946). The Butterbur in Yorkshire. Naturalist 1946: 45-46. VALENTINE, D. H. (1947). The distribution of the sexes in Butterbur. N.W. Naturalist 1947: 111-114. WATSON, H. C. (1883). Topographical Botany, 2nd ed. Quaritch, London. WEBB, D. A. (1985). What are the criteria for presuming native status? Watsonia 15: 231-236. WEBB, D. A., PARNELL, J. & DOOGUE, D. (1996). An Irish Flora, 7th ed. Dundalgan Press, Dundalk. WEST, R. G. (1961). Vegetational history, of the early Pleistocene of the Royal Society borehole at Ludham, Norfolk. Proceedings of the Royal Society B, 155: 437-453. WEST, R. G. & WILSON, D. G. (1968). Plant remains from the Corton Beds, Lowestoft, Suffolk. Geological magazine 105: 116-123. WESTHOFF, V. & DEN HELD, A. J. (1969). Plantengemeenschappen in Nederland. Thieme, Zutphen. WHEELER, B. D. & GILLER, K .E. (1982). Status of aquatic macrophytes in an undrained area of fen in the Norfolk Broads, England. Aquatic botany 12: 277-296. WILLKOMM, M. & LANGE, J. (1861). Prodromos Flora Hispanicae. E. Schweizerbart, Stuttgart. (Accepted March 1999) Watsonia 23: 197—208 (2000) 197 The distribution, ecology and conservation of Arenaria norvegica subsp. anglica Halliday (Caryophyllaceae) K. J. WALKER LT.E., Monks Wood, Abbots Ripton, Huntingdon, Cambridgeshire, PE17 2LS ABSTRACT Arenaria norvegica subsp. anglica is endemic to the Craven area of Mid-west Yorkshire (v.c. 64) where it grows exclusively amongst limestone fragments (clitter) and in shallow solution hollows with a thin covering of soil on level or slightly inclined limestone exposures. Since its discovery at Ribblehead in 1889 over 20 populations have been found, all but one of which occurs on the eastern slopes of Ingleborough. Over 70% of the entire population, of around 2000-3000 plants, occurs within two adjacent pastures where numbers are apparently stable. However, subsp. anglica has declined on tracks and bridleways where it was first recorded over 100 years ago. Although colonies within pastures appear to have fared better, their frequent small size (less than 50 plants) makes populations susceptible to drought and disturbance. The current rarity of subsp. anglica owes more to climate change (contraction of habitat) and aspects of its autecology (limited seed production and dispersal) rather than human activities. However, grazing and recreational disturbance still threaten some populations. As a result conservation measures are being introduced in order to reduce grazing levels and protect sensitive trackside populations. In addition, the creation of “safesites” and collection of seeds for genebanking will ensure its ultimate survival. KEYWORDS: taxonomy, population size, rarity, habitat management, Yorkshire. INTRODUCTION Arenaria norvegica Gunn., Arctic Sandwort (Caryophyllaceae, Subfamily Alsinoideae), is an arctic-montane therophyte which occurs in Iceland, Norway, Sweden, Finland, Shetland and north- west Scotland. Two subspecies occur in the British Isles (Fig. 1): A. norvegica subsp. norvegica' is a plant of base-rich screes, river shingle and fell-field, confined to a small number of sites in Scotland (Beinn Iadain, v.c. 97; Beinn Sgulaird, v.c. 98; Rum and Eigg, v.c. 104; Inchnadamph, v. c. 108; Unst, v.c. 112) and one somewhat problematical site in the west of Ireland (Gleninagh Mountain, v.c. H9), where it may be extinct (Webb & Scannell 1983). A. norvegica subsp. anglica Halliday, English Sandwort, (Arenaria gothica auct., non Fries hereafter referred to as subsp. anglica) appears to be endemic to England occurring in only two 10 km squares in the Craven region of Mid-west Yorkshire (v.c. 64). This subspecies can be distinguished from subsp. norvegica by its often annual (rarely biennial) habit, somewhat larger flowers, laxer habit, greater ciliation (especially on the base of the sepals) and narrower, more lanceolate leaves (Halliday 1960b). Subsp. anglica is an undoubted Yorkshire speciality which has attracted many generations of botanists to the area, not least because it shares its habitat with a number of other rarities (e.g. Minuartia verna and Sedum villosum). The population in Craven, which is in the order of (1000-) 2000—3000(—4000) plants, is confined to the Carboniferous limestone outcrops on the eastern slopes of Ingleborough (Walker 1995). Remarkably there is only one very small population outside a restricted (5 km square) area, despite the occurrence of suitable habitat elsehwere. As a result it is now recognised as “endangered” (Hodgetts et al. 1996) and afforded full protection under Schedule 8 of the Wildlife and Countryside Act (1981). This paper describes the history of the plant in Britain, in terms of its taxonomy, distribution and Status. Information on aspects of its ecology are also presented, and are discussed in relation to rarity as well as conservation measures proposed within a recent action plan (Walker & Corkhill 1996). ' Nomenclature follows Kent (1992) for vascular plants and Smith (1978) for bryophytes. 198 K. J. WALKER TAXONOMY Arenaria norvegica is included within the Arenaria ciliata L. complex, a polymorphic taxon incorporating a number of closely related montane plants with highly disjunct distributions stretching some 40° from northern Spain to 82° north in the Arctic. The taxonomic history of this complex is very confused, with up to four species and seven geographically distinct subspecies being recognised (Chater & Halliday 1993). Subsp. anglica, in particular, formerly posed a number of taxonomic problems for British botanists. On discovery in July of 1889 it was initially thought to be a form of A. ciliata, although it was later confirmed by F. A. Lees and Professor C. C. Babington as A. norvegica Gunnerus (Rotheray 1889b). At the time this species was known from only two sites in the British Isles; on Unst where it had been known of since 1837 and in West Sutherland on the Scottish mainland where it had been recently discovered (Scott & Palmer 1987). However, a number of botanists who were familiar with the Scottish plants, including J. Gilbert Baker of Kew, commented on the laxer habit, narrower leaves and differing shoot structure of the “Yorkshire sandwort” (Baker 1889). After visiting the Ribblehead populations in September of 1889 both Baker and Lees began to doubt these earlier determinations, particularly Lees who suggested that the Yorkshire plants may represent an “altered form” of A. norvegica (Rotheray 1889b). As a result he sent specimens to William Whitwell and Arthur Bennett who had a small collection of European Arenaria material sent to him by Nilsson in 1882. After comparing these with Lees’ plants and Fries’ original descriptions Bennett concluded that the Ingleborough plant was in fact A. gothica Fries (Rotheray 1889b; Bennett 1891), a very rare plant of dry limestone habitats and lake shores in southern Sweden (V4stergétland), Gotland and the Swiss Jura (Lac de Joux) (Williams 1898; Albertson 1946). The Ribblehead population represented a notable extension of its curiously disjunct European distribution although both Bennett and Baker thought that its true place was with A. norvegica either as a subspecies of A. ciliata or as a variety (Baker 1889; Bennett 1891). Despite these misgivings botanists remained faithful to Bennett’s determination until Halliday’ s study of the A. ciliata complex in the late 1950s (Halliday 1960a). Using a variety of techniques Halliday showed the Yorkshire plants to have a greater affinities to A. norvegica than either the Swedish or Jura A. gothica and remarked that “‘apart from differences in habit, the British plant is closer to A. norvegica in most other respects, in particular in leaf shape, colour, ciliation and inflorescence size” (Halliday 1960a). Furthermore, British plants were diploid (2n = 80; Swedish plants 2n = 100), could interbreed freely and produce fertile seeds whereas seed production © declined in Swedish x English hybrids (Halliday 1958, 1960a, b). On this basis Halliday concluded that the English plant was better placed under A. norvegica and proposed two geographical subspecies; A. norvegica subsp. norvegica and A. norvegica subsp. anglica, “taxa sufficiently distinct morphologically and geographically separated but capable of interbreeding freely and with a common origin” (Halliday 1960a). Subsequent taxonomic work on the A. ciliata complex has supported this division (Wyse-Jackson & Parnell 1987). In this study the two subspecies were shown to be closely related and distinct from all the other European members of the complex, including both A. ciliata s.s. and A. gothica. DISTRIBUTION HISTORICAL RECORDS: 1889-1980 Subsp. anglica was originally discovered by Lister Rotheray at Ribblehead Station, near Ingleborough, on 12 July 1889 where it was growing “in some profusion” amongst limestone “staging” on a recently constructed track (Rotheray undated, 1889a, b; Whitwell 1889). F. A. Lees who visited the site soon after located hundreds of plants, including a new population growing “300 and 400 yards [350 m] away” from the original colony on the opposite side of the railway line (Whitwell 1889). Its occurrence on a man-made track initially raised doubts as to whether the plant was native (Baker 1889; Whitwell 1889; Bennett 1892). However, the use of local stone for track-building led Lees and J. G. Baker to predict its presence on “the neighbouring hills” (Baker 1889; Whitwell 1889). Indeed a small population was discovered on a track at Selside in the ARENARIA NORVEGICA SUBSP. ANGLICA 199 FIGURE 1. The past and present distribution of Arenaria norvegica subsp. norvegica (MM 1970 onwards; A pre- 1970) and subsp. anglica (@ 1970 onwards) in the British Isles. Sites where undetermined fossil seeds of A. ciliata agg. have been found are indicated as follows: A late Devensian/Flandrian, 0) mid Devensian,O early Devensian. following August and as a result botanists began to search more widely (Whitwell 1890). Then in September of 1894 Reginald J. Farrer, the famous plant-collector of Ingleborough Hall, reported its presence at “Sulber Nook” [Nick], in a spot so remote as to confirm its native status (Farrer 1894; Rotheray 1895). Several more populations were found by him in the following year, all of which occurred on or adjacent to the Clapham Lane, a popular bridleway connecting Selside and Crummack Dale (Whitwell 1890, 1895; Rotheray 1895). Farrer communicated these new finds to Rotheray who promptly visited the following August. His account of this trip gives a good indication of the numbers and location of plants as well as describing a number of populations which were undoubtedly new to Farrer (Rotheray undated). An intriguing record was the discovery by Farrer of a small population within the grounds of his own residence, Ingleborough Hall, some 9 km from Ribblehead (Whitwell 1895). Given the numbers of alien plants which have been found in the vicinity in recent years (Abbott 1996) one wonders whether such a keen plant-collector could have moved it there? 200 K. J. WALKER Although Rotheray and Farrer undertook extensive searches of the Craven limestones over subsequent decades no new populations were found until the 1950s when Halliday discovered a small colony close to Farrer’s original Sulber Nick site (Halliday 1960a). A number of other populations were discovered by Bartley & Clark (1978) who were the first to describe the extensive flush population within Sulber pasture (refound in 1993), and the two outlying colonies to the south-east; at Foredale, in a disused quarry, and close by on pavement at Juniper Gill. To date the only verified record away from the eastern slopes of Ingleborough is the small trackside colony at Dawson Close which was discovered by Ruth Kilby during a Yorkshire Naturalists’ Union field excursion in 1952 (Sledge 1952). This site on the northern slopes of Fountains Fell was a notable extension in range, some 6 km to the east of the nearest Ingleborough population. Further afield Lees’ (1937) unconfirmed report of having “picked up two or three specimens from a limestone-metalled road near Healuagh” (Swaledale) in 1906 is conceivable given the transport of Ingleborough limestone for track-building, although this population has never been relocated. Most unlikely is Arthur Bennett’s claim of having found a fragment of A. gothica whilst collecting Potamogeton obtusifolius from Grasmere (Cumbria) given the absence of suitable habitat within the vicinity (Halliday 1960a). CURRENT DISTRIBUTION AND STATUS Since 1981 staff of the Nature Conservancy Council (and latterly English Nature) have carried out a number of detailed population censuses of all extant populations (Table 1). In 1995 all these sites were visited by the author and extensive searches made of the neighbouring limestone pastures (Foley 1995; Walker 1995). The results of these surveys have given a much clearer indication of the plant’s distribution and abundance. However, given the discovery of at least eight new populations over the last half century, one of which is situated some 6 km to the east of Sulber, it is not inconceivable that subsp. anglica may well await discovery elsewhere in Craven. In 1995 over 2700 plants were recorded in approximately 20 colonies within eleven | km squares. Over 70% of the total population (Table 1; populations 2—7 & 8—14) occurred within two adjacent pastures at Sulber on the eastern slopes of the Ingleborough massif (Walker 1995). Here a series of loosely connected populations are confined to shallow soils on limestone exposures, flushes and rough tracks all of which occur on or within 500 m of the Claphman Lane. Elsewhere small populations occur by tracks, at Dawson Close and Selside, in a disused quarry at Foredale and on limestone exposures at Juniper Gill and Thieves’ Moss. Although overall populations sizes are small (mean 143 +43; n = 18) these vary consistently by habitat. Trackside populations are by far the smallest (mean 16 +4; n = 6) and only account for 3% of the total population. As a result these populations are vulnerable to trampling and drought which presumably caused the localised extinction of subsp. anglica at Selside and Sulber Nick in 1992. Populations on gently inclining pavement, which account for 53% of the entire population, are much larger (mean 161 +54; n = 9) and apparently less prone to fluctuations in overall numbers (Table 1). In contrast the two flush colonies at Sulber, which alone account for over 40% of the entire population, show marked fluctuations in numbers from year to year presumably due to the ephemeral water-supply within spring-fed flushes. CHANGE SINCE 1889 The distribution of subsp. anglica has changed little since it was discovered in 1889 and it can still be found in the localities described by Rotheray and Farrer over a hundred years ago. Extinction has occurred in only two sites; in Spring Valley at Ingleborough Hall (where it may have been introduced) and at Ribblehead, whete over-collecting led to its demise within two years of discovery (Whitwell 1890). Rotheray’s accounts and the results of recent surveys suggest that it has also been lost, or declined, from numerous sections of the Clapham Lane, particularly those which traverse Sulber and the adjacent Borrin’s pasture (Rotheray undated; Whitwell 1895). The most likely cause of decline is presumably the increased recreational usage of tracks within the Sulber area over the past century. Since the opening of the National Park in 1953 visitor numbers have increased dramatically, particularly at Ingleborough where there are now around 250,000 visitors annually (S. Rogers, pers. comm., 1995). This is undoubtedly due to the increasing popularity of the Three Peaks footpath which passes directly through Sulber Nick (and a number of populations of subsp. anglica). This figure is likely to increase further when the Clapham Lane is incorporated into the “Pennine Bridleway’, a new long-distance routeway designed to cater for walkers, horse-riders and cycling enthusiasts. 201 ARENARIA NORVEGICA SUBSP. ANGLICA (S661) JOMBM “ET “(HO6T) STOUR “ZI (E661) STOUR “TT “(Z66T) APHOW “OT (0661) FOTARL 6 “(9861) YOSTIM '8 “(T86T) atowayeTg ‘L‘(8L6T) VRID 7% Aopreg “9 (ZS61) 28peTs “¢ “(2O96T) ABPITTPH “p “(S68T) AvroIIoY “¢€ “(parepun) Aeromoy “7 ‘(C68T) TOMY “T ‘SAOUNOS ‘juosaid ..d,, ‘POUSIA JOU ,.AU,, (C661) ABO A “LW + AOGSV “N 29 [[UIOD “g § ‘Aq pakaamng ‘ssaquinu jueyd Jo sayeurnsa uo paseg sans 4. SALON 0 AU ZI JUDUIIAR (UOT]JONPONUT) dsOTD IeIS QT §8 d WUOUIIAR ID Jodrung = LT 8 t¢ vl ¢ AU Oe S¢9 cd yqoes oso) wosMed 91 9g AU 98Z 0 AU OV CGeo? yal Toop Arend) Arend) yepotoy ST OSE 8IP lei El 9ST cS +O0S JUSUOAR swonog weyde[y pT SSL AU 86 69 $9 O6l 00C 2 JUSUTSAC Seq youuu): «Cl Id AU Ot 81 6 SS Se zSIUR[A M9J,, yoerL (1e9g SuOT) suey] weyde,y = Z] cl £1] 0 I : € ¢ yoelL (peol-sso1d) suey weydey{y TI gts AU Ol ¢ Iv GLA SOOT -,,SOU []BUIS,, yoy (a1e8) suey weydeyD =O CSV OC6 8ECl vst [8V cll t+O0E c SOU AJURDS,, ysn[y (YSN[J) YIN J9qINg = —P-46 Cl é 0 0 6 i Q¢o -< SOOM] 2 SOMI,, yorl aue’y weyde[)-YoIN Jaq(ing =e € tg Iv I JUDUIDAR SSO ,SOATIYL 8 ES O8e 8el 16 91t 8rl SOOT . Aqpngnueyd,, JUSUTOAR 10d Foqmng LL L8I AU 09-0S JUSWIOAR | (ISOM) YIN JoqIng 9 6S JUDUIDAV (yorn-daoys) YOIN Joq(ng =¢ tl AU $8 ¢ VC JUSWIOAR (s Aeptf[@H) ASIN Joqing — p 119 AU T6EI aa ysniy (, Aayea,,) aamsed saqung = 3-q¢ VIC 91 8S ¢ 68 OLI Oso d JUSTUOAR ainjsed Jaqing — BE €7 ysn[y (Ysny} MOTaq) ainjsed aqing 7 g 8 ec () () i ogo ,.duepunqe,, yorl (UddID UOJ) OpIsfag | S661 77661 £661 i C66l 0661 S861 +,1861 8L6I SsOSél S068 1 Wwyqey ouleu SHS ON ‘SNOILLV1INdOd VOITONV ‘dSGNS VOIDZAYON VINVNAYV LNV.LXd YOd SGUYOOdA SNSNEO GNV TVOIAOLSIA T ATaVt 202 K. J. WALKER 1000 ¢£ r i S 100 £ c= the A [_ ° i = 1 =F L 5 | 2. S 10 § & r © E = L a0 i = al i= = 0.1 | : 7 80 82 84 386 88 90 92 94 96 Year FIGURE 2. Changes in the mean numbers of plants of Arenaria norvegica subsp. anglica on: a. tracksides (@) (populations 1, 9a, 10, 11, 12 & 16) and b. limestone pavement/flush habitats (MI) (populations 3a, 7, 9b—d, 13 & 14). In contrast, populations on “primary” habitats, such as limestone pavements and flushes appear to be steadily increasing (Fig. 2). The extent to which this is a result of more information or relaxation in grazing pressure is unclear. In the past overgrazing has presumably reduced plant numbers, particularly in the Sulber pastures where census returns have shown a tripling in sheep numbers this century (from around 13,000 to just over 35,000 in 1989; P. Corkhill, pers. comm., 1995). Although stock are important in maintaining open conditions and transporting seed, excessive defoliation and trampling by stock has been observed to cause some mortality, especially around the edge of flushes where grazing is particularly intense (Walker 1995). ECOLOGY HABITAT, SOILS AND CLIMATE Subsp. anglica is a plant of shallow skeletal soils (lithomorphic protorendzinas) on sparsely- vegetated level or slightly inclined limestone exposures. Here plants grow amongst limestone fragments (clitter) and in shallow solution hollows with a thin covering of soil. At one site it also grows directly on tufaceous “crusts” (hydromorphic gleys) which have developed on flat or gently sloping limestone slabs around the edge of flushes and seepage lines. In addition, it has been long established within “ruts” of trackways and amongst shattered limestone on a quarry floor. Recently it has also been observed on disturbed rabbit scrapes (Walker 1995). In all of these habitats soils are shallow (usually <7 cm), circumneutral (mean pH of 7-59 + 0-06 n = 37), with both high levels of calcium (+30 mg Ca 100 g'') and organic material (mean 67% + 2). In most habitats moisture supply is limited and unpredictable and as a result populations are small and sparsely distributed. In contrast, where water-supply is more predictable, i.e. around the edges of flushes and springs, populations are larger and numbers less variable from year to year. The altitudinal range of populations is between 310-390 m (see Appendix). Here the climate is distinctly oceanic with rainfall being uniformly high (average of 1590 mm yr’; 1981-95) tending towards a spring minimum and autumn maximum. At this altitude the growing seasor (accumulated temperatures >6°C) lasts for 5-6 months (April to September) and the distribution o the plant appears to be limited to areas where the oceanic influence ensures that the summe temperatures do not exceed 23°C (Dahl 1951). ARENARIA NORVEGICA SUBSP. ANGLICA 203 TABLE 2. SPECIES ASSOCIATED WITH ARENARIA NORVEGICA SUBSP. ANGLICA IN NON-FLUSHED, FLUSHED AND DISTURBED SITES. Species Frequency Mean Non-flushedt Flushed (n Disturbed (n = 43) abundance (n= 2) = 19) (n:=:3) Arenaria norvegica subsp. anglica 43 2-2 V(1-3) V(1-4) V(1-3) Festuca ovina agg. 42 5-1 V(3-7) V(2-8) V (5-7) Linum catharticum 42 1-3 IV(i-3) V(i-3) IV(2) Sesleria caerulea 34 2:5 V(i-5) V(1i-7) (3) Minuartia verna 33 1-8 ~ V(i-4) IV(i-3) (2) Ctenidium moliuscum p24 | 1-4 IV(1-5) Nidi-4) IV(1) Thymus polytrichus subsp. britannicus 26 1-3 IV(i-4) lidi+) V(5) Carex panicea 25 7 Ii(1-2) V(i-5) V(4-6) Briza media 22. 1-0 Ili(i—3} Tii¢di-3) V(i-4) Tortella tortuosa Ze jb IV(1-7) Ii(¥i-4) Ii(1) Agrostis capillaris 16 1-1 Til (i-4) iii) H(A) Koeleria macrantha 15 1-0 Wd) Hi(3-5) Euphrasia officinalis s.1. 14 <1-0 I(1-3) (1) TV (i-2) Carex caryophyllea 10 <1-0 Iii (1-4) I1(6) Plantago lanceolata 10 <1-0 I(1) Ii(1-2) V (2-4) Cerastium fontanum 9 <1-0 II(1-3) I(i-2) li(1) Schistidium apocarpum 9 <1-0 Il(i-4) IiI(i-4) Juncus articulatus 9 <1-0 I(3) II(i—2) Prunella vulgaris 8 <1-0 1h) I(1) V(2-3) Taraxacum officinale s.1. 8 <1-0 (1) IV(1) Hypnum cupressiforme i, <1-0 Iq) Sedum villosum q| <1-0 I(t) I(1-2) II(1) Potentilla erecta 6 <1-0 II(i-3) I(1) I(5) Sagina nodosa 6 <1-0 I(1-3) (1) (2) Viola riviniana 6 <1-0 I(1) 1(4) I(t) Bellis perennis 5) <1-0 I(1) I(1) V(i-2) Carex viridula subsp. oedocarpa 5 <1-0 II(2-5) Carex flacca 5 <1-0 I(1-3) I(5) IV(3) Carex pulicaris 4 <1-0 I(1-2) I(1) II(1) Cratoneuron commuiatum 4 <1-0 Ii¢i-7) Ranunculus bulbosus + <1-0 I(1) V(i-2) Agrostis canina 3 <1-0 i(2-3) Poa annua 3 <1-0 I(1) I(3) IV(i-2) Bare soil 39 5-6 V(1-8) V(4-9) V(5) Bare limestone 36 4.9 V(1-9) IV(i-8) (5) Limestone gravel (<10cm) os 3-0 IV (3-8) IV(i-5) IV(1) Limestone gravel (>10cm) 24 2-4 IV(i-7) IV(1-9) Faeces (rabbit/sheep) 11 0-3 Ii(i—3) I(1) IV(1) Notes: + Frequency classes: I = 0—20%; II = 20-40%; III = 40-60%; IV = 60-80%, V = 80-100%. Additional species in <2 quadrats: Non-flushed; Antennaria dioica, Cirsium sp., Dicranum scoparium, Drepanocladus_ revolvens, Erophila verna, Galium sterneri, Lotus corniculatus, Polytrichum juniperinum, Pseudoscleropodium purum, Vulpia bromoides. Flushed; Bryum argenteum, Plantago media, Primula farinosa, Racomitrium lanuginosum. Disturbed; Achillea millefolium, Ajuga reptans, Cardamine pratensis, Hypericum montanum, Polygala vulgaris. 204 K. J. WALKER However, drought can have a devastating effect on plant numbers as in 1992 where low spring rainfall (83 mm for May and June as opposed to an average of 163-5 + 16-6 mm between 1981- 1995) caused a number of localised extinctions (Fig 2.). As a result subsp. anglica is most frequently found within sheltered microtopographichal niches (e.g. rock fissures, moss cushions, solution cups etc.) where water-loss and exposure to extreme temperatures are reduced (Halliday 1960a). VEGETATION The habitat of subsp. anglica typically occurs as scattered “islands” of open ground, usually only a few metres in extent, within upland limestone pastures dominated by Sesleria caerulea (CG9; Rodwell 1992). Within this localised habitat subsp. anglica occurs within three closely related assemblages with differing microtopographic, edaphic and disturbance regimes (Table 2). On non-flushed soils overlying gently inclining limestone exposures subsp. anglica occurs with a number of other diminutive species (e.g. Minuartia verna, Sagina nodosa and Erophila verna) on bare patches of soil/gravel amongst scattered clumps of Festuca ovina and Sesleria caerulea. Here vegetative cover rarely exceeds 30% and is characterised by the presence of stress-tolerants such as Thymus polytrichus and Carex caryophyliea. On flushed soils and gravels vegetative cover is less fragmentary and more species-rich. Here the constants, Sesleria caerulea, Festuca ovina, Minuartia verna and Linum catharticum are joined by a number of species of moister soils most notably Primula farinosa, Carex viridula subsp. oedocarpa and C. panicea. Presumably this is the Arenaria norvegica subsp. anglica — Sedum villosum heath described by Bartley & Clark (1978) which they considered a transitional variant between the more species-rich flushes and the surrounding calcareous grasslands. Subsp. anglica also occurs within a species-poor “disturbance” assemblage on tracks and rabbit scrapes. Here it is associated with ruderals of bare soils, such as Prunella vulgaris and Poa annua, and rosette-forming hemicryptophytes such as Bellis perennis, Plantago lanceolata, Taraxacum officinale and Ranunculus bulbosus which can tolerate heavy trampling pressure. LIFE-CYCLE Classified as an annual or biennial (e.g. Clapham et al. 1987), the majority of plants germinate in the autumn, over-winter as leafy shoots and flower and set-seed in the following spring or summer (winter annual) (Halliday 1960a; Walker 1995). Under exceptional circumstances plants have also been observed to flower and set seed within one season (summer annual) (Walker 1995), or survive two winters (biennial) (Halliday 1960a). Throughout its life subsp. anglica is exposed to extreme variations in both temperature and . moisture supply; as a consequence there is no general relationship between the time of flowering and season. This is achieved by the production of polycarpic shoots (cymes) which enable the plant to flower continuously (indeterminately) over an extended period. Thus although flowering typically takes place between early May and September, it is not uncommon to find plants flowering as late as December if conditions are favourable (Whitwell 1895). This trait (sometimes termed uniseasonal iteroparity) allows annuals such as subsp. anglica to avoid the potential dangers of synchronous flowering within an unpredictable habitat, and ensure that some seed is dispersed under favourable conditions (Harper 1977; Symonides 1988). On average plants produce 2—3 hermaphrodite white flowers. As with many other members of the Alsinoideae, reproduction may be entirely autogamous (i.e. within the same flower) due to the automatic transfer of pollen between the homogamous reproductive surfaces. On maturity both the inner and outer whorl of anthers are forced into contact with the maturing stigmas ensuring that if they have not already been pollinated any remaining pollen grains are picked up (G. Halliday, pers. comm., 1998). In addition, plants which are artificially self-pollinated, and isolated from insect visitors with a fine muslin, are able to reproduce and set seed (Walker 1995). However, the production of obvious signals (simple white receptive flower, sweet scent) and rewards (nectaries) suggest that insects do visit the plant and that cross-pollination may take place. Subsp. anglica produces relatively large seeds (0.8 x 0.9 mm), the tough black testa having numerous raised tubercles which presumably serve to reduce desiccation. The viability of fresh seed is usually very high, with over 90% germinating readily when placed on moist filter paper (S. Terry, pers. comm., 1995; Halliday 1960a). However, experimental studies carried out by the author have shown that germination declines with increased illumination and temperature, and at low moisture levels (Walker 1995). This suggests that optimal conditions for germination are in sheltered micro-sites where moisture supply is not limiting. ARENARIA NORVEGICA SUBSP. ANGLICA 205 Capsules contain on average 3—4 seeds (mean 3-59 per capsule, n = 291) although under cultivation this can be much higher (Walker 1995; G. Halliday, pers. comm., 1998). After fertilisation it can take up to three weeks for the pericarp to split and reveal the seeds inside. Although some seeds are released at this stage the majority remain firmly attached by placentae to the capsule until detachment and decomposition of the pericarp on the surrounding soils. As a result dispersal is extremely limited and usually not more than a few centimetres from the parent plant. However, the presence of subsp. anglica on tracks, in particular the Clapham Lane and the outlying colony at Dawson Close, suggests that seed may be transported over greater distances, presumably within soil attached to humans, vehicles or sheep which are often “away-wintered” from the Sulber pastures (P. Corkhill, pers. comm., 1995; Sledge 1952). In addition, like Thlaspi perfoliatum it may have been widely dispersed within limestone ballast used in the construction of road and railways in the past (Rich et al. 1989). However, although both species are now well established on artificial sites (such as quarry floors, tracks and railway embankments) the lack of any specialised seed dispersal mechanisms means there is little tendency for further spread to occur. Although experiments suggest that relatively few seeds are incorporated into the seed-bank, the sporadic reappearance of the plant within a number of trackside sites suggest that some seed may persist for short periods. CONSERVATION It seems likely that subsp. anglica may well have evolved, through isolation, from a more widely distributed and variable population (Pigott & Walters 1954; Raven & Walters 1956). The occurrence of fossil seeds of Arenaria cf. ciliata, to which subsp. anglica undoubtedly belongs, within Mid- and Late Devensian deposits in the south and east of England and the east of Ireland suggests a much wider glacial distribution (Fig. 1). These species presumably survived the height of glaciation within periglacial refuges free of ice (i.e. unglaciated massifs, coastal areas) or on the periglacial “park-tundra” to the south of the ice limits (Godwin 1975). However, as a result of climatic changes during the post-glacial this once continuous lowland distribution has been restricted to a number of disjunct montane and coastal refugia where immature soils still persist (Godwin 1953; Rose 1957; West 1988). Today the rarity of these habitats, as well as poor dispersal ability has presumably compounded this restriction in range. As such subsp. anglica is likely to remain extremely rare being restricted by biological and ecological factors rather than as a direct result of human activities. CURRENT THREATS Although there is no evidence for an overall decline in numbers this century many trackside populations have been lost or severely degraded as a result of the increased usage by walkers, “‘off- road” vehicles, scramble and mountain bikers. This has led to severe soil erosion and a decline in plant numbers at numerous sites along the Clapham Lane. In particular track-widening and Shortcuts created by walkers threaten large colonies at Sulber Gate and Crummack Dale. At Selside, where the plant grows on the edge of a popular track leading to Alum Pot, car-parking has reduced numbers to the verge of extinction. Many of the pastures in which the plant occurs were formerly overgrazed (sheep numbers tripled between 1940 and 1980). As a result sheep numbers have been reduced through management agreements since designation as a S.S.S.I. in 1986. However, many populations are still highly susceptible to disturbance caused by off-road vehicles. There are no 20th Century reports of collecting of subsp. anglica and disturbance caused by visiting botanists is unlikely to pose a threat to many of the larger populations. However, smaller, more accessible colonies (some with less than ten plants) are extremely vulnerable to trampling, particularly at Selside which is frequently visited by botanists. CONSERVATION MANAGEMENT Subsp. anglica has been included within English Nature’s Recovery Programme as a species considered to be in danger of extinction and requiring special conservation measures (Whitten 1990). As a result an action plan has recently been proposed in order to ensure its long term self- sustained survival in the wild (Walker & Corkhill 1996). 206 K. J. WALKER By the early 1990s all the sites in which subsp. anglica occurred received statutory protection either as part of the Ingleborough National Nature Reserve (N.N.R.) or as Sites of Special Scientific Interest (S.S.S.L) (Ingleborough, Pen-y-Gent and Foredale Quarry). As a consequence grazing levels are now agreed with farmers (in some cases voluntarily under the Wildlife Enhancement Scheme), and ensure that sheep numbers do not exceed two ewes (and followers) per hectare throughout the year. In addition, one pavement on which the plant occurs (Juniper Gill) has been recently stock-proofed (P. Corkhill, pers. comm., 1995). The steady increase in plant numbers recorded within these pastures (Fig. 2) suggests that these measures are having favourable results. In order to arrest the decline of subsp. anglica on vulnerable track-side localities a number of measures have been proposed. At Sulber Gate on the Clapham Lane a diversion of the Pennine Bridleway has been agreed with the Countryside Commission in order to protect a large colony. In addition, better way-marking of routes will reduce the numbers of “unofficial” paths which have appeared in recent years. At Selside limestone boulders have been placed around sensitive areas in order to discourage car-parking. In the long term ex-situ measures, such as seed-banking and the establishment of populations in “safe-sites”’, will safeguard the plant from ultimate extinction as well as providing material for further research and public viewing. In 1992, 500 seeds were collected from two extant populations and are now held at the Royal Botanic Gardens Kew Seed Bank at Wakehurst Place. In the same year seed was also artificially introduced to Scar Close N.N.R. where it has apparently failed to establish. Introduction into further “safe-sites” is currently being discussed with a view to creating a public viewing facility in order to reduce pressures on extant populations (P. Corkhill, pers. comm., 1995). In order to assess the success (or otherwise) of these management activities a sample of populations are being monitored every three years. The results of recent surveys suggest that the efforts of conservationists and landowners have ensured the survival of this endemic plant for many years to come. ACKNOWLEDGMENTS I am especially grateful to Peter Corkhill at Ingleborough without whose support and invaluable assistance this project would not have been possible. I would also like to thank Mike Foley, Barrie Goldsmith, Adrian Soames, Janet Terry and Peter Welsh for information and advice on various aspects of this work. Chris Preston, Geoffrey Halliday and Owen Mountford all made valuable ~ comments on an earlier draft of this paper and Dave Roy at Monks Wood produced Fig. 1. Fieldwork was supported by grants from English Nature as part of the C-E-L Scheme and Species Recovery Programme. REFERENCES ABBOTT, P. (1996). Alien plants from Yorkshire. BSBI news 73: 47. ALBERTSON, N. (1946). Osterplana hed. Acta phytogeographica Suecica 29: 156. BAKER, J. G. (1889). On the varieties of Arenaria (gothica) ciliata. The naturalist 14: 337-339. BARTLEY, D. D. & CLARK, S. C. (1978). A vegetation survey of Ingleborough Site of Special Scientific Interest. Unpublished report by the University of Leeds for the Nature Conservancy Council. BENNETT, A. (1891). Notice of the occurrence of Arenaria gothica, Fries, in Great Britain. Botanical Society of Edinburgh 18: 252-254. BENNETT, A. (1892). Arenaria gothica. Journal of botany 30: 86. BLAKEMORE, J. (1981). Survey of Arenaria norvegica subsp. anglica. Unpublished report for English Nature. CHATER, A. O. & HALLIDAY, G. (1993). Arenaria in TUTIN, T. G. et al., eds. Flora Europaea 1: 140-148, 2nd ed. Cambridge University Press, Cambridge. CLAPHAM, A. R., TUTIN, T. G. & Moore, D. M. (1987). Flora of the British Isles, 3rd ed. Cambridge University Press, Cambridge. DAHL, E. (1951). On the relation between summer temperature and the distribution of Alpine vascular plants in the lowlands of Scandinavia. Oikos 3: 22-52. DANIELS, A. C. (1993). A survey of Arenaria norvegica Ee: anglica in the Ingleborough area in 1993. Unpublished report for English Nature. ARENARIA NORVEGICA SUBSP. ANGLICA 207 DANIELS, A. C. (1994). A sample survey of Arenaria norvegica subsp. anglica in the Ingleborough area in 1994. Unpublished report for English Nature. FARRER, R. T. (1894). Arenaria gothica. Journal of botany 32: 344. FoLey, M. J. Y. (1995). Survey for Arenaria norvegica subsp. anglica Halliday (Caryophyllaceae) in the Ingleborough area, North Yorkshire. Unpublished report for English Nature. GODWIN, H. (1953). British vegetation in the full and late glacial periods, in LOUSLEY, J. E., ed. The changing flora of Britain, pp. 54-74. Botanical Society of the British isles, Oxford. GODWIN, H. (1975). History of the British flora, 2nd ed. Cambridge University Press, Cambridge. HALLIDAY, G. (1958). The northern species of Arenaria ciliata complex and some interspecific hybrids. Proceedings of the Botanical Society of the British Isles 3: 90. HALLIDAY, G. (1960a) Taxonomic and ecological studies in the Arenaria ciliata and Minuartia verna complexes. Ph.D. thesis, University of Cambridge. HALLIDAY, G. (1960b) The identity of Arenaria gothica auct. angl. Watsonia 4: 207-209. HARPER, J. L. (1977). The population biology of plants. Academic Press, London. HODGETTS, N., PALMER, M. & WIGGINTON, M. (1996). The pink book of plants. Lists of vascular and non- vascular plant species which are nationally threatened, localised or protected in Great Britain. J.N.C.C., Peterborough. KENT, D. H. (1992). List of vascular plants of the British Isles. Botanical Society of the British Isles, London. LEES, F. A. (1937). The vegetation of Yorkshire and supplement to the floras of the County. The naturalist 62: 233. MORLEY, J. (1992). Rare plant survey of Arenaria norvegica subsp. anglica in the Craven District of North Yorkshire. Unpublished report for English Nature. PiGoTT, C. D. & WALTERS, S. M. (1954). On the interpretation of the discontinuous distributions shown by certain British species of open habitats. Journal of ecology 42: 95-116. RAVEN, J. & WALTERS, S. M. (1956). Mountain flowers. Collins New Naturalists Series, London. RICH, T. C. G., KITCHEN, M. A. R. & KITCHEN, C. (1989). Thlaspi perfoliatum L. (Cruciferae) in the British Isles: distribution. Watsonia 17: 401-407. RODWELL, J. S. (1992). British plant communities. Vol. 3. Grasslands and montane communities. Cambridge University Press, Cambridge. ROSE, F. (1957). The importance of the study of disjunct distributions to progress in understanding the British flora, in LOUSLEY, J. E., ed. Progress in the study of the British flora, pp. 61—78. Botanical Society of the British Isles, London. ROTHERAY, L. (1889a). Arenaria gothica Fries, a plant new to West Yorkshire. The naturalist 14: 314. ROTHERAY, L. (1889b). The discovery of Arenaria gothica in West Yorkshire. The naturalist 14: 335-339. ROTHERAY, L. (1895). Further discoveries of Arenaria gothica. The naturalist 21: 303-306. ROTHERAY, L. (undated). Botanical rambles; 188S—1900. Photocopied manuscript held by the author. ScoTr, W. & PALMER, R. (1987). The flowering plants and ferns of the Shetland Islands. The Shetland Times, Lerwick. SLEDGE, P. (1952). Litton v.c. 64, June 14th: flowering plants. The naturalist 72: 179. SMITH, A. J. E. (1978). The moss flora of Britain and Ireland, \st ed. Cambridge University Press, Cambridge. SYMONIDES, E. (1988). Population dynamics of annual Plants, in DAVEy, A. J., HUTCHINGS, M. J. & WATKINSON, A. R., eds. Plant population ecology, pp. 221—264. Blackwell, Oxford. TAYLOR, I. (1990) A survey of Arenaria norvegica subsp. anglica. Unpublished report for English Nature. WALKER, K. J. (1995). The ecology and conservation of Arenaria norvegica subsp. anglica Halliday), Caryophyllaceae. M.Sc. thesis, University College London. WALKER, K. J. & CORKHILL, P. (1996). Species Action Plan: English Sandwort (Arenaria norvegica subsp. anglica Halliday) {draft version]. English Nature, Peterborough. WeBB, D. A. & SCANNELL, M. J. (1983). Flora of Connemara and the Burren. Royal Dublin Society & Cambidge University Press, Cambridge. WEST, R. G. (1988). A commentary on Quaternary cold stage floras in Britain. Journal of biogeography 15: 523-528. WHITTEN, A. (1990). Recovery: a proposed programme for Britain’s protected species. English Nature, Peterborough. WHITWELL, W. (1889). Arenaria gothica Fries in Britain. Journal of botany 27: 374-378. WHITWELL, W. (1890). An additional station for Arenaria gothica in West Yorkshire. The naturalist 15: 257. WHITWELL, W. (1895). More about Arenaria gothica. The naturalist 20: 257-260. WILLIAMS, F. N. (1898) A revision of the genus Arenaria. Botanical journal of the Linnean Society 33: 326-437. WILSON, P. (1986). Arenaria norvegica subsp. anglica. Unpublished manuscript for the quinquennial review of rare species. Nature Conservancy Council, Peterborough. WYSE-JACKSON, M. B. & PARNELL, J. A. N. (1987). A biometric study of the Arenaria ciliata L. complex (Caryophyllaceae). Watsonia 16: 373-382. (Accepted March 1999) K. J. WALKER 208 ‘POBOOTOI JOADN *, ORs] QUOJSOWTT] PaTfeIoW UO sjuRTd Jo JOqUINU [[PLUS,, B 1OJ PIODII J[SUTS ‘(6Q8T [OMITYM) 1OUUSeID WO po1dd{[O [eLIOVeUT JUL, PoUTULIOJOpuN Uv JO JUSWISe.L] ‘Sn [LUOTILIIOII puv SUIZLIS 0} ONP JOUNXS dUIRIOq A[Quqolg “You dy] JO ,SINA,, UTYITAA ‘poonpo.ul ATQIssog “(S68 [[JOMIIYM) FP] OY) 0} OSO[D JUOUTOARd UC “BUTIII[[OD -IDAO 0} ONP UOTOUNXY “UOTLIS ABMIILI DY} 0} ISO[D YOR. UO 91S ATOAODSIP [BUISLIO “YST[qeyso 0} popley sey 1 Isassns SADAINS JUDIOY “USWIOALA TJ OJUO Poyeoo[sue. pods ‘pooudy-yoo s pue poywoojos ATWUDIOY “Aoy[eA ,.Aup,, B UIYIIM UONeindod Tfeurs ‘asn [euONeaIoaI Ag poudyeamy_], You.) v 0} JUDIelpe dUO}SOUT] JeLJ UO UOTe[Ndod [RUS ‘ajsem Aarenb jo 3urdumnp Aq poudjeaiyy, “Joop Alsenb uo JoyI[9 yssuoure suoneindod moj ‘saypem Aq ynd-yoys Jo asn Aq pouaqywoaiyy, ‘Aopea , AIp,, Ul soov..10] UO uoNL[ndod ose] ‘(posnsIp MOU) YORI} [[IY B UTYIIM DUOSOWIT] JO Bore [[eWUs UO UONL[Ndod s51e'7] ‘asn [euONvaia Aq poudyVaIy], “Yov.) dy) 0} XoU somnsodxs ouOsoUNT] UO UOTLINdod ase] ‘asn jeuonvaioas Aq pousywary,], ov] oy) 0} Jxou , JUOWUARQUID,, UT UONe[Ndod [[eus Ara A, ‘SQ6HRT WU) UL BAIR JATSUDIXO JOAO popJOIIY “Yor. JO ,.SiNs,, UII uOTe[Ndod [[eursg ‘posodoid usoq sey uonendod sty) punore Aeme ARMOIIPLIG DUIUUA BY} JO UOISIOAIP Y “(CHZ] ALIOIOY) SOS] OY) Ul DOURpUNGe UT popsosay “WraIsAs SITLU/YSNTF Snosoevyn} UTM sainsodxd duojsoUT] ITF UO UONe[Ndod o5.1e7T ‘JOUNNXS AT[ENIA MOU ‘SQGRI UL PopslOoy “yorsy JO syns uIyIM uOTe[Ndod [[eUsS ‘QITU 9SALT 0} JUDDe[pe saInsodxd dUOJSOUT] UO UONL[Ndod [feus AIO A, ‘SQ6QI Ul Jo.ue,y Aq pardAoosip suo oy} ATqvunsoig ‘WUoWoAed eT uo UoNL[ndod a31e'7T ‘TTEM MOTOG SddK.LI9} DUOJSOUWT] JTF UO UONL[Ndod ss1e7T * Aayyea MOTeYS,, pue | YOIN,, Wy Usamjoq youn-dsoys suoye soinsodxo uo uoneindod {ews ‘Jap JUoUIOARd Udyog UO SUIMOIS UOTFe[Ndod [][eUIs AIBA, ‘sadvios yiqqei pue yuoWdAed UO OsTR SIOQUINU [TRU “ATUI/YSN]J SNOddeJn} ISI] SUIIIM pue ‘puNnore puNoIs }STOW UO JUepUNqe SOP *,,AMPCA MOTL[LYS,, UTYIIM (PYT 9) Bde DATSUD}XI IBAO paroyeVds uoNe[ndod jsosie] ‘Qovjins JUOWOARd UO pur YSN spYyI0a]/q JO Ispo Uo UONe[Ndod pozis UMNIps/y : ra (LE61) $991 “VA =) 0019 (6881) Nouusg “Vy = A0S¢ (6881) Jou “Y / AvsoyOY “T = --00¢ (S681) Jou Y aes: (6881) Avrouoy “7 UWNN Ore (7661) IPOD ‘d TSSS O00 (8L61) HPID O'S & Ao ‘dd TSSS Ov (ZS61) AQIEN “UY TSSS OS (T86]) soursyelg ‘f TSSS O1€ = (8L61) HPID D'S @ Aotueg “dd TSS'S 09€ (TZ6]) wowsyrlg_ ‘f TSSS O8€ = (8L61) HPID D'S @ Aotueg ‘dd TS'S’S. CRE (T861) Mowayeyg_ ‘f TSSS 09¢ (C681) Jouey “Y / AvaIOY “J UNN O9¢ (C681) ouey “Y / AviOyIOY “J TSSS OLE (C681) Jouey “Y / AvIOyIOY “T TSSS SSE (7661) HEMPHOD ‘d UWNN 06€ (S681) ouey Y NN S8¢ (7661) JOXTEM “MI UNN O09 (C661) JOMIPM “M UNN SSE (OC61) APpITeH ‘D UNN So-Spe (8L61) VRID O'S 7 AouRg ‘dd UNN SSe€ (8461) HRID D'S % Age ‘dd ayepoyeMs ‘ysneleay VLIQUING ‘O1OUUSeIDH Pel HoAu/) (Jaq(ns) amseg suLIOg Age A sutids uonris peaya|qqry yOunX| (UOTONPONUT) IsO[D eds I[!D Jodrung dso[D UOsMEG Alseng) o[epotoy sulonog weydely gyeq yorurunig (1e9S$-sUOT) dUeT WeYyde[y (peol-ssois) oueT weydeyy (a}e3) our] weydr[D (Yysn{J) YSIN JoqIngs oueT weyde[a-YoIN Jaqins SSOJ ,SOADTU_L 0g Joqins (SOM) YIN Toqing (yoe.n-daoys) YOIN Joqins (s Aepr][eH) oanjsed ssqins (, Aaypea,,) aimsed raqins ainjsed Jaqins "Ysnyy [[euus & JO aspa oy} Aq uONe[Ndod [[eWIS “WN N OVE (S661) JOMLM “M (Ysn Mopoq) sinjsed ssq[ns ‘sIVdXK JUDIAI UL PoUuTpOap sey Inq JUBpuNge sdUG “AeMoTpLIg Ie_ndod Aq uonendod [Tews - C662 (O68) uosdwuoYyL “4 yw 's (UIDID UOJ) IPISTAS yUvIXA uONISOg smeig opmn[yv Teak /1aplooayy uone[ndog ‘VOITONV ‘dSANS VOIDFAYON VIYVNAYV MOS SCYOOAA GHIWIAHANN GNV LONILXA (LNVLXH “XIGNAddV Watsonia 23: 209-212 (2000) 209 Annual variations in the size of a population of Cardamine impatiens L. L.R. WILLIAMS Brent Parks Service, c/o Brent Town Hall, Forty Lane, Wembley, Middlesex, HA9 9HD ABSTRACT A population of Cardamine impatiens Narrow-leaved Bittercress was monitored by annual counts of the number of plants during the decade 1988-1998 at Fryent Country Park, Middlesex (v.c. 21). There were large annual variations in the number of individuals, both in total and at each colony. Variations appeared to be due to the cyclic nature of populations at each colony, to unexplained variations possibly related to the weather, and to ground disturbance. The species increased in response to occasional ground disturbance. The reappearance of plants where it had not been observed for several years, suggested the role of a buried seed bank. The average spread, in a south-easterly direction, was 17 ma year over the decade. KEYWORDS: Narrow-leaved Bittercress, Brassicaceae, monitoring, Middlesex, hedgerows. INTRODUCTION Cardamine impatiens, Narrow-leaved Bittercress is a scarce plant in Britain with scattered populations nationally (Stewart et al. 1994). It is a plant of varied habitats including shady woodland, rocks and screes, shady banks of streams and rivers and of damp roadsides. It occurs on both calcareous and neutral soils. In London and in Middlesex (v.c. 21) it was considered extinct (Kent 1975; Burton 1983) until a population was rediscovered in 1985 at Fryent Country Park, (TQ/198.881), in the former parishes of Kingsbury and Wembley. Prior to that the most recent record had been ‘near Harrow’ in 1901. Wembley was part of the Harrow parish until 1894. Monitoring of C. impatiens commenced in 1988 with the aim of providing information on how best to conserve the species locally. METHOD The investigation was undertaken at Fryent Country Park, approximately 15 km north-west of central London and in the London Borough of Brent. The Country Park covers an area of approximately 100 hectares of lowland countryside surrounded by suburbia and bisected by the A4140 road. The landscape is of hay meadows cut annually, with hedgerows, woodlands, ponds, and areas of scrub, rough grassland, amenity grassland and horse grazing. The park is on London Clay, with a pH of about 6-5, though Barn Hill, to the south, is capped with gravel and more acid soils. The altitude varies between 45-83 m O.D. The flora of the park was described by Williams (1996) and the landscape by Williams & Northcroft (1992). Counts were made by the author to estimate the number of individual plants. Each more-or-less continuous group of plants was mapped as a “colony”. Obviously there were limitations to this approach, since some of the colonies merged during the investigation. Counts were made towards the end of May (to early June) in each of the years 1988 to 1998. RESULTS The number of individuals at each colony are given in Table 1. The species was found in the northern, central part of the Country Park. Records were from both the Kingsbury and the Wembley parish areas, and from the park on both sides of the A4140 road. There was considerable 210 L. R. WILLIAMS variation in the population size from year to year (Table 1 and Fig. 1). The highest count was in 1997 (9977) and the lowest in 1995 (504), though the 1994 count (1076) would probably have been lower if it had not been for two new colonies, on disturbed ground, totalling 900 plants. There was an increase in the number of colonies during the investigation, mainly within the original range. However the linear spread, in a south-easterly direction was approximately 170 m, or an average of 17 m a year. At the end of the investigation the range of the species covered about 25 ha, or about a quarter, of the park. Cardamine impatiens was predominately associated with hedgerow edge habitats. In almost all other cases it could be found within 10 m of a hedgerow, for example, on ditch and streamsides, in nearby scrub and on ground that had been disturbed near to hedgerows. C. impatiens preferred the herbaceous hedgerow edge, but was occasionally found under the hedgerow canopy in the centre of the hedgerows. The flora and history of the hedgerows has been described by Williams & Cunnington (1985), and in Williams et al. (1987). A few plants were found in the adjacent hay meadows, probably as a result of ground disturbance or poaching of the ground. It was however, unable to survive in hay meadows beyond the first year. A colony established on an adjacent railway cutting. It survived as a weed in the garden of one property backing onto a hedgerow. Several plants were found in the drying mud of seasonal ponds. In addition to the annual variation in the total population there was considerable variation in the counts of each colony. The relative variation in the numbers at individual colonies did not necessarily reflect the overall changes. Three main factors appeared to be responsible for the year to year variations in the population at each colony. These factors were cyclic variation, the weather and ground disturbance. In addition, the total population may have increased as a result of the establishment of new colonies. At each colony there appeared to be a cycle as the population increased and then decreased. The duration of these cycles was difficult to estimate due to the other factors involved, but two to four cycies per decade appeared to be the norm. To some extent the cyclic nature would be expected from a biennial species, but the cycies could aiso be a facet of the weather-related variation. The weather was possibly a significant factor affecting the population, particularly when the populations at most colonies increased or decreased in synchrony. Thus 1994 and 1995 were years of low populations, while the highest number of plants was recorded in 1997. It is not known which weather factors were responsible for these effects and at which stage/s in the life cycle. In 1990 many of the plants were smaller than usual, possibly as a result of the dry conditions of that year. Nevertheless, many plants were flowering even though they were less than 20 cm tall. In | 1995 many of the young rosettes were affected by a late frost. 1888 . 1989 1896 1982 1983 1984 1995 1996 1987 1988 FIGURE 1. Changes in the population of Cardamine impatiens at Fryent Country Park, Middlesex, 1988-1998. 211 VARIATION IN A CARDAMINE IMPATIENS POPULATION 1Z91 LL66 9OEE 70S 9LOT ELBE ETT O8EE ZST1 619T O£0Z s[ej0 |, z 9 epispuog OV 8Z 0z ia yueq MOJospopy qv = 0z Z apis puog eV - OS OTT espe MOIOSPOL] AR - cs 7 os OOE poqmisip/suoiqes Wes xX 6 08Z s : 009 MOIOSPay/apIsuMIvINS NM vSS S9EI 8701 v7 2 OL Youp/espe MoraspoH] A 3 ¢ 09 O¢ 07 062 opis Aemprey a = = - T - C MOIOSPY/SpIspLoy LL 4 O£Z 161 v ral 866 pueysseis ysnor/qn.iog S - - - - - 7g ospo2 MOIOSpeH] WY - - - - - O€E puv[ssvis ysnol/qniss re) - - - - - ¢ - C 0} poonpo.nui - o3pa MOIEspoH] N 0Z CT 06 09 5 = pe CS aspe MOIOspaH] 7, 16 O€T LIE Os 97 OLT O8T OL ad3pa MOIOSpoH d 8 OL 09T SE : OV 5 002 aspe MOJOSpa}] O 8p OS : = : (6 € EL a3po MOJospoH W 2 GI = : OO CI 00T OI 0ZI purjssers ysnor/qniog TI 061 O¢T ELI 5 = 0z 6v CE €8 CL OOT purysseis Ysnoy | OI = €T 06 4 Sr : Or €] O9T OT spIsuuvans/MOIas pay f OL s > ; - ZS 07 OO€ L OI 9 aspe MOIISpoH] I 00Sz - : = OSP OT BETZ Wi 0zI O€ qnuios/MorIaspaH H Z 080r IZ = a Ors IST 079 0zI 06r O19 qnios/MorospoH 9) - = = = 4 = = = = = I d3Pd MOJOSPI}] J 01 07 8 - - O¢ OSE OTT LE L6E MOJOSpoy - puNoss paqinjsiq d Sol CSL 18z O€ SZ OS COZ OOL 61€ 61 LZ9 a3pe MOJOSpoH] d OLE TEZ O€Z : = CLG SI 0sz 09 87 OST aspo MOJOSpoH e) 8Z S OSY OLT 0z 6€ O¢€ 07Z 09¢ OST cs espe MOIaspoH] q ; - OST OI 09 09T 78 001 O¢ Or bb a3pe UspIes/MOIOSpoH] V 8661 L661 9661 S661 7661 £661 7661 1661 0661 6861 8861 eoh yeuqey AuojoD 8661-8861 “XASATICCIN “TuVd AULNNOO LNAAY LV SNAILVdWI ANINVGYV) AO SLNNOO TT ATEaVL 212 L. R. WILLIAMS Ground disturbance was responsible for some major fluctuations in populations. During the investigation ground disturbance was often a result of mechanical works on the park. The effect was also noted after work by conservation volunteers on hedgerow edges, ponds and streamsides. This suggested germination of seeds from a buried seed bank and/or from seeds released by adjacent plants. The evidence for a buried seed bank gains support from the observation that in some years no plants could be found at some colonies, and yet the colonies would revive in later years (Table 1). DISCUSSION It is accepted that counting errors could affect the results, though the results probably give a (1991) suggested a number of recording problems during counts of plant populations. Possible easons for error include reductions in the number of seedlings surviving through the course of the growing season. Whilst the survey was undertaken in late May (to early June) each year, it is possible that in a ‘late’ season, more seedlings would have appeared subsequently. Hutchings (1991) emphasised that the growing plants do not necessarily reflect the size of the population of any dormant viable seeds in the soil. Indeed, it was obvious from the resuits that the seed bank olayed an important role in the reappearance of C. impatiens at sites where it had not been observed for one or more years. The effect of ground disturbance on the maintenance of populations of C. impatiens was noted by Stewart et al. (1994) and by Briggs (1990). Where C. impatiens is present in southern ashwoods, seedlings appear in cleared sites within the year following felling. After three years, very few plants survive against the competing vegetation. The dormant seed buried in the soil, however, remains viable for many years. Corner (1988) described the habitat of a population of C. impatiens in Scotland where it occurred on scree in old woodland and noted that there was some ground disturbance at that site as a result of cattle and sheep grazing. Sinker et al. (1985) also noted that C. impatiens was intolerant of competition, but invasive in recently disturbed habitats. At Fryent Country Park, C. impatiens was recorded on recently re-dug garden soils adjacent to a hedgerow colony of the plant. Whilst the species spreads mainly by the explosive release of seeds, it is possible that some of the new colonies, assuming that they are new, were established by seeds carried by other means. In the London investigation, C. impatiens was largely associated with the hedgerow edge; a habitat additional to those listed by Rich (1991), but consistent with the shade preference of this species. The results suggest that locally, the species is best able to survive in light herbaceous vegetation, and on recently disturbed or exposed ground. It prefers moderate shade and limited competition from other plants. ~ REFERENCES BriGGs, M. (1990). Sussex plant Atlas supplement. Booth Museum of Natural History, Brighton. BURTON, R. M. (1983). Flora of the London area. London Natural History Society, London. CORNER, R. W. M. (1988). Cardamine impatiens L.: A native Scottish locality. Watsonia, 17: 91-92. HUTCHINGS, M. J. (1991). Monitoring plant populations: census as an aid to conservation, in GOLDSMITH, F. B., ed. Monitoring for conservation and ecology, pp. 61-76. Chapman and Hall, London. KENT, D. H. (1975). The historical Flora of Middlesex. The Ray Society, London. RICH, T. C. G. (1991). Crucifers of Great Britain and Ireland. Botanical Society of the British Isles, London. SINKER, C. A. et al. (1985). Ecological Flora of the Shropshire region. Shropshire Trust for Nature Conservation, Shrewsbury. STEWART, A., PEARMAN, D. A. & PRESTON, C. D. (1994). Scarce plants in Britain. J.N.C.C., Peterborough. WILLIAMS, L. R. (1996). The flora of Fryent Country Park, Middlesex. The London naturalist, 75: 45-66. WILLIAMS, L. R. & CUNNINGTON, W. (1985). Dating a hedgerow landscape in Middlesex: Fryent Country Park. The London naturalist, 64: 7-22. WILLIAMS, L. R., MCLAUCHLIN, J. & HARRISON, T. G. (1987). Hedgerows surviving in suburban Kingsbury. The London naturalist, 66: 35-39. WILLIAMS, L. R. & NORTHCROFT, K. (1992). Heritage land: the case for Fryent Country Park. The London naturalist, 71: 33-42. (Accepted March I 999 ) Watsonia 23: 213—220 (2000) 218 Notes UPDATE ON BIRDSEED ALIENS (1985-1998) In the 13 years since our joint paper (Hanson & Mason 1985), which listed a total of 425 species, of which 318 were actually cultivated by the authors from various birdseed mixtures, I have continued to examine packets purchased in supermarkets, pet shops, garden centres and street markets. The less common seeds have been extracted for cultivation and the remainder mixed with Sunday-roast fat for the garden bird table. Many of the resulting exotic weeds have turned out to be the usual familiar aliens but several new, upgraded and overlooked records can now be added to the list of definite birdseed aliens as the sources of birdseed constantly alter due to changing world commerce. The number of such additions amounts to a further 44 taxa while the local sewage works has produced another six “probables”’, i.e. species growing together with known birdseed aliens. In addition to the above, Clement & Foster (1994) and Ryves et ai. (1995) include another 34 species as possible birdseed aliens although some of these are much more likely than others thus making a grand total of just over 500 taxa. Continued cultivation experiments will certainly enlarge the proportion of definite birdseed aliens to more than the 70% of this grand total which will itself also certainly marginally increase. Any further additions on a major scale to the total number of taxa now seems most unlikely. DETAILS OF ADDITIONAL BIRDSEED ALIENS BETWEEN 1985-1997 Cult. cultivated by CGH C. G. Hanson FH Mrs F. Houseman Miiller recorded in Miiller (1950) gardens recorded in gardens in association with other birdseed aliens tips recorded on rubbish tips with other birdseed aliens occasional a few plants seen most years sporadic single plants appearing irregularly The following abbreviations are used: The nomenclature and sequence of families follow Flora Europaea. The genera and species are listed alphabetically. Polygonaceae Persicaria glabra (Willd.) M. G6mez (Polygonum glabrum Willd.): Tips and parks; sporadic. P. pensylvanica (L.) M. Gémez: Tips; sporadic. Chenopodiaceae Atriplex prostrata Boucher ex DC.: Cult. CGH. A. sagittata Borkh.: Cult. CGH. Bassia scoparia (L.) Voss (Kochia scoparia (L.) Schrader): Sewage works and tips; sporadic. Amaranthaceae Amaranthus bouchonii Thell.: Cult. CGH; tips; sporadic, probably overlooked as A. hybridus. A. capensis Thell. subsp. uncinatus (Thell.) Brenan: Cult. CGH. Ranunculaceae Nigella arvensis L.: Tips; sporadic. N. sativa L.: Tips; sporadic. Cult. CGH. Papaveraceae Papaver dubium subsp. lecogii (Lamotte) Syme: Cult. CGH. 214 NOTES Crucifereae Brassica rapa L. subsp. oleifera (DC.) Metzger: Cult. CGH; tips; sporadic. B. tournefortii Gouan: Cult FH; also observed as a birdseed alien in Belgium (F. Verloove pers. comm.) Lepidium campestre (L.) R. Br.: Cult. CGH. Sisymbrium loeselii L.: Sewage works and tips; sporadic. Caesalpiniaceae Senna obtusifolia (L.) Irwin & Barneby: Sewage works; sporadic. Fabaceae Glycine soja Siebold & Zucc.: Cult. CGH. Lotus glaber Miller: Cult CGH. Lupinus sp.: Cult. CGH from niger seed (Guizotia) but not yet identified. Onobrychis viciifolia Scop.: Cult. CGH. Sesbania exaltata (Raf.) Cory: Sewage works and tips; sporadic. Vicia hirsuta (L.) Gray: Cult. CGH. V. sepium L. var. eriocalyx Ceiak.: Cult. CGH. Vigna angularis (Wilid.) Chwi & Ohasmi (Adzuki bean): Cult. CGH. V. radiata (L.) Wilczek: Cuit. CGH; tips; sporadic. Umbelliferae Apium graveolens L.: Cult. CGH. Conium maculatum L.: Cult. CGH. Boraginaceae Amsinckia micrantha Suksd.: Cult. CGH; tips; sporadic. My own specimen at least is this and not A. intermedia as listed in Hanson & Mason (1985). Borago officinalis L.: Cult. CGH; tips; sporadic. Verbenaceae Verbena officinalis L.: Cult. CGH. Lamiaceae Ballota nigra L.: Cult. CGH. Solanaceae Lycopersicon esculentum Miller: Cult. CGH; very frequent on tips and sewage farms but not as a birdseed alien. Solanum nigrum L. subsp. schultesii (Opiz) Wessely: Cult. CGH; tips; sporadic, probably overlooked in the past. S. physalifolium Rusby: Cuit. CGH; tips; sporadic. My own specimen at least is this and not S. sarrachoides as listed in Hanson & Mason (1985). Polemoniaceae Navarretia squarrosa (Eschsch.) Hook. & Arn.: Gardens; sporadic. Reported as a birdseed alien in Holland. Sesamum capense Burm. f.: This is probably the species which occurs sporadically at our local sewage works. Asteraceae Anthemis altissima L.: Recorded from a Kentish field with other birdseed aliens after a dressing of sewage sludge. A. arvensis L.: Cult. CGH. Cichorium pumilum Jacq.: Sewage works and tips; sporadic. Tripleurospermum inodorum (L.) Schultz-Bip.: Cult. CGH. Xanthium strumarium L.: Miller; gardens; tips; sporadic; cult. CGH q i r NOTES 215 Poaceae Bromus diandrus Roth.: Cult. CGH and FH. - Echinochloa crus-galli (L.) P.Beauv. var. oryzoides: Cult. CGH. Eleusine coracana (1...) Gaertn.: Cult. CGH; recorded also at Ghent (Belgium) as a birdseed alien by F. Verloove (pers. comm.). A cultivated millet in tropical Africa. Lolium x boucheanum Kunth (L. multiflorum x L. perenne): Cult. CGH. Oryzopsis miliacea (L.) Benth. & Hook.: Cult. CGH and FH; tips; sporadic. Panicum subalbidum Kunth: Cult. CGH; tips; sporadic. Paspalum dilatatum Poir.: Recorded from several urban areas with other birdseed aliens. Phleum exaratum Griseb.: Miiller; reported as a rare birdseed alien at Roeselvare, Belgium 1994 by F. Verloove (pers. comm.). Rostraria cristata (L.) Tzvelev: Gardens; tips; sporadic; reported as a rare birdseed alien at Ghent, Belgium in 1994. Triticum durum Destf.: Miller: cult. CGH; tips; sporadic T. turgidum L.: Miller; tips and waste ground; sporadic. Note 1 Suggestions are strong that the following species could well be included as birdseed aliens but definite proof is still awaited. Confirmation from readers would be most welcome. Aegilops neglecta Lolium persicum Amaranthus spinosus Medicago scutellata Crepis suberostis M. truncatula Eleusine muitiflora ~ Nigella hispida Galactites tomentosa Phleum paniculatum Lathyrus inconspicuus Picris cupuligera L. ochrus Polygonum bellardii Leptochloa uninervia Silene conoidea Note 2 The following aliens are claimed as of possible birdseed vector in the references below but I have been unable to corroborate them in my studies to date. E. Clement (pers. comm.) agrees that a significant number appear to be distinctly unlikely and also that several identification errors are suspected. Confirmation is again welcome. Agrostis gigantea Cucumis myriocarpus Oenothera canovirens Anthemis austriaca Euphorbia falcata Paspalum distichum Anthoxanthum ovatum E. segetalis Physalis ixocarpa Artemisia annua Gilia capitata Salvia viridis Bowlesia incana Ayoscyamus albus Schkuhria pinnata Brachiaria eruciformis Juncus pallidus Scolymus maculata Carduus thoermeri Lathyrus tingitanus Tetragonolobus purpureus Centaurea diffusa L. tuberosus Triticum monococcum Chenopodium missouriense Lepidium virginicum Verbascum phoenicium Clarkia quadrivulnera Linum grandiflorum Vulpia geniculata Convolvulus sabatius Malope trifida ACKNOWLEDGMENTS Once again I am greatly indebted to Eric Clement for checking the identity of most of the taxa and to Filip Verloove who has kept me informed of his birdseed discoveries in Belgium. REFERENCES CLEMENT, E. J. & FOSTER, M. C. (1994). Alien plants of the British Isles. Botanical Society of the British Isles, London. HANSON, C. G. & MASON, J. L. (1985). Bird seed aliens in Britain. Watsonia 15: 237-252. MULLER, K. (1950). Die Vogelfutterpflanzen. Mitt. Ver. Math. Naturw, Ulm 23: 3-33. RYVES, T. B., CLEMENT, E. J. & FOSTER, M. C. (1995). Alien grasses of the British Isles. Botanical Society of the British Isles, London. C. G. HANSON I Coltsfoot Road, Ware, Herts. 216 NOTES HIERACIUM KENTII SP. NOV. (ASTERACEAE) During our long spell of working on the British hawkweeds, Cyril West and I had to lump together some species into groups which we were unable to deal with at the time. The largest of these groups was what we referred to as the Hieracium exotericum aggregate. The two botanists who have done the most work on this group are Alex Jordan, whose final account was published by A. Boreau (1857), and N. Hylander (1943). The plant here described as H. kentii belongs to this group and is particularly distinct in the stellate hairiness of the underside of the cauline leaves. We could not match it with any of the species in Boreau (1857) or Hylander (1943). Hieracium kentii P. D. Sell, sp. nov. HOLOTYPE: Busbridge, Godalming, Surrey, v.c. 17, 28 May 1896, E. S. Marshall in Set of British Hieracia by E. F. & W. R. Linton, no. 37, as H. murorum var. pellucidum (CGE). Ab H. grandidenti Dahist. foliis latioribus, pilis stellatis densis in pagina inferiore foliorum caulinorum, involucri squamis pallidioribus, pilis glanduliferis in involucris pedunculisque luteonigrioribus distinguitur. Herba perennis phyllopoda, caudice ramoso. Caulis 40-70 cm altus, pallide luteoviridis, ad basin saepe purpurascens, stricte erectus, striatus, in parte inferiore pilis simplicibus eglandulosis numerosis, sursum gradatim paucioribus, mediocribus vel longis pallidis et in parte superiore pilis stellatis numerosis pilisque glanduliferis numerosis brevissimis vel brevibus luteonigris vestitus. Folia in pagina superiore mediocriter luteoviridia, in pagina inferiore pallidiora neque purpurascentia, in paginis ambabus pilis simplicibus eglandulosis numerosis brevibus_ vel mediocribus, et in marginibus costaque infra longioribus, subrigidis pallidis vestita, caulina pilis stellatis numerosis vel densis infra vestita; folia basalia pauca vel satis numerosa, exteriora 4—10 cm longa, 2-5-6 cm lata, subrotunda vel late ovata vel late elliptica, ad apicem late rotundata mucronuiataque vel obtusa, dentibus magis minusve mammiformibus undulatodentata, ad basin truncata dentium pari basali saepe deflexo, interiora (quando adsunt) 6-10 cm longa, 2-5-5 cm lata, elliptica vel oblongoelliptica, ad apicem subobtusa vel magis minusve acuta, exterioribus similiter dentata, petiolis saepe brevibus, sed interdum ad 8 cm longis et pilis simplicibus eglandulosis numerosis longis pallidis vestitis; folia caulina 0O—1(—2), lanceolata vel ovata, ad apicem acuta vel acuminata, plerumque acute sed interdum dentibus mammiformibus dentata, ad basin truncata, sessilia vel breviter petiolata. Inflorescentia corymbosa, saepe ramo inferiore longo praedita, acladio breviusculo; pedunculi plerumque aliquantum breves gracilesque, pilis stellatis densis pilisque glanduliferis numerosis brevissimis vel brevibus inaequalibus luteonigris pilisque simplicibus eglandulosis sparsis brevibus pallidis vestiti. Capitula 30-35 mm diametro, ad basin rotundata. Involucri squamae ante anthesin incumbentes, 4-9 mm longae, 1-0-1-2 mm latae, olivaceae, interiores marginibus pallidis, anguste linearilanceolatae, ad apicem magis minusve acutum gradatim decrescentes, pilis glanduliferis densis brevissimis vel brevibus maxime inaequalibus luteonigris et secus margines adque basin pilis stellatis numerosis vestitae, sine pilis simplicibus eglandulosis. Ligulae flavae glabrae. Styli paulo obscuri. Receptaculi alveoli margine breviter dentati. Cypselae 3-0—3-5 mm longae, rubronigrae. Phyllopodous perennial herb with a branched stock. Stem 40-70 cm, pale yellowish-green, often purplish towards the base, strictly erect, striate, with numerous, medium to long, pale simple eglandular hairs below, which become gradually fewer up the stem, and numerous stellate and numerous, very short to short, yellowish-black glandular hairs in the upper part. Leaves medium yellowish-green on the upper surface, paler beneath and not purplish, with numerous, short to medium, subrigid, pale simple eglandular hairs on both surfaces and longer ones on the margins and midrib beneath, the cauline with numerous to dense stellate hairs beneath; basal few to fairly numerous, the outer 4-10 x 2-5-6 cm, subrotund, broadly ovate or broadly elliptical, broadly rounded and mucronulate to obtuse at apex, undulate-dentate, the teeth more or less mammiform, and truncate at base with the basal pair of teeth often deflexed, the inner, when present, 6-10 x 2-5-5 cm, elliptical to oblong-elliptical, subobtuse to more-or-less acute at apex, toothed like the outer, the petioles often short, but sometimes up to 8 cm and with numerous, long, pale simple eglandular hairs; cauline leaves 0—1(—2), lanceolate to ovate, acute to acuminate at apex, dentate, NOTES ONT, usually sharply but sometimes with mammiform teeth, truncate at base, sessile or shortly petiolate. Inflorescence corymbose, often with a long lower branch, with the acladium fairly short; peduncles mostly rather short and slender, with dense stellate hairs, numerous, very short to short, unequal, yellowish-black glandular hairs and occasional short pale simple eglandular hairs. Capitula 30-35 mm in diameter, rounded at base. Jnvolucral bracts incumbent in bud, 4-9 x 1-0—1-2 mm, olive green, the inner with pale margins, narrow linear-lanceolate, gradually narrowed to a more-or-less acute apex, with dense, very short to short, very unequal, yellowish-black glandular hairs and numerous stellate hairs along the margins and at the base, without simple eglandular hairs. Ligules yellow, glabrous. Styles slightly discoloured. Receptacle pits with margin shortly dentate. Achenes 3-0-3-5 mm, reddish-black. Flowering May and June. V.C. 13. W. SUSSEX Shady path near Pitts Hill, August 1918, W. C. Barton (BM). Bexley Hill, near Lodsworth, 22 June 1957, F. R. Browning & C. West (CGE). V.C. 14. E. SUSSEX Bexley Heath above Eastbourn, 10 August 1912, C. C. Lacaita (BM). Mickleham, 8 June 1938, H. W. Pugsley (BM). Roadside east of Holtye Common, near East Grinstead, 30 June 1958, B. A. Miles (CGE). V.C. 16. KENT Tunbridge Wells, 1855, A. G. More (BM, CGE). Wood near Shoreham, 29 May 1922, L. B. Hall (BM). Near Ightham, 11 June 1938, F. A. Browning (BM). Ashurst Park, 30 August 1942, H. W. Pugsley (BM); June 1951, C. West (CGE). Shadwell Wood, near Sheldhurst, 6 June 1943, H. W. Pugsley (BM). V.C. 17. SURREY Banks ascending to Hindhead, 4 June 1889, E. S$. Marshall 479 (BM, CGE). Church Lane, Whitley, 7 June 1890, F. J. Hanbury (BM); 28 June 1958, B. A. Miles (CGE). Whitley, 23 May 1896, E. S. Marshall (BM, CGE). Busbridge, Godalming, 28 May 1896, E. S. Marshall (BM, CGE). Under trees, White Hill, near Caterham, 21 June 1906, C. E. Salmon (BM). Wotton, 12 July 1916, C. E. Britton (BM). Margery Wood, Colley Hill, 8 June 1919, C. E. Salmon (BM). Pilgrims Way, Caterham, 28 June 1932, C. E. Britton (BM). Roadside on Gravelly Hill towards White Hill, Caterham, June 1948, C. D. Pigott (CGE). I am very pleased to name this plant after Duggie Kent. Hieracium was not a genus he was particularly interested in, but he spent a large amount of time searching out original publications for me so that we could get their nomenclature straight. His contribution to British botany was prolific over many years and I am personally particularly indebted to him for meticulously reading the proofs of volume 5 of Flora of Great Britain and Ireland. | am indebted to Philip Oswald for supplying the Latin description and diagnosis and to David McCosh and Chris Preston for sorting out and arranging the loan of specimens from the BM. A complete revision of the British Hieracia will appear in volume 4 of Flora of Great Britain and Ireland and is almost finished. REFERENCES BOREAU, A. (1857). Flore du centre de la France, 3rd ed. Roret, Paris. HYLANDER, N. (1943). Die Grassameneink6mmlinge Schwedischer Parke mit Besonderer Beriicksichtigung der Hieracia silvaticiformia. Symbolae Botanicae Upsalienses 7: 106-274. PD SEL Dept. of Plant Sciences, University of Cambridge, Downing St, Cambridge CB2 3EA 218 NOTES CHARA BALTICA BRUZ. (CHARACEAE), BALTIC STONEWORT, REDISCOVERED IN HICKLING BROAD, NORFOLK 5 The stoneworts are a small group of unusually complex aquatic green algae which make up the family Characeae. Baltic Stonewort, Chara baltica Bruz., is one of 16 species on the British Red List (Stewart 1996), and was once widely distributed around the coast of Britain, with records from 14 sites in Cornwall, Devon, Kent, Norfolk, Anglesey and the Western and Northern Isles of Scotland. The species has declined in recent years, and there are only six modern records, some of which are doubtful and need confirmation (N. Stewart, pers. comm.). Its sole site in Norfolk is Hickling Broad, which is the richest site for stoneworts in the UK. The earliest documented record of C. baltica at Hickling was in 1898 by G. R. Bullock-Webster, one of the pioneer workers on charophytes of the British Isles. Bullock-Webster recorded the species again in 1899 and 1904 (specimens at NMW) and in 1902 (specimen in BM), and noted that the plant had “occurred freely for a number of years in company with Chara hispida and Chara aculeolata (now Chara pedunculata) in the northern area of Hickling Broad” (Groves & Bullock-Webster 1924a). Although it was abundant around this time, C. baltica was not observed in subsequent visits to Hickling Broad prior to 1924 (Groves & Bullock-Webster 1924b). There are sporadic reports of C. baltica for the middle of the century; one is attributed to E. A. Ellis in the mid-1940s (Ellis 1965), and G. H. Rocke recorded it in 1955 and 1956 (Wallace 1956, 1957). A sample of the material collected in 1956 is also in BM. The last record of this species at Hickling was made by Rocke in 1965 (Petch & Swann 1968). All historical records are for the variety rigida (originally described by Bullock-Webster) of which Hickling is the type and only known locality. The distinctive characters are supposed to be its firm, rigid habit, comparatively long internodes and patent (standing out at right angles) spine cells (N. Stewart, pers. comm.). Hickling Broad was known to be dominated by stoneworts for the first 60 years of the 20th century, with 15 species being recorded. Water quality subsequently deteriorated due to eutrophication, increasing salinity and accelerating rates of organic sedimentation. The main causes were nutrient input from a large winter roost of Black-headed Gulls in the late 1960s, and deep drainage of the surrounding arable land which breached the saline water table (George 1992). The changes in water quality resulted initially in a periodic explosion in growth of filamentous algae, especially Cladophora sauteri (Nees von Eisenbeck ex Kiitzing) Ktitzing, and the replacement of the charophyte-dominated communities by waterweeds typical of the more fertile conditions. The 1965 record of C. baltica in Hickling Broad was prior to the near total disappearance of aquatic macrophytes and the shift to turbid, phytoplankton-dominated water which accompanied further nutrient enrichment. Charophytes were rarely recorded from then until the late 1970s, when they were an occasional component of the submerged plant communities (Phillips & Moss 1978). By the 1980s, nutrient levels were decreasing due to the decline in the size of the gull roost and progressive washing-out of the residual gull guano from the Broad’s sediment. It is uncertain whether C. baltica continued to survive at very low densities in Hickling after 1965, but the species was not detected by the Broads Authority’s annual macrophyte surveys which commenced in 1982. It may, however, have been overlooked, since these surveys sample only selected transects across the Broad. In 1995 the Broads Authority implemented a weedcutting regime for Hickling to remove Spiked Water-milfoil Myriophyllum spicatum, which was impeding movement of boats across the open water. Intensive sub-aquatic monitoring of the aquatic macrophytes (including charophytes) is an integral part of the weedcutting programme and has yielded much detailed information about species distributions, growth characteristics, physical structure and composition of the plant communities. In June 1998, C. baltica was rediscovered in the northern arm of the Broad which had not been monitored since 1994. This location is consistent with the historical records which were also for the “northern area” (Groves & Bullock-Webster 1924a, b) and suggests that oospores of this species may have survived in the sediment from the 1960s. C. baltica was found to be co- dominant with the Intermediate Stonewort Chara intermedia, forming dense, closed lawns up to 0-8 m tall. The species was only associated with soft, organic sediments and was growing at depths of 0-8 to 1-2 m. Oogonia and antheridia were abundant in June and July on branchlet nodes at the surface of the charophyte lawn which experienced the highest light levels. This species is almost certainly perennial, overwintering with Chara intermedia. Other occasional associates were Chara connivens, Nitellopsis obtusa, Potamogeton pectinatus, Najas marina and Myriophyllum spicatum. NOTES 2119 C. baltica has been recorded predominantly from standing water bodies up to 7 m deep and is usually associated with sandy substrates (Stewart & Church 1992). It is restricted to sites near the coast, which suggests a requirement for salt, and in the Baltic Sea it is known to tolerate salinities up to 1-8% (approximately 10, 000mgCl 1') (Olsen 1944). Salinity levels in Hickling Broad ranged between 1,000 and 2 400mgCl l' for the period 1978 to 1998. Total phosphorus concentrations were approximately 0-1mgP I’ from 1980 to 1990, but showed a steady fall to about 0-05mgP I’! by summer 1998 (Schutten & Davy 1998). There has also been a gradual increase in water clarity in Hickling Broad during the 1990s, culminating with a switch to clear water in summer 1998. Further surveys are needed to determine whether C. baltica maintains its current pattern of abundance and distribution in the northern area of Hickling Broad, or whether it appears only sporadically. Experience has shown that it is sometimes difficult to distinguish sterile material of this species from long-spined plants of C. intermedia. The two species are, however, readily separated by the larger size of the antheridia of C. baltica, which are over 0-5 mm in diameter. ACKNOWLEDGMENTS I am most grateful to Nick Stewart for confirmation of C. baltica, historical records and much helpful advice. Thanks are also due to Dr Martin George for providing historical information about Hickling Broad. REFERENCES ELLIS, E. A. (1965). The Broads. Collins, London. GEORGE, M. (1992). The land use, ecology and conservation of Broadland. Packard, Chichester. GROVES, J. & BULLOCK-WEBSTER, G. R. (1924a). Notes on the British Charophyta. Journal of botany 62: 33- 3i5 GROVES, J. & BULLOCK-WEBSTER, G. R. (1924b). The British Charophyta, Vol. 2. Ray Society, London OLSEN, S. (1944). Danish Charophyta. Chorological, ecological and biological investigations. Det Konggelige Danske Videnskabernes Selskab. Biologiske Skrifter 3(1): 1-240. PETCH, C. P. & SWANN, E. L. (1968). The Flora of Norfolk. Jarcold, Norwich. PHILLPS, G. L. & Moss, B. (1978). The distribution, biomass and productivity of submerged aquatic macrophytes in the Thurne Broads, Norfolk, 1975-1977. Report to the Nature Conservancy Council. SCUTTEN, J. & DAvy, A. J. (1998). Hickling clear; can we understand why? Technical Research Report 98/02. University of East Anglia, Norwich. STEWART, N. F. (1996). Stoneworts — connoisseurs of clean water. British wildlife 8: 93-99. STEWART, N. F. & CHURCH, J. M. (1992). Red Data Books of Britain and Ireland: Stoneworts. J.N.C.C., Peterborough. WALLACE, E. C. (1956). Plant records. Proceedings of the Botanical Society of the British Isles 2: 26-44. WALLACE, E. C. (1957). Plant records. Proceedings of the Botanical Society of the British Isles 2: 245-268. J. E. HARRIS Kepwick Cottage, Wymondham Road, East Carleton, Norfolk, NR14 SJB 220 NOTES THLASPI PERFOLIATUM L. NEW TO WORCESTERSHIRE (V.C. 37) On 13 April 1999 whilst recording near the Gloucestershire border for the forthcoming Flora of Worcestershire, KB found a large population of a Thlaspi species on the edge of a remote arable field, which was tentatively identified as T. perfoliatum. The identification was subsequently confirmed by TCGR and two specimens have been deposited in the National Museum & Gallery of Wales herbarium (NMW). This is the first confirmed record for Worcestershire (v.c. 37). The only previous record for Worcestershire sensu lato is an undated one from Evenlode (c. SP/22.29) by W. Cheshire (Lees 1867), which at the time appears administratively to have been a detached parish of Worcestershire in East Gloucestershire (v.c. 33). This is a remarkable discovery of a rare, declining and statutorily protected species, hitherto surviving in only nine native and three introduced sites in Britain (Rich et al. 1998). The locality is about 20 km from the nearest sites in the Cotswoids, situated in an otherwise unremarkable area with many arable fields and straight, hawthorn-dominated hedges on a low ridge of Blue Lias. The soil was derived from the Blue Lias bedrock with loss of the upper horizons due to ploughing and soil creep, and formed a shallow, stony, greyish clay soil pH 7-3 (measured in a 50:50 slurry with distilled water using a Hanna pHep2 meter). Rich et al. (1989) noted T. perfoliatum was generally only persistent for long periods of time on Oolite Limestone soils, though it had persisted on similar formations (including Blue Lias) elsewhere. The arable habitat is unusual in Britain, the species only having been recorded in arable sites four times previously (Rich et al. 1989). The plants were exceptionally tall and robust, probably resulting from fertiliser. They occurred both in the crop and on the adjacent bank, which presumably acts as a refuge when unsuitable crops are grown. The bank was 1-1-5 m tall indicating a long history of ploughing and soil creep at the field margin. Associated species in the arable crop were Convolvulus arvensis, Polygonum aviculare, Tripleurospermum inodorum and, on the bank, Anisantha sterilis, Arrhenatherum elatius, Bromus hordeaceus, Knautia arvensis, Lamium purpureum, Malva moschata and Senecio vulgaris. The population was estimated to be about 5000 plants, most of which occurred in an area of c. 77 m x 3 m on the edge of the arable field, but significant numbers were also present along an adjacent bank. This is thus the third or fourth largest population in Britain (Rich et al. 1998). The information available to date gives no obvious reason to suspect it is an introduction, and the site is accepted as native. It is close to the known native range on similar geology and soils, and the landscape context suggests it could have survived in situ from the Enclosure of this part of Worcestershire. There were no obvious alien species present, and the large population indicated it had been present for some time. The location has been notified to Plantlife, English Nature and the Worcestershire Trust for Nature Conservation, and efforts are being made to conserve the site. In 1998, a few plants of 7. perfoliatum were also reported from the Gloucestershire Trust for Nature Conservation Reserve of Cutsdean Quarry. They are rumoured to have been sown (pers. comm. C. Studholme to M. and C. Kitchen 1998), but the person responsible is unknown as is the origin of the seed. Plants were last recorded here in 1967 for certain and possibly in 1971, and were lost when Buckle Street was realigned along the west side of the quarry (S. C. Holland, pers. comm., 1988). No plants were seen in repeated searches in recent years by TCGR and co-workers (1983, 1986, 1992, 1994 and 1996). In 1999, there were six plants scattered in the middle of deliberately created bare open banks. REFERENCES LEES, E. (1867). The botany of Worcestershire. Worcester Naturalists’ Club, Worcester. RIcH, T. C. G., KITCHEN, C. & KITCHEN, M. A. R. (1989). Thlaspi perfoliatum L. in the British Isles: distribution. Watsonia 17: 401407. RICH, T. C. G., LAMBRICK, C. R., KITCHEN, C. & KITCHEN, M. A. R. (1998). Conserving Britain’s biodiversity: Thlaspi perfoliatum L. (Brassicaceae), Cotswold Pennycress. Biodiversity and conservation 7: 915-926. K. BARNETT 49 Hastings Road, Malvern, Worcestershire, WR14 2XE | T. C. G. RICH Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff, CF 10 3NP Watsonia 23: 221—230 (2000) 271 Book Reviews A Flora of Norfolk. G. Beckett and A. Bull pp. 320. G. Beckett 1999. Price £38.00. ISBN 0- 9534999-0-1 Just when recent county Floras, such as Hampshire and Cumbria, had set new standards in content, mapping and presentation that seemed unlikely to be easily surpassed, along comes this new Flora, covering all political and Watsonian areas of Norfolk and incorporating bryophytes as well as vascular plants and ferns. It is a superbly integrated work, produced over a relatively short period at a price I consider quite acceptable. The introductory chapters have a particularly valuable section on man’s influence on the various habitats, showing a deep and welcome understanding of the subject. This is supplemented by chapters on habitats using the National Vegetation Classification (NVC) and on soils, reminding readers of the wealth and diversity of habitats, particularly wetlands, created by drift deposits, calcareous and acid, overlaying chalk. There is an excellent section on Norfolk botanists which manages to place predecessors into context rather than just be a recital of names. These introductory chapters end with an excellent and welcome map, an unaccountable rarity in modern Floras. Because of this there is no gazetteer, which would have been useful if only to locate the places named in the illustrations. It would also have been useful to have had an indication of where political and vice-county boundaries differ. The species accounts cover the bulk of the book; vascular plants and ferns, 215 pages, and bryophytes 22 pages. These are supplemented by maps on a tetrad scale for the plants and on a 10 km scale for the bryophytes, showing only records made during the survey (1985 to, I assume, 1998). This latter is my main reservation over the book — the complete lack of older records on the maps, although declines are often noted in the text. By and large I feel that “snapshot” floras, mapping only those species and sites seen during the survey, omit an important dimension by not illustrating change and it is this that is a weak point in recent Floras, of, say, Kent and Wiltshire. Purists might argue that there is a degree of difficulty in allocating tetrads to old records. I think this problem is overstated. In this flora historical maps of species such as Lycopodiella inundata, Lathyrus palustris, Trifolium ochroleucon and many others would have added much to the text. A minor point would have been a wish that the authors had resisted temptation and stuck to their original 1987 start date — to tie in with the BSBI Monitoring Scheme and the new Atlas 2000. A further concern is the extent of incorporation of published records. I noted, following a query on some aquatic plants, that records on Potamogeton compressus, in Scarce Plants in Britain (Stewart et al. 1994) and on Potamogeton X cooperi, in Aquatic Plants in Britain and Ireland (Preston & Croft 1997) were not included. Also, it did not seem that extensive surveys by C. Doarkes (on aquatic plants) or R. J. Driscoll (on Luronium and other subjects) had been used. The text of the accounts is first rate, and is one of the few I would use when writing ecological accounts on a countrywide scale for the new Atlas 2000. There is also a very welcome injection of horticultural knowledge, particularly in the accounts of aliens. The germination details are also very useful and rarely encountered elsewhere. Relatively unimportant criticisms here would include a frequent lack of any comment on native or alien status, numerous errors and omissions (over 25) in the national status of rare plants, and occasional difficulties in tying up tetrad frequencies with the text for the unmapped species. The coverage of Brambles is absolutely first- class, and not at all too long. The maps and the superimposing of the tetrad distribution over coloured soil types where relevant, or over rivers or roads for aquatic or wayside species, are really elucidating and pleasing to the eye. They add explanation to the distribution of all but a very few species. I found it difficult, by eye, to assign a tetrad dot to a precise grid reference and have made myself a little overlay. I wish there were more maps of aliens, but it may be that the authors considered those they mapped told a story, whereas those they omitted did not. Since tetrad totals are given for the species not mapped it would have been useful to have had totals for those mapped. There are many more records, particularly in East Norfolk, of species that were mapped in Scarce plants in Britain 222 BOOK REVIEWS (Stewart et al. 1994). Finally the colour illustrations are first-rate, and incorporated into the text with very little loss of quality — a most pleasing arrangement, and a great improvement over the traditional and obligatory block. [t is not often that my non-botanist friends enthuse over the county Floras that they see strewn around our dining room. A Flora of Norfolk is a rare exception, with the species pages setting new standards in presentation and integration of maps, plates and text. It is instantly accessible, with the coloured underlays of soil types and rivers and roads a real bonus. Just as its predecessor was the first to carry colour (Flora of Norfolk, C. P. Petch and E. L. Swann, 1968), so this takes us into the computer age in new ways that must arouse admiration and trepidation in equal measure in those who would emulate in other counties. D. A. PEARMAN Jilustrations of the British and Irish orchids. D. M. Turner Ettlinger. Pp. 214. Published privately by the author (Royden Cottage, Cliftonville, Dorking, Surrey). 1998. Price £24.50 + £1.10 p & p. ISBN 0—9530380-1-7. My recent review in Watsonia (22: 201-202) of what I termed “an iilustration-free iconograph” by Derek Turner Ettlinger (Notes on the British and Irish Orchids, 1997) has, thanks to the fiscal generosity of an insightful uncle, now been joined by an illustration-rich, genuinely iconographic companion. This welcome A5-sized volume consists of 519 well reproduced photographs organised in 100 colour plates — the images, drawn from a vast reservoir, are excellent. Their primary role is to document the wide range of morphological variation observed within the 49 bona fide species recognised as native by the author; coverage varies from one plate for the cisappoitingly uniform to five for notoriously variable species such as Dactylorhiza incarnata and D. fuchsii, plus E. helleborine if considered to include the genetically undifferentiable E. “youngiana.” Colour morphs, achlorophyllose plants and, where possible, pollinating insects are included, supplemented with a few “accidental” and extinct species. Opposite every plate is a page of information relating to specific photographs; each carries approximate locality, habitat, date, and magnification for close-ups. Here, much space is wasted and the opportunity to highlight diagnostic characters of the taxa is largely eschewed. Rather, brief, footnote-style commentaries are occasionally provided, most relating to conservation sensu lato; examples include under- and over-grazing, de- and re-afforestation, hoof-poaching, draining, vandalism, landfill, development, over-collection and re-introduction. Most taxonomic insights from recent DNA studies (e.g. Pridgeon et al., Lindleyana 12: 89-109, 112-141, 1997) are supported in principle, though in practice the classification of Ettlinger (1997) is followed for consistency (if not brevity, given that it includes 72 often-obscure varieties). The taxonomy is especially brave when dealing with critical genera such as Epipactis and Dactylorhiza. One exception to the uniformity between the two volumes is the upgrading of Stace’s (New Flora 2nd ed., 1997) three “varieties” of G. conopsea to full species on the basis of extensive (albeit mostly unpublished) molecular data, though they remain notional subspecies in the associated account of hybrids. Flicking between selected plates can be instantly informative; for example, it usefully reinforced my prejudices regarding the synonymy of the orchidological “Celtic fringe” represented by the Scottish Dactylorhiza majalis “ssp. majaliformis” and the Welsh “ssp. cambrensis.” The sparse treatment of hybrids (focusing on those between distantly related species) and omission of the many wonderful teratologicai “hopeful monsters”, each a potential species in its own right (Bateman and DiMichele, in D. S. Ingram & A. Hudson, eds., Shape and form in plants and fungi, 63-102, 1994), lead this reviewer to hope that a third volume may one day emerge. In summary, like its predecessor, this book could be viewed as targeted primarily at committed orchidophiles. However, I trust that its extensive use of colour will encourage the wider readership that it merits. R. M. BATEMAN BOOK REVIEWS 223 The Box Hill Book of Box. Lalage Grundy. Pp 334. Friends of Box Hill, Dorking. 1998. Price £3.00*. ISBN 0—9534430—0-0 Box Hill in Surrey was one of the earliest properties of the National Trust, acquired at the turn of the century. In spite of its enormous popularity, it retains a wildness and a rich and varied flora and fauna, including many rarities. Of these the eponymous box (Buxus sempervirens) is locally so plentiful that its special interest deserves to be drawn to the attention of visitors. This attractive, well illustrated and informative booklet with a superb colour photograph on its front cover will surely fulfil this purpose. The booklet is, however, much more than a glossy advertisement. It is written in an engaging style, which should make it accessible to a wide range of people, and it is full of facts about the botanical interest of box: its relatives world-wide, its toxicity to mammals and the specialised invertebrates which nevertheless feed upon it, its extraordinary smell, its geography, ecology, history and its uses to man. The wood of box is valuable and from time to time the trees on Box Hill have been cropped and there is evidence that some areas on Box Hill are old plantations. For this reason it has often been questioned whether a sub-Mediterranean species is truly native so far north, even though a place-name, Boxlands, shows that it was a significant feature in the fourteenth century. There is, however, a remarkable feature of Box Hill which makes the native status much more plausible: the river Mole undercuts the south-western slope to produce a very steep curved amphitheatre facing the afternoon sun. Here, as the author points out, temperatures can rise to 40°C! On this slope, known as the Whites because of its long-established stripes of eroded chalk, box thrives and is dominant over several hectares and forms, with yew and privet, a sombre dark green scrub which is unique in Britain. One of the beautiful photographs shows this special place with its two sentinel whitebeams. This type of booklet has great value, both educationally and for adding to the visitor’s enjoyment, and, as a credit to the National Trust who protect Box Hill, it mirrors the handsome guide-books which are provided for historic houses. Box is of course unusual because it is a rarity that is not endangered by advertisement but the Friends of Box Hill and the author are to be congratulated on setting a precedent which should surely be followed with attractive guides to the geology, vegetation and animals of other open-space properties. C. D. PIGOTT (* obtainable from the Friends of Box Hill, Pixham Mill, Pixham Lane, Dorking, Surrey, RH4 1PQ at £3.50 to include postage and packing.) Three-language list of botanical name components. A. Radcliffe-Smith. Pp. vi+ 143. Royal Botanic Gardens, Kew. 1998. Price £9.99. ISBN 1—900347-50-4. In 1675 John Ray produced his Dictionariolum trilingue, a vocabulary of physical, biological and other terms under 32 headings in English, Latin and Greek, mainly for the use of schools, best known today through the Ray Society’s facsimile edition of 1981. This might well have been the inspiration for the present list, but Mr Radcliffe-Smith mentions in his introduction that he heard of it only in 1984, the year after he embarked on a 16-part list of Greek “roots or elements”, with their Latin and English equivalents, published in Kew’s Nomenclatural forum. This formed the basis of his new list (or rather lists, for, thanks to “the amazing alphabetizing ability of the Spreadsheet’, the list is repeated with the Latin roots first and then with the English meanings first). The original list was apparently compiled “in order that mixtures of Latin and Greek in new compound-coinages might be avoided”, but “an additional purpose, namely to help its users understand the meanings of Latin and Greek compound-coinages already in the literature”, seems of wider relevance, at least for the ‘Greek-first’ and ‘Latin-first’ sections. The main lists, it should be noted, are of ‘building-blocks’ for compound words. Thus “poly- multi- many” and “- phylla -folia -leaved” allow the creation of the specific epithets polyphylla and multifolia but not multiphylla or polyfolia. But it is nowhere explained that in adjectival terminal elements the lists usually give precedence to the feminine form: Lupinus (masculine) must be polyphyllus. There is a strong bias towards initial elements: for example, I looked in vain for “-prasum’” (as in the specific epithets of three Allium species) and ‘“-scordum’” (as in the names of the related genera Nectaroscordum and Nothoscordum). 224 BOOK REVIEWS There follow plant-name and numerical supplements, each again in three sections. It would have been easier to find one’s way about the book if the resulting nine sections had carried running heads. The first supplement contains 297 Ancient Greek plant-names for which plausible Latin and English equivalents can be found, not always according with modern botanical usage (e.g. “daphne laurus laurel”). The second is a mixture of indispensable roots such as “proto- primi- first” and “amphi- bi- two’, useful extras (e.g. “didymo- gemini- twin’), and impossibly complex monstrosities like “diplasiepitetrapempto- duo et quattuor partes [not in fact an element usable in a botanical name, even supposing one wanted to!] two and four-fifths”. It was wise to transliterate the Greek (unlike Ray) “to keep costs down’, but surely also to be accessible to present-day readers, but crazy to do it so literally, with “k” for kappa, “ai” for alpha iota, and “gch” for gamma chi (usually but not consistently). The Romans invented a consistent system of transliteration, which is generally used in botany: thus “chamaikissos” (groundivy) would be better written as “chamaecissos”’, “orobagche” (broomrape) as “orobanche”, and “sogchos” (sowthistle) as “sonchos”’. Another mistake was to present the second and third sections of each part as exact replicas of the first (apart from their word order). This has produced five successive entries for “cup” followed by four for “curl”, with much repetition in the Latin column also, because of the richness of the Greek language. (The Greeks often had several words for it!) Again, because “tip” appears in the ‘Greek- first’ section only as a synonym for “point” (under “-acme’’), it cannot be found alphabetically in the ‘English-first’ section, while there are ten entries there for “point”. So much for that “amazing ability of the Spreadsheet’! The moral is surely not to trust a computer to operate without close supervision. One final grumble: in the plant-name supplement, 13 Latin names have abbreviated (but unexplained) generic names. Especially odd are “p. cerasus” and “pr. cerasus”, with different English and Greek equivalents and separate “alphabetized” places in the “Latin-first’ section. But these are all things which, with a little human interference, could easily be set right in a second edition. Mr Radcliffe-Smith deserves warm congratulations on his stupendous labour and the Royal Botanic Gardens on a tidily presented book at a very modest price. P. H. OSWALD The Plants of Nottingham, A City Flora. P. Shepherd. Pp. v + 76. Wildtrack Publishing, Sheffield. Price hbk. £15.00, pbk. £8.00. ISSN 1354 0270. This is a source of much interest and information. Accounts are given of Nottingham’s very varied geology and of the impact of the urban environment on plants. There are descriptions of the City’s habitats and vegetation, following, as far as possible, the National Vegetation Classification. The final chapter lists, in Kent order, the 700 plant taxa recorded since 1985. Localities are indicated but no grid references given. There are, however, numerous mis-spellings, (including the rather apt “ravishes” of Dutch elm disease), ambiguities and inaccuracies in the text, for example, Saxifraga granulata is said to be “Restricted to Church Cemetery and Forest Recreation Ground .....” and then four more localities are given; Cochlearia danica 1s said to be alien but some known non-natives are not so designated; Sagittaria sagittifolia is called Marsh arrowgrass — and are all the city’s railways really laid on clinker and ash instead of the more usual ballast? At least two cited references are missing from the list. Two sketch maps have no scale and the rivers are represented differently on-all three maps. There is repetition; “Calcicole” is defined twice, though not on first appearance, (but we are never told what a forb is, nor what is meant when Fumaria officinalis is described as a “widespread segetal species of flower beds, cultivated ground and recently landscaped areas”) and I wonder if it is necessary to give both English and Latin names each time a plant is mentioned. I would have preferred to have the parentage of hybrids given in the text and to have the major roads shown on the maps. This very useful publication would have been greatly improved by more rigorous editing but, nonetheless, I look forward with eagerness to the promised historical account of the flora of Nottingham. A. BURNS BOOK REVIEWS 225 John Lindley 1799-1865 gardener-botanist and pioneer orchidologist. Edited by William T. Stearn. Pp. 232. Antique Collectors’ Club in association with the Royal Horticultural Society, Woodbridge. 1999. Hardback. Price £29.50. ISBN 1—85149-296-8. John Lindley is probably most familiar through his association with the Royal Horticultural Society. The Lindley Library has his own library at its core, while the Lindley Medal is awarded for exhibits of special scientific or educational interest at RHS shows. His name is also linked to a few of our native plants, including the eponymous Rubus lindleianus, and many more garden plants — he was responsible for such generic names as Chaenomeles, Photinia, and Victoria. However it is in the world of sumptuous, exotic orchids that he is most notable. John Lindley was born at Catton near Norwich on 5 February 1799, the son of a nurseryman and pomologist — this symposium volume is published to mark that bicentenary. He was a contemporary of William Jackson Hooker, and his life overlapped with that of Sir James Edward Smith. This Norwich-born triumvirate dominated English botany through the early nineteenth century. While Smith’s The English flora (1824-1828) stuck doggedly to the obsolescent, artificial, Linnaean system of classes, Hooker in his Flora Scotica (1821), with Lindley’s help, and later Lindley himself, championed natural orders (families) as a better method for classifying flowering plants. Lindley, appointed Professor of Botany at the University of London in May 1828, went so far as to forbid his students from using any of Smith’s botanical texts because of the archaic classification system, and so he published A synopsis of the British flora; arranged according to the natural orders ... in 1829 (new editions appeared in 1835 and 1841). This celebratory book contains six original essays. The first is an invaluable biography by Professor William T. Stearn, incorporating information from the hitherto obscure archive of family papers now in British Columbia (these papers are the subject of a brief essay by Kathryn Bridge). Lindley the botanist, perpetually writing and publishing, is interwoven with Lindley the family man, administrator and inter alia defender of Kew Gardens. Dr Phillip Cribb discusses Lindley’s contribution to orchidology, concluding that he really was the founding father of orchid taxonomy. To illuminate this, the book is illustrated with 35 colour plates of orchids — there are also six colour plates showing plants from other families. Chris Brickell expands on Lindley’s career as a horticulturist, and Professor W. G. Chaloner discusses The Fossil flora of Great Britain (1831-— 1837) that was jointly produced by Lindley and William Hutton. Dr Brent Elliot, the present Librarian of the Lindley Library, describes the history of the library and John Lindley’s own books. A history of the Lindley Medal by William L. Tjaden is reprinted from Archives of natural history. Miss J. M. Allford’s bibliography of John Lindley, compiled in 1953, forms the tenth chapter and, to quote Stearn, “graphically demonstrates [his] astonishing industry”. Yet this bibliography of 238 works is unlikely to be complete as it takes no account of Lindley’s 20 years as editor of Gardeners’ chronicle which he co-founded. Two of Lindley’s lectures, delivered at the University of London in 1824 and 1834, are also reprinted. All told, this is a scholarly, handsomely produced tribute to a botanist whose studies and publications become even more astonishing when it is remembered that, as a boy, he lost the sight of one eye. E. C. NELSON Alternative Agriculture. A History. From the Black Death to the Present Day. Joan Thirsk. Pp. 365. Oxford University Press. 1997. Hardback £25.00, ISBN 0—19-820662-3. This book can be taken as precisely what its title suggests, a scholarly examination of a rather obscure aspect of social history. The author, now President of the British Agricultural History Society, was formerly a Reader in History at Oxford. It came to me highly recommended by a fellow botanist, and I looked forward to it as a useful reference book, a help with questions about the historic status of plants. We all now accept that our precious and declining arable weeds must largely have arrived from the continent as crop seed impurities, and that plants from monkish gardens and apothecaries’ satchels become long-term denizens, so I certainly hoped that Alternative Agriculture would be useful. I had not anticipated a real delight, a book so rich in interest and information that the writer has joined Pennington, Rackham and Salisbury as one of the authors who has changed my life. 226 BOOK REVIEWS Dr Thirsk writes with a sparse donnish elegance. Each chapter is concise, factual, lucid, and immaculately referenced. Her general argument is that agriculture follows historical cycles. Mainstream agriculture, concentrating on producing beef and corn, reaches a stage when supply exceeds demand, prices drop, and farmers then have to diversify. This thesis is extremely topical, as since the 1980s Britain has been entering an alternative phase, and it is both fascinating and reassuring to read that the current “ruin” moans of farmers have been uttered “at least three times before in our documented history’. The phases of alternative diversity have been roughly 1350-— 1500, 1650-1750, 1879-1939, and we are now in the fourth phase. A glance at the popular press can confirm this current trend, as crops such as borage, linseed, evening primrose and calendula have been attracting notice. For instance oilseed rape, castigated since the 1970s as a modern blight on the landscape, and currently even more of a bugbear as part of the GMO controversy, was actually introduced in the 1560s, and was grown for some 300 years before going out of fashion in the 1880s. Some of the species involved, such as woad, have direct relevance to county recording, as they survive as rare denizens, and some fascinating possible lines of spread (and countries of origin) can be postulated from this book. Others such as madder and saffron are gone, but remain in place names and folk lore. (Saffron caused my only complaint: Latin names are not always given, and I had a panicky little search of other references before I was sure that Crocus sativus was the plant in question, not Colchicum autumnale). The only direct botanical reference in the bibliography is Syme’s English Botany 1863-86, which isn’t perhaps quite familiar to most of us, but the botanical (and in particular horticultural) information is accurate and to the point. Both native plants, such as teasels or weld Reseda luteola, grown commercially for a time to supply changing industries or trade shortages, and more exotic species such as safflower or vines are discussed. The accompanying details are one of the great pleasures of the book, and often illuminate historical county records. Not only is this a clear and readable account of the characteristic crop plants in the British landscape for more than six centuries, but the material for the book is enriched by a multitude of ripping yarns. Diversification produces entrepreneurs, people of vision, imagination, daring, and the new crops they tried, the deals, journeys, tithe wars, quirks of fate make wonderful reading. The larger patterns of wars, peace, international trade, disease and immigration are clearly presented, but within this matrix individual stories have entrancing life. For instance early in the nineteenth century in Lincolnshire, a Mr Cartwright improved his father’s business based on movable woad mills by working a fixed 200 acres by the New Forty Foot Drain near Boston, and the labourers were housed in a specially built hamlet called “Isatica’! Treasures like this abound, with much “serious” information which is directly relevant to field botany, such as the use of Anthyllis vulneraria to fodder rabbit warrens, or the novelty of crop legumes in the seventeenth century (the treatise Sainfoin Improved was printed in 1671). My many unsuccessful searches for plants such as Mentha pulegium or Damasonium alisma were explained when I read that William Cobbet saw 10,000 geese on one Surrey common in the 1820s, where now there may be only dogs on leads. I was also delighted to read of the feminist sensibility of a poultry commissioner in 1896 acknowledging the superiority of French egg production where women are in charge and “men there do not hinder as in some cases here”’. Enthusiasm makes for an unbalanced review, but this really is a book with relevance for most of the botanical community. Tales of individual species, and fashions in land use, and the origins of imported plants, all give clues to current plant distributions. For conservationists, reassurance is perhaps the most welcome lesson. A paradox affecting many of us is that we work to conserve, but believe that it’s already too late. Dr Thirsk’s calm and elegant exposition of “things coming around and around” is most comforting, especially against the current background of millennium hype. And nobody with any curiosity about the oddities of nature, or with a liking for a good story, should resist this book. R. FITZGERALD BOOK REVIEWS 227 Flora of the Russian Arctic, Vol. I, Polypodiaceae — Gramineae, ed. J. G. Packer, transl. G. C. D. Griffiths. Pp. xxxviii + 368. University of Alberta Press. 1995. Hardback $65, ISBN 0-88864— 2695. (A translation of the original Russian edition ed. A. I. Tolmachev.) The ten volumes of the original Russian work were published in softback between 1960 and 198 and cover the vascular flora of the Russian arctic islands from Franz Joseph Land to Wrangel Island and the generally 200-300 km wide mainland from the Murmansk peninsula to the western shores of the Bering Sea. The Flora contains a wealth of information of interest to western botanists but this has remained largely untapped on account of language difficulties. It was, therefore, a bold and imaginative initiative on the part of the University of Alberta Press to publish in hardback an English edition in six volumes under the editorship of Professor J. G. Packer. The first volume covers the first two Russian volumes and deals with the Pteridophyta, Gymnosperms and monocots from Sparganiaceae to Gramineae. In his preface Packer gives projected publication dates for the remaining five volumes concluding in 1998. However, this laudabie and longed-for project terminated prematurely with the publication of volume 2, Cyperaceae — Orchidaceae, in 1996. At least these volumes cover the two largest families in the arctic: Cyperaceae and Gramineae; the latter includes such exclusively arctic genera as Arctophila, Dupontia and Phippsia and is represented in the Russian arctic by at least 162 species. A welcome feature are the detailed keys, sometimes even unduly detailed, and the expansive remarks on many critical species, there being no attempt to produce accounts of uniform length. OF especial interest for western botanists are the long accounts of the critical genera Calamagrostis (including all the British taxa) and Puccinellia. The remarks under Lycopodium annotinum and Huperzia selago should help to focus western attention on their neglected subspecies. There are useful habitat and ecological notes and data on distribution outside Russia. About half the species are provided with dot maps and the distributions of all the species in the 16 districts are summarised by tables in the Appendix. Since nearly 40 years have elapsed since the original Russian edition it is not surprising that several statements are now outdated. The northern limit for Potamogeton (P. filiformis in Greenland) is now 77°N, Puccinellia tenella occurs in Spitsbergen, Poa hartzii in Spitsbergen and Canada, Alopecurus borealis in England, and Poa pseudoabbreviata is not a Russian endemic since it occurs in Alaska. There are few other errors: Hierochloe odorata and Alopecurus aequalis occur in Greenland, Cystopteris dickeana is stated to occur in the mountains of Scotland and for some reason Puccinellia capillaris is said to occur in Britain “only as an introduction”. The species concept is generally similar to that in the west. A glaring exception to this is the inclusion of Festuca baffinensis in F. brachyphylla. No chromosome numbers are cited and there is no indication that only tetraploid Polypodium vulgare occurs in the arctic. The nomenclature has in many cases been superseded. Dryopteris species are erroneously called shield ferns and I was blissfully unaware that Roegneria was the “rhizomeless wheat grass” genus. The format is pleasing and clear although the margins are excessively wide and the running heads are in smaller type than the text. While applauding the appearance of this fine volume one must deeply regret the demise of such an imaginative project. One can only guess at the disappointment experienced by Professor Packer and his able translator and hope that an altruistic publisher will come to their rescue. G. HALLIDAY Carmarthenshire rare plant register. R. D. Pryce. Pp. xvi + 88. Published by the author and available from him at Trevithin School Road, Llaneli, Carmarthenshire, SA15 4AL. 1999. Price £15.00. No ISBN number. Just as the Atlas of the British Flora encouraged the production of many county local plant atlases after its publication in 1962, so too has the publication of national Red Data Books encouraged the production of local equivalents. In some cases these have been published books dealing with rare species in many different plant and animal groups, such as the Red Data Books for Cambridgeshire and Cornwall reviewed in Watsonia 22: 289-290 (1999). Other counties have decided not to follow this route, preferring instead to produce “rare plant registers”, working documents listing the nationally and locally rare species in a county and detailing the sites in which they occur. 228 BOOK REVIEWS The current volume covers the vice-county of Carmarthenshire. It has its origins in a rare plant list produced by Richard Pryce in 1984, “the product of the early days of home computing”. It has been revised and produced as a spirally bound report which makes full use of current computer facilities. It includes internationally rare species, nationally rare and nationally scarce species and species which are rare, scarce or declining in the vice-county. Red text is used to labei international and national rarities and green for the local rarities. Thus the criteria for inclusion are set out clearly, and species such as Liparis loeselii are marked by a plethora of conservation designations, detailed in red. The register is richly illustrated by colour photographs and also includes tetrad distribution maps of the more frequent species. The criteria for inclusion are said to follow those recommended by Perring & Farrell in BSB/ news 71: 10-11 (1996), though they have been interpreted very liberally. One example is that species are included if they qualify under the criteria regardless of whether or not they are native or alien in the county, so that Galanthus nivalis is included as an internationally protected species. even though only native populations are protected and the species occurs in the county as an alien, and Arenaria balearica, Erinus alpinus and Lagarosiphon major are listed as locally rare species. A particularly interesting inclusion is Yucca recurvifolia, thought to have become established on sand dunes from seeds or vegetative fragments which have drifted from North America. If so this would establish the Agavaceae as a family native to Europe! Subspecies, varieties and hybrids are included where appropriate. Occasionally the criteria appear to have been “bent”, as for the orchids Anacamptis pyramidalis and Gymnadenia conopsea, which are frequent on the coast but included as they are rare inland. The accounts of individual species include a list of records made since 1950, detailing the grid reference, native status, date of last record, recorder, site protection details, site name and habitat for each record. The species accounts are a model of clarity, an exception being the baffling treatment of Asparagus officinalis subsp. officinalis, which identifies several native populations of this introduced subspecies and applies to it an English name (“Wild Asparagus”) and a Biodiversity Action Plan priority rating which presumably refer to subsp. prostratus. Many of the accounts include a brief paragraph describing trends in the distribution and current threats to the plant. These provide a valuable insight into the changes taking place in the flora of this area. The individual accounts could have been usefully supplemented by an introductory chapter drawing together these themes, and explaining the various acronyms and initiatives mentioned in the tex: (CADW, Tir Cymen scheme, Millennium Coastal Park scheme) which might be familiar to the residents of the area but less well known outside Wales. This register is designed to be comprehensible to both botanists and others such as county . planners who might need it, and it will undoubtedly provide a useful tool for conservationists in the vice-county. It will be interesting to see how effective it will be in promoting the conservation of the threatened species which are identified. For Lemna trisulca, for instance, will the “urgent remedial measures required by local authority in order to maintain this species in the vice-county” be undertaken in time? In addition, the register will certainly be of considerable interest. to botanists outside the vice-county. Most of the records in this register come from Carmarthenshire Flora Database, built up by recorders working towards a Flora of Carmarthenshire. This publication increases the anticipation with which we await this important Flora. C.D PRESTON Field Flora of the British Isles. C. A. Stace. Pp. xiii + 736. Cambridge University Press, Cambridge. 1999. Price £17.95. ISBN 0-521-—653150-0. The New Flora of the British Isles by C. A. Stace was first published in 1991 and appeared as a second edition in 1997. It has, as predicted, now become the standard British Flora. Only in Ireland, where many botanists remain loyal to Webb’s An Irish Flora, does it have a serious rival. The aim of the new Field Flora is, in the words of the author, “to present the essential data contained in my New Flora...in a more compact and portable form suitable for taking into the field or on holiday”. Thus it includes “the means of identification of all the species and subspecies contained in the’ New Flora” (although I note that no attempt is made to separate the three subspecies of Dryopteris affinis recognised in the latter!). The Field Flora is a light and compact book with a flexible plastic cover. It weighs one pound, less than a third of the weight and occupying less than a third the volume of its predecessor. This BOOK REVIEWS 229 condensation has been achieved by reducing the keys and descriptions which formed the bulk of that book to brief descriptions of the families and genera and to annotated keys. In addition to the details required to key out a species, the keys include a brief indication of the size of the plant as well as ecological and distributional information about the species. Thus the dichotomy which separates two Helianthemum species appears as: 4 Petals white; leaves grey-tomentose on upperside. Stems to 50cm. Native; dry limestone grassland; N Somerset and S Devon White Rock-rose - H. appeninum (L.) Mill. 4 Petals yellow; leaves green-pubescent on upperside. Stems to 50 cm. Native; base-rich grassland; common in suitable places over most of Br, E Donegal Common Rock-rose - H. nummularium (L.) Mill. The hybrid between these species is mentioned in a brief note after the key. Where there is only a single species in a genus the data are presented as a numbered “monachotomy” rather than immediately after the generic description. Thus for Tuberaria: 1 Stems + procumbent to erect, to 30cm but often <10 cm; Native; dry barish ground near sea; W & SW Ir, NW Wa, Jersey and Alderney Spotted Rock-rose - T. guttata (L.) Fourr. The line drawings of individual (often alien) species which appear in the New Flora are omitted, but the comparative illustrations of features such as the leaves of Alchemilla and Sorbus, the utricles of Carex, the fruits of Rumex and the labella of Dactylorhiza are included. Although the format of this Flora is designed for use in the field, the content remains similar to that of the larger Flora. Rare native and alien species are given the same treatment as commoner plants. This is a wise decision, as although these species are less likely to be encountered while undertaking routine recording, they are just the species one might need to check if one thinks one has come across them. As Stace points out, this is a major difference from Clapham, Tutin & Warburg’s Excursion Flora of the British Isies, which orovides descriptions of commoner species but simply keys out the rarer species. Perhaps more debatable is Professor Stace’s decision to retain the same characters to identify the species as are used in the New Flora; no attempt is made to concentrate on characters which are easy to observe in the field. Many of the keys do make good use of vegetative characters, but sometimes (as with Trifolium) I feel that more use could have been made of vegetative characters at the early stages of the key to species. The paradox that sO many species in our flora are instantly identifiable to the experienced botanist on sight but can only be keyed out in a Flora by a series of difficult technical questions will continue to puzzle and frustrate beginners. I would also have liked to see, in addition to the keys which are included in the book, some extra aids such as the vegetative key to commoner grasses included in C. E. Hubbard’s British Grasses. Most ecologists would expect some help with the identification of vegetative grasses in a Field Flora, but little is provided here. All one is told about vegetative characters in the account of Arrhenatherum is that A. elatius is a loosely tufted perennial. It is always easy to suggest additional features which a book might contain. I must therefore make it clear that, having carried the book in the field for several weeks, I already regard it as totally indispensable. Not only does it help identify unknown plants, it is an ideal aide-mémoire for dim-witted botanists like me who can never remember whether Viola reichenbachiana or V. riviniana has the darker spur, and what are the differences between the fruits of Rumex conglomeratus and R. sanguineus. Even though botanists who have bought the first two editions of the New Flora may initially feel they have already paid twice for the information in the Field Flora, it is sure to prove a worthwhile purchase. It is in some ways a more satisfactory book than the New Flora itself, for whereas the absence of extensive descriptions in the larger Flora is disappointing, the format of the Field Flora is perfectly suited to its intended purpose. Whereas I have retained my copy of Clapham, Tutin & Warburg’s Flora to use alongside the New Flora, | am sure that I can now discard my copy of the CTW Excursion Flora. (Indeed, I have already tried to throw it away, but I succumbed at the last minute to sentimental affection for this battered and water-stained companion of 18 years fieldwork.) Professor Stace has again placed us greatly in his debt. C. D. PRESTON 230 BOOK REVIEWS Urban Flora of Belfast. S. Beesley and J. Wilde. A Project of the Belfast Naturalists’ Field Club. Pp. viii + 196. The Institute of Irish Studies and The Queen’s University, Belfast 1997. Price £8.50. ISBN 0 85389 695 X. Ever since writing the Flora of Inner Dublin (Wyse Jackson & Sheehy Skeffington 1984) I find it hard to walk in any city without examining what is emerging from the pavement cracks or glancing at a promising bit of waste ground. I was therefore delighted to be able to review the Urban Flora of Belfast. Belfast is, in many ways, similar to Dublin, as both are large cities on the east coast of Ireland with a large commercial port and therefore comparisons are inevitable. The Flora is very systematically surveyed. The city, which has been divided into 76 one kilometre squares, covers the whole of urban Beliast, much more than the Dublin inner city flora, which covers an area of approximately 9 km’ and is defined along “natural” lines. The Belfast Flora also includes suburban and more rural areas. As with the Dublin city flora, many aliens and garden escapes are present. The criteria for their inclusion are well set out and a list of garden escapes is also included. The Flora inventory was carried out over a short time interval so that it could act as a baseline for future study of this ever-changing environment. Each square was visited at least twice over the period 1993-1995, with seasonal differences taken into account where possible. The total number of species is 592, 63% of which are native. The proportion of non-native species is high, as in inner Dublin, where the number of casuals and aliens was also recorded as approximately 35%. The Dublin city Flora has less records (358), but this is not proportionally less in relation to the area covered (c.9 km? vs 76 km’), which suggests the existence of an intrinsic urban flora, regardless of the area in question. Twenty-five species were found in every square, 4! very predictable (e.g. Taraxacum, Plantago spp., Sonchus spp. and several grasses) and fence 19 one tree, the elder, Sambucus nigra. Though an opportunist, it seems quite an achievement for it to turn up in all 76 squares. Buddleja is also common, as in Dublin, though it is not so common in other Irish cities. There is an impressive list of records new to the counties of Antrim and Down, but aiso many are listed as new to the Census Catalogue of Ireland (Scannell & Synnott 1987). However the majority of these last are adventives, mostly garden escapes ranging from potato Soianum tuberosum to ornamentals such as Mahonia aquifolia and nasturtium Tropaeolum majus. | would not have expected these to be included in the Irish census, which includes native species or “established aliens”. However, it does bring into question whether some, such as the Mahonia, merit inclusion in the Census. I would not consider potato or nasturtium as established, though they probably do get renewed from cultivation very regularly. Other species, such as Sanguisorba. minor are native to Ireland, but are still considered adventive to the area - as is the surprising record for Euphorbia hyberna - almost certainly a garden escape. The introductory chapters deal with the methodology and summarise the results in terms of records per square and new records. A most valuable section discusses each square in turn, listing its main habitats and species of note. This is not only very useful for the current botanist, but also for future comparison of changes. It also gives some insight into the problems of urban recorders, permission being necessary to enter some grounds. In the case of the Balmoral Golf Club, recording involved getting up at 5a.m., presumably to avoid early golfers! Conducting a botanical survey of any inner city area requires a certain amount of courage and great credit is due to the Belfast Naturalists’ Field Club for this survey of a not entirely untroubled spot. There are some interesting asides like the reference to the Malone Road residents who (sadly!) keep their gardens and pavements weed-free! But it is clear too that everyday life, including that of botanists, continued throughout a very rough time, hopefully now relegated to the past. The Flora is very accessible to the amateur or professional, whether resident or not in Belfast and readers should derive much pleasure in dipping into this book. It is not much bigger than An Irish Flora (Webb et al. 1996) and therefore can also be carried about the streets of Belfast. It is also illustrated by some attractive line drawings of plants in a Belfast setting —which are very similar in style to those of the Flora of Inner Dublin. I am therefore disappointed that they do not refer at all to this sister Flora, even as suggested further reading. M. S. SKEFFINGTON RITISH ISLES (B.S.B.1.) ES Ve Bobl The B.S.B.I. traces its origin to the Botanical Society of London founded in 1836 and has a membership of 2,850. It is the major source of information on the status and distribution of British and Irish flowering plants and ferns. This information, which is gathered through a network of county recorders, is the basis for plant atlases and of publications on rare and scarce species and is vital to botanical conservation. The Society published Atlas of the British flora in 1962 and revised editions and is working towards a new atlas, to be completed in 2000. It organises plant distribution surveys, publishes handbooks on difficult groups of plants and has a panel of referees available to members to name problematic specimens. The B.S.B.I. arranges conferences and field meetings throughout the British Isles and, occasionally, abroad. It welcomes as members all botanists, professional and amateur alike. Details of membership and any other information about the Society may be obtained from: The Hon. General Secretary, Botanical Society of the British Isles, c/o Department of Botany, The Natural History Museum, Cromwell Road, London, 5W7 5BD. BSBI Handbooks Each Handbook deals in depth with one or more difficult groups of British and Irish plants. No.1 Sedges of the British Isles A. C. Jermy, A. O. Chater & R. W. David. Revised edition, 1982. 272 pp., with descriptions, line drawings and distribution maps for all 73 species of Carex. Paperback. ISBN 0 901158 05 4. No.2 Umbellifers of the British Isles T. G. Tutin. 1980. 200 pp., with descriptions and line drawings of 73 species of Apiaceae (Umbelliferae). Paperback. ISBN 0 901158 02 X. No.3 Docks and knotweeds of the British Isles J. E. Lousley & D. H. Kent. 1981. 208 pp., with descriptions and line drawings of about 80 native and alien taxa of Polygonaceae. Paperback. Out of print. New edition with distribution maps in preparation; orders recorded. No.4 Willows and poplars of Great Britain and Ireland R. D. Meikle. 1984. 200 pp., with descriptions and line drawings of 65 species, subspecies, varieties and hybrids of Salix and Populus. Paperback. ISBN 0 901158 07 0. No.5 Charophytes of Great Britain and Ireland J. A. Moore. 1986. 144 pp., with descriptions and line drawings of 39 species and varieties of Characeae and 17 distribution maps. Paperback. ISBN 0 901158 16 X. No.6 Crucifers of Great Britain and Ireland T. C. G. Rich. 1991. 344 pp., with descriptions of 148 taxa of Brassicaceae (Cruciferae), 129 of them with line drawings, and 60 distribution maps. Paperback. ISBN 0 901158 20 8. No.7 Roses of Great Britain and Ireland G. G. Graham & A. L. Primavesi. 1993. 208 pp., with descriptions and line drawings of 13 native and nine introduced taxa of Rosa, descriptions of 76 hybrids, and 33 maps. Paperback. ISBN 0 901158 22 4. No.8 Pondweeds of Great Britain and Ireland C. D. Preston. 1995. 352 pp., with descriptions and line drawings of all 50 species and hybrids of Potamogeton, Groenlandia and Ruppia, most of them with distribution maps; detailed introductory material and bibliography. Paperback. ISBN 0 901158 24 0. No.9 Dandelions of Great Britain and Ireland A. A. Dudman & A. J. Richards. 1997. 344 pp., with descriptions of 235 species of Taraxacum, most of them illustrated by silhouettes of herbarium specimens; drawings of bud involucres of 139 species and 178 distribution maps. Paperback. ISBN 0 901158 25 9. Other publications English names of wildflowers J. G. Dony, S. L. Jury & F. H. Perring. 1986 (2nd ed.). 126 pp. Recommended English names for British and Irish vascular plants (Latin-English and English-Latin). Paperback. ISBN 0 901158 15 1. List of vascular plants of the British Isles D. H. Kent. 1992. 400 pp. The same nomenclature and sequence as in Clive Stace’s New Flora of the British Isles (1991, 1997), with selected synonyms. Paperback. ISBN 0 901158 21 6. Supplied with five errata lists and Supplement I (December 1996) of 36 pp. Alien plants of the British Isles E. J. Clement & M. C. Foster. 1994. 616 pp. Lists 3,586 recorded non-native species (of which 885 are established), with English names, frequency of occurrence, status, areas of origin, location of voucher specimens, references to published descriptions and illustrations, and selected synonyms. Paperback. ISBN 0 901158 23 2. Alien grasses of the British Isles T. B. Ryves, E. J. Clement & M. C. Foster. 1996. 234 pp. A companion volume to the last, listing over 700 non-native grasses. Includes keys to bamboos and eight of the larger and more difficult genera and 29 pp. of illustrations. Paperback. ISBN 0 901158 27 5. Plant crib 1998 T. C. G. Rich & A. C. Jermy. 1998. 400 pp. An identification guide for some 325 ‘difficult’ taxonomic groups, with explanations, keys and illustrations of plant details. A4 paperback. ISBN 0 901158 28 3. British Red Data Books I Vascular plants M. J. Wigginton, ed. 1999 (3rd ed.). 468 pp. Up-to-date information and maps for 408 taxa (including 118 microspecies) regarded as threatened in Great Britain, with details of their habitats and associated species, maps showing their pre-1970, 1970—1987 and post-1987 records by 10-km squares, and 1-km square frequency maps for 63 taxa. Hardback, published by JNCC, Peterborough. ISBN | 86107 451 4. Scarce plants in Britain A. Stewart, D. A. Pearman & C. D. Preston, comp. & ed. 1994. 518 pp. Accounts of 254 nationally scarce taxa (occurring in 16-100 10-km squares in Great Britain) and of 71 taxa formerly thought to be so, with details of their habitats and associated species, reproductive biology, changing British distribution and world range; updated distribution maps. Hardback, published by JNCC, Peterborough. ISBN 1 873701 667. Aquatic plants in Britain and Ireland C. D. Preston & J. M. Croft. 1997. 365 pp. Summarises the distribution, habitat and reproductive biology of 200 aquatic plants in 72 genera, with 200 distribution maps and 72 line drawings. Hardback, published by Harley Books, Colchester. ISBN 0 946589 55 0. Available from the official agents for BSBI Publications, Summerfield Books, (John & Sue Atkins) Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX. Fax & Tel: 017683 41577. E-mail: summerfield @iclweb.com 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. 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://members.aol.com/bsbipubs/watsonia. 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: Mr D. A. Pearman, The Old Rectory, Frome St Quintin, Dorchester, Dorset DT2 OHF Plant Records: the appropriate vice-county recorder, who should then send them to Dr C. D Preston, Biological Records Centre, Monks Wood, Abbots Ripton, Huntingdon, PEI7 21S. Obituaries: Mrs M. Briggs, 9 Arun Prospect, Pulborough, West Sussex, RH20 IAL. Back issues of Watsonia are available from the official agents for BSBI Publications, Summerfield Books, (John & Sue Atkins) Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX. Fax & Tel: 017683 41577. E-mail: summerfield @iclweb.com to whom orders for all issues prior to Volume 22 part 1 should be sent. Recent issues (Vol 22 part 1 onwards) are available from Mr M. Walpole, B.S.B.L., 68 Outwoods Road, Loughborough, Leicestershire, LE11 3LY. Watsonia February 2000 Volume twenty three Part one Douglas H. Kent Memorial Issue Contents PRESTON, C. D. & SANFORD, M. N. Douglas H. Kent Memorial Issue of OBITUARY Douglas Henry Kent (1920-1998) ALLEN, D. E. & STACE, C. A. Botanical publications of Deusten. Fees Kent ALLEN, D. E. A selective supplement to British and Irish Herbaria CROMPTON, G. & NELSON, E. C. The herbarium of William Skrimshire (1766- 1829) of Wisbech BOWLT, E. M. & BOWLT, C. Tobi Bieta (713- 1753) a wader apothecary and botanist of his time NELSON, E. C. A history, mainly noneleoian of St ahene S Heath PRESTON, C. D. Engulfed by suburbia or destroyed by the er the ecology of extinction in Middlesex and Cambridgeshire HOLYOAK, D. T. Hybridisation between Rumex heart ie Gall (Polygonacese) and other docks . BAILEY, J. P. & CONNOLLY, A. P. Prize- -winners to ae Knotweed s.l. (Polygonaceae) in the British Isles Q Biseory of j apanese HOLLINGSWORTH, M. L. & BAILEY, J. P. Hybridisation and clonal diversity in some introduced Fallopia species (Polygonaceae) ABBOTT, R. J., JAMES, J. K., IRWIN, J. A. & COMES, H. P. Hybrid origin of the Oxford Ragwort, Senecio squalidus L. JAMES, C. M., WURZELL, B. S. & STACE, C. A. A new or eee a European and a Chinese species of Artemisia (Asteraceae) sae OSWALD, P. H. Historical records of Lactuca serriola L. and L. virosa L. in Baltain with special reference to Cambridgeshire (v.c. 29) BRUMMITT, R. K. & CHATER, A. O. ues amie tic. hybrid in West Wales wee CLEMENT, E. J. a eee x ae E. Je clement a new fataril saves. FLETCHER, R & STACE, C. A. A new section and species of Festuca (Poaceae) naturalized in England ee : OTHER PAPERS FORBES, R. S. Assessing the status of Stratiotes aloides L. Silene ic: in Co. Fermanagh, Northern Ireland (v.c. H33) : WALKER, K. J. The distribution, ecology and conservation of ee norvegica subsp. anglica Halliday (Caryophyllaceae) WILLIAMS, L. R. Annual variations in the size of a Asepelaiion of Coe impatiens L. i sone > “ime NOTES Hanson, C. G. Update on birdseed aliens (1985-1998) Sell, P. D. Hieracium kentii sp. nov. (Asteraceae) Harris, J. Chara baltica Bruz. (Characeae), Baltic Stonewort, valves in Hickling Broad, Norfolk Barnett, K. & Rich, T. C. G. Thlaspi ee L. new to Worcseteichiie e ce 37) BOOK REVIEWS Published by the Botanical Society of the British Isles Typeset by D. K. & M. N. SANFORD Printed in Great Britain by THE BOOK COMPANY, PO BOX 243, IPSWICH, SUFFOLK 139-147 (49-150 14 161-165 167-172 173-177 179-196 197-208 209-212 213-215 216-217 218-219 220 221-230 met WUT NN! A Botanical Society of the British Isles Volume 23 Part 2 August 2000 Editors: M. Briggs, D. L. Kelly, D. R. McKean, D. A. Pearman, M. N. Sanford, D. A. simpson = Botanical Society of the British Isles Patron: Her Majesty Queen Elizabeth the Queen Mother 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 2000—2001 President, Dr G. Halliday Vice-Presidents, Mr M. Walpole, Mr D. J. McCosh, Prof. C. A. Stace Acting Honorary General Secretary, Miss A. Burns Honorary Treasurer, Mr M. E. Braithwaite Editors of Watsonta Papers and Notes, D. L. Kelly, D. R. McKean, M. N. Sanford*, D. A. Simpson Book Reviews, D. A. Pearman Obituaries, M. Briggs *Receiving editor, to whom all MSS should be sent (see inside back cover). © 2000 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 Meconopsis cambrica (L.) Viguier (Welsh Poppy) was drawn by Rosemary Wise. Watsonia 23: 231-235 (2000) Zo Presidential Address, 1999 MARY BRIGGS LIBRARIES THE STUDY OF A LOCAL FLORA An appreciation of the wild plants of the British Isles The theme of this Address is to show how the study of a local flora can lead one to a greater understanding of the larger themes of botany, and how a local botanist can contribute directly to these larger themes from his or her own local knowledge and investigations. It is based on my own experience as a B.S.B.I. member and as County Recorder for West Sussex. The B.S.B.I. entered the technological age in our Atlas of the British Flora (Perring & Walters 1962). For this, records collected by members from all corners of the British Isles were edited by Max Walters and Frank Perring, and punched onto cards to produce maps. There was immediate excitement as patterns of distribution were clearly demonstrated, often for the first time, as the maps appeared. The Atlas led directly to the first legislation for the protection of wild plants in Great Britain in the Wild Creatures and Wild Plants Act 1975, followed by the Wildlife and Countryside Act 1981, with the schedules of protected plants reviewed every five years since. The Society was deeply involved with both these Acts of Parliament and the subsequent Quinquennial Reviews. It is difficult for us now to realise that before the first Atlas, there was no quick reference to the distribution of individual species across the British Isles. Reference to the distribution of each species gave the first general indication of rarity. However, in the conservation field, as far back as the “Battle for Teesdale” in the 1960s, when we attempted - without success - to save the rare plants at Cow Green from inundation by a new reservoir (Lousley et al. 1965; Tutin et al. 1965), we learned then to our disadvantage that learned opinions based on experience were insufficient as legal evidence for Planning Inquiries. In the thirty years since we have steadily endeavoured to assemble our information on plants into a more rigid form, with the statistics required for legislation. This has progressed through the recording methodology used for our Monitoring Scheme sample survey 1987-1988 organised by Tim Rich and reported in Rich & Woodruff (1990). Currently we have another leap forward in recording, record storage and the availability of records planned for the future through our present Atlas 2000 project, and on into the next century with the Threatened Plants Database Project, in which the B. S.B.I. is heavily committed. Inevitably this standardisation of recording format could well lead to the recording of plant distribution becoming largely a number-crunching exercise. A sub-theme of this Address is to show that, in spite of advanced modern technology, there can be more to the study of a local flora than just the accumulation of records. My sub-title proposes that there is as much a place for the wider study of the natural history of our plants, and for the enjoyment of the plants, at the turn of the millennium as ever before. For a deeper perspective and understanding of a local flora, an obvious start is with the geology. Taking Sussex in the context of the Weald, we have a number of habitats: chalk downland, coastal plain, narrow bands of upper and lower greensand and extensive heavy weald clay. In small local areas the clay is lightened by bands of Paludina limestone and Horsham stone (Gallois 1965). In Mesozoic times the Weald was a lake, in which a snail, Viviparus sussexiensis flourished along with dinosaurs. The snails sank on death into the sediment at the water’s edge, and their shells can still be seen now embedded in the local “winkle stone”, which is sufficiently firm to cut into blocks for building. Some local houses were built from this stone, for example Hungerhill House built in 1911 from winkle stone quarried from a nearby field on the farm and the wall blocks clearly show the rounded shells. Winkle stone can also be polished, with the shell outlines giving a marbled marking; it is known locally as Sussex, Petworth or shelly marble. In local churches memorials are framed with polished local “marble”, and many churches have baptismal fonts carved from winkle stone. At the time of the Wealden Lake the vegetation was largely of conifers, cycads, ferns and horsetails; now we look in the bands of winkle stone for plants not common on (Aue 17 wo | 232 M. BRIGGS the surrounding clay, e.g. Adoxa moschatellina, Lathraea squamaria and Asperula cynanchica. A similar local sandstone is Horsham stone, a thinly bedded ripple-marked calcareous sandstone. which was also used in building, as roof tiles for houses built with oak rafters sufficiently stout to carry the weight of the stone. In long-drained Horsham stone quarries the occasional Ophrys apifera has been recorded, far from the usual chalk downland localities for this orchid. Adoxa moschatellina has also been recorded in fields marked “brickfield” on old maps at the Records Office. In the north-west of Sussex there is a narrow wedge of Bargate beds, a sandgate bed in the lower greensand; the map for Fallopia dumetorum in Sussex Plant Atlas (Hall 1980) can be exactly superimposed on the Bargate beds on a geological map of Sussex. The heavy weald clay supported large areas of deciduous forest, and historically Sussex oaks were used to build the naval sailing ships at the time of the Armada. Still today Sussex has a higher percentage of deciduous woodland than most English counties. For many centuries the sticky clay was impassable in the winter months to men, horses and horse-drawn vehicles, and a Domesday Book map of mid-Sussex shows the absence of human settlements along the Forest Ridge. Ifield and Worth are the only place names mapped in the mid-north Weald in 1086 (Morris 1976). A recently discovered link with geological distribution is in Tilia platyphyllos - nationally scarce as a native tree, and until 1987 considered to have been only planted in Sussex. In 1987 Francis Rose realised that a succession of copses of T. platyphyllos discovered along north scarps of the South Downs in northwest Sussex all had features of ancient woodland (Rose 1991). Donald Pigott accepted these as native trees and there is a paper in preparation. Another feature of the local flora is the Mediterranean element. Many Mediterranean species are recorded that are dependent on hot summers for good seed production, this seed in turn being dependent on early spring rainfall for germination. When recording population size of these species from year to year the variation, from a handful of individuals to some thousands in occasional years, is very evident. The unpredictability of orchid populations is well known. This year at a meadow reserve (not in Sussex) a local recorder was apologising for the Orchis morio pictured on a notice listing the plants to be seen, saying that this had now not been seen for many years at this site, when in the meadow that day 20,000 plants of Orchis morio were counted. Similar variations in population sizes in different years are shown by Gentianella anglica in East Sussex, but even more dramatic are figures from Dorset, where in 1994 numbers of this species were estimated at 500,000, but in spite of searches at the same sites in 1996 only 50,000 were recorded. This was referred to by my predecessor, David Pearman, in his Presidential Address (Pearman 1997). To some extent these variations in population size cast doubt on the validity of “snapshot” recording as a monitoring tool. In the New Naturalist monograph on Lords and Ladies by Cecil Prime, the pattern of Arum italicum subsp. neglectum in southern Britain is explained by a map correlating the distribution with the area in which snow lies for less than five days in the year (Prime 1960). Since then a possible weather pattern change due to global warming could extend the range of this plant further north following a retreating snowline, but the further restrictions of habitat requirements to either the springlines between chalk and greensand, or the coastal plain on brick-earth and gravels, could limit any dramatic change. A recent survey of the distribution of this plant in West Sussex was carried out by members of the Sussex Botanical Recording Society and reported by Ron Clough (1998). The author concluded that additional records on the Downs in this survey were almost certainly due to diligent searching rather than an increase in frequency of the plant. Phyteuma orbiculare is totally confined to the chalk downs, and the distribution map in the Sussex Plant Atlas exactly overlies the line of the South Downs. Locally it is known as “The pride of Sussex” — rather inappropriately, as travelling in Europe we find that it is widespread, being recorded from 15 countries in Flora Europaea (Webb et al. 1964-1980). More distinctive to Sussex is the wood of Taxus baccata at Kingley Vale, recognised as the finest yew wood in Europe. Steep chalk downland slopes provide a stronghold for these woods, and at Kingley Vale the slopes are south- and west-facing which, being dryer, are less favourable to the beech and ash trees which are also typical of downland slopes. For many centuries the yew supplied the wood for Britain’s strategic weapon, the longbow; Homer and Virgil had recorded that yew was one of the best woods for bows. On the Reserve there are old gnarled trees and in the main valley at Kingley Bottom there is the dark “sinister and fantastic” forest of Victorian times. Ring counts have shown the older trees now to have been growing for 500 years (Williamson 1978). PRESIDENTIAL ADDRESS 233 In contrast to the strict habitat requirements of these chalk species is Hypericum hirsutum. Thinking of this as a downland plant we were surprised to find that the Sussex Plant Atlas distribution map shows the species as widespread in West Sussex with many fewer downland records in East Sussex. Norman Robson interpreted this for us by pointing out that the species needs not only neutral soil and good drainage, but also some shelter, such as hedgerows and wood edges rather than open fields. The bare downland of East Sussex is too windswept for Hypericum hirsutum. Another Hypericum species, H. perforatum has been in the news as a treatment for depression. Used externally in massage and as an effective wound healer for centuries, modern research has revealed volatile oils, hypericins and prenylated phloroglucinol derivatives, and the plant has been tested for internal use as an anti-depressant with encouraging success. Ashdown Forest was the Sussex stronghold for Genista pilosa, but this species has dwindled in the second half of this century and is now presumed extinct. David Coombe has suggested (pers. comm.) that the heathland ecotype, the form which was found in Sussex and on the Suffolk heaths (where it is also now extinct), was less vigorous than the more robust ecotype of the coastal heaths in Cornwall; the eastern heathland form thus had less potential to recover after fire - a frequent hazard in the Forest. Tim Rich reported on this in the Flora of Ashdown Forest (Rich et al. 1996). Cardamine impatiens, another local plant restricted to woodland on neutral soil and to river edges, germinates in sections of ash wood in the year following coppicing. We found it in cleared soil near my home at Slinfold, when collecting springy sucker shoots of the ash (very suitable for bow and arrow play). The Cardamine flowered in the second year, but by the third was crowded out by the colonising vegetation. This local species is widespread in the northern hemisphere; I have seen it growing in grykes on limestone pavement at Fell End Clouds in Cumbria (Halliday 1997), also in the Swiss and Italian Alps, and from a pony in a Himalayan wood in Kashmir. Sussex is still a stronghold for Carex vulpina, now a Red Data Book species. Only separated in Britain from C. otrubae in 1939, the two species can be difficult to distinguish; fruiting plants apparently have a distinctive “jizz”, but comparison of the epidermal cells of the utricle under the microscope, as described in the B.S.B.I. Handbook (Jermy, Chater & David, ed. 2, 1982) is required by the Carex referees. The sedge is endangered in Sussex through loss of habitat, especially now that so many river edges have been artificially built up with steep flood banks. It is always exciting to find a completely unknown flower. Downingia elegans at Ardingly Reservoir was a complete mystery, until the clues of Hordeum jubatum and a tall robust form of Mentha pulegium (now presumed extinct as a Sussex native) suggested an American species. Botanists at Wakehurst Place, Kew were able to find out for us that the edges of the newly constructed reservoir had been sown with a grass mixture from the U.S.A., designed for fluctuating water levels (Briggs 1986). Mary Elizabeth Parsons (1897) described these plants as growing in wet places, “where they often make the ground blue. The showy white-centred flowers are familiar along the roadsides upon the borders of puddles”. Another introduction is Elodea callitrichoides, a waterweed from South America. The Sussex site in the Chichester Canal may now be the only English record. There is also one in South Wales (Simpson 1986). Historical records have a special fascination. In East Sussex Teucrium chamaedrys was recorded on Camber Castle in 1690. J. Petiver and J. Sherard (1714) wrote “Upon that side of the castle wall that looks to Rye we saw full plenty of chamaedrys”’. It could be seen there still until 1991, but in this century the Castle has been excavated, first by the Ministry of Works, then by the Department of the Environment. Despite the efforts of local plant recorders for 40 years, the last plants were “cleaned off’ the castle walls with herbicide spray in 1991. In 1945 A. W. Graveson found T. chamaedrys in short cliff-top turf at Cuckmere Haven. Francis Rose (pers. comm) accepts this as native by comparison with chalk grassland sites in France. A herbarium sheet in the Department of Botany, The Natural History Museum was first labelled Polypodium vulgare var serratum, “Hbm. of J. Forbes Young MD 1796-1860”, and “from ye old tower of Arundel Castle Sussex”. At the Castle, Sara Rodger researched her archives to identify the “old tower” where the specimen was collected, as the present keep. On a more recent label added to the herbarium sheet, a “redet.” reads: “P. australe Fee R. H. Roberts” - the species now known as Polypodium cambricum L. It was renamed by our current Polypodium referee, R. H. Roberts, as recent work had established the key character of paraphyses mixed with the sporangia, enabling specific distinction. This well demonstrates the value of keeping herbarium specimens in good storage, as reference from the past and for future work. 234 M. BRIGGS As a professional pharmacist, my personal interest in medicinal plants has often involved those which are poisonous. A local primary school phoned for advice when a small boy had eaten Mercurialis perennis, fortunately with no lasting harm as he had been very sick. However, a tale is told in Philosophical Transactions (Anon 1693) of the wife who went into the fields and gathered M. perennis as a pot-herb to serve “Fryed with bacon (having first Boyled them)’. After eating this all the family were ill, the children vomited and were purged and “slept from Thursday to Monday” when sadly the youngest gave “2 sprunts” (convulsions) and died. The father, W. Matthews, overslept too but went to work saying that “he thought his Chin had bin all the Day in a Fire, and (he) was forced to keep his Hat full of Water by him all the Day long, and frequently dipt his Chin in it as he was at Work.” Mother was reported to be “also very sick, but forced to rise to look after her little family concerns”. In 1983, in a modern case of poisoning by M. perennis the doctors described the reddening of cheeks and jaw as a “curious malar erythema”. That time the M. perennis had been eaten in mistake for Veronica beccabunga (an edible herb) (Rugman et al. 1983), highlighting the importance of correct identification before eating. In August 1977 three Dutch (botany) students youth hostelling and self-catering in Sussex added Oenanthe crocata leaves to their stew, assuming it to be Apium graveolens which was familiar to them from their local countryside at home. Although O. crocata is widespread in Sussex, and throughout southern and western Britain and Ireland, it is a very rare plant in Holland. There it was first recorded only in 1975. O. crocata contains highly dangerous oenanthetoxin and all the young men collapsed within ten minutes and were taken to hospital; one seriously ill (after two helpings) and in intensive care, but happily they all recovered. Similarly an Eastern European family working in West Sussex went berry-picking and baked a pie of mostly Atropa belladonna. All were seriously ill, but recovered after hospital treatment. Looking to the future, the most advanced work on molecular botany, using characters which are indistinguishable in the field, still relies in many cases on records from old collections, and living material provided by local botanists. An example is the paper by Johannes Vogel, Fred Rumsey et al. (1999), which describes Asplenium ruta-muraria subsp. ruta-muraria as widespread across Europe and autotetraploid, but the diploid A ruta-muraria subsp. dolomiticum as coinciding in area with the glacial refugia in northern Italy. The diploid subspecies needs at least two gametophytes produced by different spores, giving a slower rate of colonisation which leads to a restriction of diploid populations to ancient glacial refugia, and the colonisation of the rest of Europe by polyploids. Collections of material by travelling and local botanists are critical for this ongoing work on Asplenium. Also looking to the future, B.S.B.I. records will be included in the World Database mapping scheme at The Natural History Museum to provide information on biogeography. This, when the number of sites, sizes of populations, climate and other factors are taken into consideration, will lead to an improved evaluation of threats to rare species for conservation. Petrorhagia nanteuilii is a Red Data Book species with its only British sites in West Sussex. Morphological, cytological and geographical data strongly suggest that P. nanteuilii is an allopolyploid derived from P. prolifer and the non-native P. velutina. The distribution of P. nanteuilili is shown on the Atlas Florae Europaeae map as scattered along Mediterranean coastlines, and on the Iberian Peninsula as widespread both on the coast and inland. In the British Isles there are only two localities, both in West Sussex and coastal, at Pagham Harbour and Shoreham; the latter is an old locality refound since the publication of the Red Data Book ed. 3 (Wigginton, 1999). A most valuable section of B.S.B°I. activities is the publication of journals and identification handbooks; these are a continual source of information on our plants, and most local floras are written by B.S.B.I. members. Also informative, a series of B.S.B.I. surveys has been organised through the years, some on a single genus or species e.g. Mistletoe, Atriplex, Sorbus torminalis and Populus nigra; others on arable weeds, churchyards etc. All members can participate in these network research surveys, which often lead to unexpected plant finds locally. The programme of field meetings and conference meetings has enabled me to visit botanically most areas of the British Isles during 39 years as a B.S.B.I. member, so that although this Address refers mainly to my local county of Sussex I have had good opportunity to appreciate the wide variation in the wild plants of Britain and Ireland. PRESIDENTIAL ADDRESS 235 My family learned when young to identify ‘pin and thrum’ primroses; but only later at B.S.B.L. lectures did I learn the intricate details of their morphology. Jack and Yolande Heslop-Harrison brought scanning electron micrographs of pin and thrum stigma surfaces, showing the long papillae of pin and the short papillae of thrum stigmas, which together with differences in size and shape of the pollen grains effectively ensures that cross-fertilisation is the norm. Further papers with diagrams on this by John Richards (1986, 1989) explained this complex mechanism simply. What wonderful years to have been able to combine family life with the opportunity to learn such detailed information on our plants! But I hope I have shown that, as well as providing such personal enjoyment, careful study of a local flora can contribute to much wider knowledge, and can put local botanists in touch with work of considerable scientific importance. This in turn brings feed-back, leading to a deeper appreciation of the plants recorded locally. Through the B.S.B.I. we all have the means to become active members of the wider botanical community. ACKNOWLEDGMENTS For the use of slides on 8th May 1999 [ would like to thank Miss Alison Paul, Department of Botany; Prof. Chris Humphries, Biogeography and Conservation Laboratory, and the Photographic Studio at The Natural History Museum, London, The British Library, Mrs Sara Rodger, Archivist at Arundel Castle and Dr Alan Knapp. I would also like to thank all who helped with encouragement and support. REFERENCES ANON., (1693). VIII. Part of a letter from Mr. T.M. in Salop, to Mr. William Baxter, concerning the strange effects from the eating Dog-mercury. Philosophical Transactions 17(203): 875-876. Briaccs, M. (1986). Downingia elegans refound in East Sussex. BSBI News 44: 20. BricGs, M. (1990). Sussex Plant Atlas Supplement. Booth Museum of Natural History, Brighton. CLOUGH, R. (1998). Arum italicum (Italian Lords-and-Ladies) in Sussex v.c. 13 & v.c. 14. BSBI News 79: 55-58. GALLOIS, R. W. (1965). The Wealden District. Ed. 4. British Regional Geology. H.M.S.O., London. HALL, P. C. (1980). Sussex plant Atlas. Booth Museum of Natural History, Brighton. HALLIDAY, G. (1997). A Flora of Cumbria. Centre for North-west regional studies, University of Lancaster. JERMY, A. C., CHATER, A. O. & DAVID, R. W. (1982). Sedges of the British Isles. B.S.B.I. Handbook No 1. 2nd ed. Botanical Society of the British Isles, London. LOUSLEY et al., eds. (1965). The threat to Upper Teesdale. Botanical Society of the British Isles, London. Morris, J., ed. (1976). Domesday book 2, Sussex. Phillimore, Chichester. PARSONS, M. E. (1897). The wild flowers of California. Cunningham, Curtiss & Welch, San Francisco. PEARMAN, D. A. (1997). Presidential address, 1996. Towards a new definition of rare and scarce plants. Watsonia 21: 225-245. PERRING, F. H. & WALTERS, S. M., eds. (1962). Atlas of the British flora. T. Nelson, for Botanical Society of the British Isles, London. PETIVER, J. & SHERARD, J. (1714). Adversaria Petiveriana. Journal to Hastings, a journal of a botanical tour from London to Dover etc., by Tunbridge Wells and Hastings. Sloane MS. No. 3340, British Library and (1862) in Phytologist. PRIME, C. T. (1960). Lords and Ladies. Collins New Naturalist, London, and (1981) Pendragon, Cambridge. RICH, T. C. G. et al. (1996). Flora of Ashdown Forest. Sussex Botanical Recording Society, East Grinstead. RICH T. C. G. & WOODRUFF, E. R. (1990) The B.S.B.I. Monitoring Scheme 1987-1988. Nature Conservancy Council CSD Report No. 1265. RICHARDS, A. J. (1986). Plant breeding systems. Allen & Unwin, London. RICHARDS, A. J. (1989). Primulas of the British Isles. Shire Natural History. Shire, Aylesbury. ROSE, F. (1991). Neolithic trees on West Sussex Downs. BSBI News 57: 10-11. RUGMAN, F., MEECHAM, J. & EDMONDSON, J. (1983). Mercurialis perennis (dog’s mercury) poisoning: a case of mistaken identity. British medical journal 287: 24-31. SIMPSON, D. A. (1986). Taxonomy of Elodea Michx. in the British Isles. Watsonia 16: 1-14. TUTIN, T. G. et al., eds. (1964-1980). Flora Europaea. Cambridge University Press, Cambridge. VALENTINE, D. H. (1965). The natural history of Upper Teesdale. Northumberland & Durham Naturalists’ Trust Ltd. VOGEL, J. C., RUMSEY, F. J., SCHNELLER, J. J., BARRETT, J. A. & GIBBY, M. (1999). Where are the glacial refugia in Europe? Evidence from pteridophytes. Biological journal of the Linnean Society 66: 23-37. WEBB, D. A. et al. eds. Flora Europaea. (1964-1980). Cambridge University Press, Cambridge. WIGGINTON, M. J., ed. (1999). British Red Data Books I Vascular Plants, ed. 3. Joint Nature Conservation Committee, Peterborough. WILLIAMSON, R. (1978). The great Yew forest. Macmillan and Reader’s Union, London. Watsonia 23: 237—256 (2000) 237) Mistletoe Viscum album L. (Loranthaceae) on oaks in Britain J. D. BOX 17 Jockey Bank, Ironbridge, Telford, Shropshire TF& 7PD ABSTRACT Information on oaks with parasitical mistletoe (Viscum album L.) was sought from herbaria, county botanical recorders and biological record centres in conjunction with site visits and an extensive search of published records and reports. Currently there are eleven confirmed mistletoe-oaks in Britain. Evidence from past records, reports and general accounts of mistletoe on oak indicates that the rarity of mistletoe-oaks in Britain has not changed since the 17th century. The current results, together with the various reports of mistletoe-oaks in the 19th century and early 20th century, suggest a population of mistletoe-oaks which is apparently fairly stable in number but with some turnover as losses are compensated by the discovery of new sites. The existing mistletoe-oaks are concentrated in Herefordshire and the current range is very much more restricted than the historical range. Pedunculate oak (Quercus robur L.) is the most frequent host amongst the existing mistletoe- oaks. The presence of mistletoe on red oak (Q. borealis Michx. f.) at three separate sites, as well as on a scarlet oak (Q. coccinea Muenchh.), is notable. The estimated ages of the existing oak hosts range from 90 to 400 years. The existing mistletoe-oaks are not found in woodland but prefer hedges, woodland edges, parks and a churchyard. KEYWORDS: pedunculate oak, common oak, parasitism, plant distribution, red oak, Quercus. INTRODUCTION Mistletoe (Viscum album L.) has a wide range of hosts in Britain (Bull 1864a, 1864b, 1907; Somerville 1914; Nicholson 1932; Perring 1973). The commonest hosts are apple (Malus sylvestris (L.) Mill.), lime (Tilia x europaea L.), hawthorn (Crataegus monogyna Jacq.) and hybrid black poplars (Populus x canadensis Moench) (Perring 1973). Despite the traditional association between mistletoe, oaks and Druids in the minds of many people (Frazer 1922; Kanner 1939; Box 1995), oaks have always been considered to be a rare host for mistletoe in Britain (Evelyn 1664; Ray 1670, 1677; Withering 1796; Loudon 1838; Bull 1864a, 1864b, 1907; Nicholson 1932; Tubuef 1923; Perring 1973). The Botanical Society of the British Isles (B.S.B.I.) carried out a survey of mistletoe and its hosts from 1969 to 1972 using both B.S.B.I. members and the general public (Perring 1973). Mistletoe was not recorded on sessile oak (Q. petraea (Mattuschka) Lieblin) and was recorded on pedunculate oak (Q. robur L.) in only twelve 10 km squares in Britain (2-2% of the 539 10 km squares recorded). In contrast, the most recent survey of mistletoe carried out from 1994 to 1995 by Plantlife and the B.S.B.I. produced some 140 sightings of mistletoe on oak, 1-8% of the 8,000+ sightings of mistletoe sent in by the public and by botanists (Briggs 1995), although there are likely to be errors due to the misidentification of oak in winter (Perring 1973). An examination of historical records and the confirmation of existing sites of mistletoe-oaks were undertaken in 1996 and 1997 in order to provide firm evidence of their current and historical distribution in Britain. METHODS B.S.B.I. RECORDERS AND BIOLOGICAL RECORD CENTRES B.S.B.I. vice-county recorders, the national Biological Records Centre at Monks Wood, and local Biological Record Centres (usually county based) in England and Wales were asked for any past and present records. 238 J.D. BOX PUBLISHED SOURCES Mistletoe-oaks referred to in the published literature were followed up, if there was sufficient information to identify the site, through correspondence with the landowner and, if necessary, by a site visit. County Floras for England and Wales, from Victorian times to the present, were examined for references to mistletoe on oak as were local Floras for Bristol (Swete 1854: White 1912) and Chepstow (Shoolbred 1920) which are in mistletoe-rich areas. Given the density of mistletoe records in Herefordshire and surrounding counties (Perring 1973: Briggs 1995), sources of local natural history records were checked: botanical records published by the Caradoc & Severn Valley Field Club (Shropshire) from 1892 to 1970 when the botanical records petered out; the Transactions of the Woolhope Naturalists’ Field Society (Herefordshire) from 1851 to the present; the Transactions of the Worcestershire Naturalists’ Club (1847 to the present); and the Proceedings of the Somerset Archaeological & Natural History Society (1850 to the present). HERBARIA Enquiries were made with those herbaria which were considered to have large collections, to be in mistletoe-rich areas, or to be rich in historical material: Bristol Museum & Art Gallery (BRISTM); Dept of Plant Sciences, Cambridge University (CGE); Royal Albert Memorial Museum, Exeter (RAMM): Royal Botanic Gardens, Kew (K); Leicester Museum (LSR): Liverpool Museum (LIV); Luton Museum (LTN); Maidstone Museum (MNE); Manchester Museum, University of Manchester (MANCH): National Museum of Wales (NMW): Natural History Museum (BM): Norwich Museum (NWH); Daubeny and Field-Druce Herbaria and Morrisonian Collection, Oxford University (OXF); University of Reading (RNG); South London Botanical Institute (SLBD; Warwickshire Museum (WARMS). EXISTING MISTLETOE-OAKS The oaks listed in Table | were visited with the permission of the landowner or occupier. The girth of each tree was measured at 1-3 m above ground level (Hamilton 1975) and used to derive the diameter at breast height (dbh). The age of each tree was estimated using the dbh taking into account the location in which each tree was growing (White 1994); additional estimates of age were derived from published accounts. Mistletoe is dioecious and the mistletoe on each oak was examined for berries in December 1996/January 1997 through 12 x 25 binoculars: mistletoes without berries were checked again in November 1998. Records of the existing mistletoe-oaks in this paper include the 10 km square of the national grid and the nearest village. Further details are deposited with the Biological Records Centre as confidential records. RESULTS BIOLOGICAL RECORDS CENTRE There were 21 records of mistletoe growing on oak. One was the published report of mistletoe on several oaks along the Towy and Cotti rivers in Carmarthenshire (Webster 1885) [see below]. The other records date from 1963 to 1976 with the majority of records dating from 1969 to 1971 as a result of the B.S.B.I. mistletoe survey (Perring 1973). These records were followed up wherever possible in order to confirm the identification of the host tree and its location (Table 2). PUBLISHED LISTS OF LOCATIONS ; Lees (1842, 1851a, 1851b) gives brief details about various sites. Edward Jesse, Surveyor of the Royal Parks & Palaces, lists four sites (Jesse 1844). He made enquiries about mistletoe on oaks in the Royal forests and parks but without success; timber merchants assured him that they had never seen it on oak. Bull (1864a, 1864b, 1907) gives details of various locations; Bull (1869, 1870) gives accounts of newly discovered mistletoe-oaks together with a list of the other known sites (with errors in the spellings). Anon. (1873a) gives a list of sites, and Webster (1885) gives a list without sources (with errors in the spellings of sites) together with further details of one site. Elwes & Henry (1907) give a list as well as some further details for some of the locations; Tubuef (1923) gives a list based on Elwes & Henry (1907). Nicholson (1932) repeats these lists and adds two more reports. Durham (1935) refers to some of these sites, notes that many of the recorded specimens now no longer exist, and adds the report by Anon. (1933). MISTLETOE ON OAKS IN BRITAIN 239 TABLE 1. CONFIRMED EXISTING MISTLETOE-OAKS IN BRITAIN Dbh Estimated Location Oak species (cm) age (yrs)! Brief Description Brampton Bryan Q. coccinea 65 90° One “bunch” of mistletoe (probably male) c. 7 (Herefordshire) m up in the oak which is in a group of three [SO/3.7] oaks in parkland. Bredwardine Q. robur 185 300-400" Two “bunches” of female mistletoe c. 9.5—-10 (Herefordshire) m up in the hedgerow oak. [SO/3.4] Brinsop Q. robur 107 160-170 Three “bunches” of female mistletoe c. 9.5 m (Herefordshire) and c.15 m up in the streamside oak in arable [SO/4.4] land. Deerfold Forest Q. petraeax 142 190-260° One “bunch” of female mistletoe c. 10.5 m up (Herefordshire) Q. robur in the hedgerow oak. [SO/3.6] Eastnor Q. robur 85 210-240° Many female plants c. 8 m up in the woodland (Herefordshire) oak which is adjacent to a drive. [SO/7.3] Frampton-on-Severn (QQ. robur 146 250-270° One “bunch” of female mistletoe c. 6 m up in (Gloucestershire) the hedgerow oak. [SO/7.0] Gwehelog Common Q. robur 97 140-150 One “bunch” of mistletoe (probably male) c. (Monmouthshire) 10 m up in the oak beside a lane in a small [SO/3.0] copse. Leintwardine Q. borealis 63 90 One “bunch” of female mistletoe c. 2m up in (Herefordshire) the churchyard oak. [SO/4.7] Putley Q. borealis 74 110 One “bunch” of female mistletoe c. 8 m up in (Herefordshire) the hedgerow oak. [SO/6.3] Stretton Sugwas Q. robur 180 380’ One “bunch” of female mistletoe c. 3 m up in (Herefordshire) the hedgerow oak. [SO/4.4] Windsor Q. borealis 81 115 One “bunch” of mistletoe (probably male) c. 6 (Berkshire) m up in the roadside oak. [SU/9.6] Notes ' Estimated ages are calculated using the methods set out in White (1994) for common/sessile oaks based on “average site’ conditions supplemented by additional estimates described below. > Planted around 1910. * Range includes the estimated ages calculated using both the present dbh and the girth given in Anon. (1870). * Range includes the estimated ages calculated using the present dbh as well as the girths given by Bull (1869), Anon. (1930) and Tonkin (1984). > Range includes the estimated ages calculated on the basis of “inside normal woodland” conditions (White 1994) using both the present dbh and the girth given by Anon (1866a), as well as the estimated ages derived from the reports by Lees (1842), Bull (1864a, 1864b, 1907) and Anon. (1866a). ° Range includes the estimated age calculated using the present dbh as well as the estimated age derived from the report by Lees (185 1a). ” Age estimated without any allowance for the oak being a very old pollard which has been in a stag’s-horn condition since at least the 1930s. 240 J.D. BOX TABLE 2. RECORDS OF MISTLETOE-OAKS HELD BY BIOLOGICAL RECORDS CENTRE Location (county) Woodyates Manor (Dorset) Hatfield Forest (Essex) No location given (Gloucestershire) No location given (Herefordshire) Deerfold (Herefordshire) No location given (Herefordshire) No location given (Herefordshire) No location given (Herefordshire) No location given (Herefordshire) No location given (Lincolnshire) No location given (Monmouthshire) Neen Savage (Shropshire) Neen Savage (Shropshire) No location given (Shropshire) No location given (Shropshire) Rookery Farm, Drinkstone (Suffolk) NGR SU/01.19 TL/5.1 SO/7.0 SO/3.4 SO/3.6 SO/4.4 SO/6.3 SO/6.3 SO/7.3 TF/0.2 SO/5.1 SO/67.77 SO/67.77 SO/6.7 SO/6.9 TL/95.61 Details of record June 1970. Recorder: M. P. Yule. “Ancient oak tree growing on edge of wood. Two plants obviously dying fast though large’. 1970. Tetrad.J, Recorder: P.C. Devlin (Bishop’s Stortford). 1974. Recorder: S. C. Holland. “In lit 29/10/74”. 1971. Recorder: A. C. Powell. “First recorded in 1870. Hfd. C. C. Preservation Order 1954. Now has 1 large bunch - 7 small — some berries”. 1963. “Extinct c. 1963. Miss M. Bardel in lit. to D. McClintock 4/12/69”. 1971. Recorder: F. H. Perring. 1970+. Recorder: D. Moxley (Abertillery). 1970. Recorder: N. Hadfield (Putley). “Quercus borealis. One bunch”. 1970+. Recorder: F. H. Perring. 1976. Recorder: P. B. Grimes (Grimsthorpe, Bourne). “Oak (only one and the tree has been felled by a gale)”. 1971. Recorder: F H Perring. 1967. “In lit. to D. Mc Clintock”’. 1976. Recorder: Mrs Heywood- Waddington. “Photograph given to F. H. Perring”’. 1967. Tetrad T. Recorder: Thomson in lit. to D. McClintock. 1970. Recorder: A. W. Smith (Shrewsbury). 1971. Recorder: G. Thomson. Current status Followed up in 1997 by letter to the owner of Woodyates Manor (Woodyates, Verwood) without any reply. Followed up in 1997 with P. C. Devlin who has revisited Hatfield Forest but cannot find a mistletoe- oak. This is the Frampton-on-Severn tree. This is the Bredwardine tree. Presumably the Deerfold Forest tree. This is the Stretton Sugwas tree (F. H. Perring, pers. comm.). Followed up in 1997 but Mrs Moxley had moved away. Followed up in 1997 by letter to Mrs Hadfield but without any reply. Presumably the Putley tree. This is the Eastnor tree (F. H. Perring, pers. comm.). Followed up in 1997 but Mr Grimes’ successor at the Grimsthorpe Estate has no knowledge of a mistletoe-oak. Followed up in 1997 but Franklyn Perring does not remember a mistletoe-oak. Presumably the same tree as below. Branch with mistletoe fell off some two years after record sent in (Mrs Heywood-Waddington, June 1997). This is probably the Neen Savage tree: Followed up in 1997 with Mr Smith who cannot remember the oak. Followed up in 1997 with present owners of Rookery Farm who have no knowledge of a mistletoe-oak. MISTLETOE ON OAKS IN BRITAIN 241 TABLE 2. CONTINUED Location (county) NGR Details of record Current status Orvis House, East TM/07.34 1971. Recorder: K. Dodgson. Followed up in 1997 with the Bergholt Mr F. R. Dodgson, the owner of Orvis (Suffolk) House, who remembers mistletoe growing on an oak from the 1930s (when it was artificially planted on the tree) until the late 1970s. There are no other mistletoe-oaks in the vicinity. No location given SO/7.7 1974. Tetrad T. Recorder: F. Followed up in 1997 but Mr Fincher (Worcestershire) Fincher (Bromsgrove). “Prob. Q. had moved away. petraea in T approx 45 metres”’. “T This may be the Bewdley tree. only visited this square to check up on oak, the only Worcs Mistletoe Oak known to me and seen some years ago. The original branch fell off about 3 years ago but Mistletoe has been successfully re-established and is now 3 years old, small but app. healthy. This was the only one to take of 30 attempts, all on oak”. No location given SO/8.5 1972. Recorder: H. M. Caddick Followed up in 1997 with Ms (Worcestershire) (Chipping Camden). Caddick who was not able to provide precise information. No location given SO/8.6 1970. Tetrad Q. Recorder: I. F. Followed up in 1997 but Mrs (Worcestershire) Gravestock (Bristol). Gravestock has died. RECORDS AND REPORTS The sites of records and reports are given below in order of the relevant vice-county. For each site, the first record or report is given. Thereafter, records and reports are only given if they add to the knowledge about the mistletoe-oak; those accounts that just include the site in a general list are not given. The following terminology has been used: Recorded: first-hand observation of mistletoe growing on oak. Reported: second-hand report of mistletoe on oak (these vary from reports of observations by reliable observers to apparent mis-identifications of the host tree). DEVON (V.CC. 3 & 4) Plymouth. Gissing (1855) reports that Edwin Lees had seen a single bush in July 1854 in the topmost branches of a lofty though scraggy oak in the first wood by the South Devon Railway some 2-3 miles north of Plymouth. Dr R. A. Stevens (Nature Conservation Officer, Plymouth City Council) has not found any mistletoe in the small woods along the old railway, which is now abandoned (pers. comm., September 1996). Tiverton. Nicholson (1932) notes a report of mistietoe on oak at Cruwys-Morchard by Mr H. M. Fowler in 1910. The landowner at Cruwys Morchard House has no knowledge of this tree (pers. comm., January 1997). SOMERSET, NORTH (V.C. 6) Bristol. Roper (1916) records a large bunch of mistletoe near the top of Q. intermedia Boenn. [Q. robur x Q. petraea| in Leigh Woods; the oak was of moderate girth, about 50 feet (16 m) in height, and probably more than a century old; it grew in a fringe of ancient woodland on the Somerset side of the Avon gorge. In her Presidential Address to the Bristol Naturalists’ Society on 20 January 1916, Roper records that she had very recently found a magnificent bunch of mistletoe growing near the top of a lofty Q. intermedia in Leigh Woods (Roper 19138). 242 J2D BOX Roe (1981) notes that mistletoe has only ever been recorded on oak in Somerset by I. M. Roper in 1916 in Leigh Woods. Currently, there is no local knowledge of this oak (Tony Robinson, English Nature Site Manager for the Avon Gorge N.N.R., which contains Leigh Woods, pers. comm., July 1996; Tony Titchen, local botanist, pers. comm., November 1996). WILTSHIRE, SOUTH (V.C. 8) Salisbury. A report in Webster (1885) of mistletoe on an oak at Clarendon Park. The estate currently has no records of mistletoe on oak (pers. comm., December 1996). DORSET (V.C. 9) Good (1984) refers to one record of mistletoe on Quercus with no location given. Edmonsham. The Dorset Environmental Records Centre has a record for Edmonsham from C. Didham dated Nov. 1977—Mar. 1978 at SU/062.119 “on oak trees, about 12”. It seems likely that this is the record referred to by Good (1984). A visit to Edmonsham House in February 1997 found three mature limes near the house and one mature lime on the edge of the wood to the north of the house, all with many bunches of mistletoe, but no mistletoe on oak; the owner has confirmed that mistletoe occurs on lime, maple, sycamore and apple but not on oak (Julia Smith, pers. comm., March 1997). Woodyates. Record in the Biological Records Centre dated June 1970 for Woodyates Manor (SU/01.19) annotated “Ancient oak tree growing on edge of wood. Two plants obviously dying fast though large.” HAMPSHIRE (V.CC. 11 & 12) Basingstoke. A herbarium sheet of mistletoe from BM is inscribed “On an Oak in Lord Bolton’s Park at Hackwood, Hampshire. Rev. P. Roberts. 1808”; there is also a note which indicates that this specimen originally came from Sir Joseph Banks’ herbarium. Bull (1864a, 1864b, 1907) refers to a report of a mature oak with mistletoe at Hackwood Park which was described by Sir Joseph Banks in the Philosophical Transactions. This reference could not be found after searching both subject and author indexes to Philosophical Transactions at the Royal Society using various keywords (Banks, Hooper, Lord Bolton, Hackwood Park, mistletoe). Townsend (1883, 1904) refers to another herbarium record of mistletoe - “Hackwood park, on ‘the mistletoe oak’, 1864; Herb. Hill” and indicates that the source for this record was the herbarium of R. S. Hill of Basingstoke which was in the possession of Winchester Natural History Society. Hampshire County Council Museums Service has no knowledge of the Hill Herbarium (pers. comm., Chris Palmer, September 1996) nor does Winchester College which has had a natural history society since 1871 (J. M. Cooper, pers. comm., November 1996). Morris (1917) reports receiving two recent photographs of the tree bearing a bunch of mistletoe. Rayner (1929) states that “The famous plant on the Mistletoe Oak at Hackwood Park has now (January 1928) disappeared, having apparently been carried off during the last few days, though those on adjacent thorns were untouched.” The present owner of Hackwood Park has no knowledge of mistletoe on oak (B. C. Spicer, Agent to the Hackwood Estate, pers. comm., August 1996). Winchester. Britten (1870) refers to mistletoe in K labelled “From the Oak near Winchester, T. O. Duke.” Arnold (1887a) refers to a specimen in BM labelled “From the oak near Winchester, 1870.” No record can be found of the Winchester specimen in K (Sandy Atkins, pers. comm., November 1996). The only herbarium records for mistletoe on oak in BM are from Maidstone (Kent) and Hackwood Park (Hampshire) (Roy Vickery, Curator of Flowering Plants, pers. comm., September 1996). SUSSEX, WEST (V.C. 13) Chichester. Bull (1864a, 1864b, 1907) refers to a report of mistletoe on oak on an estate near Shopwyke, Chichester. Lurgashall. Wolley-Dod (1937) notes a report of mistletoe on an oak near Hempstead Farm, Lurgashall, in 1913 but adds that none was seen in 1915 by E.W.S. on this tree. Northchapel. Arnold (1887a, 1887b) records seeing mistletoe on a branch of oak in Petworth Rectory which had been collected from Northchapel in the Sussex Weald and taken to the Rectory in the 1840s. Northchapel is close to Lurgashall and may have involved the same tree. MISTLETOE ON OAKS IN BRITAIN 243 KENT (V.CC. 15 & 16) Ashford. Bull (1864a, 1864b, 1907) refers to a report of an oak at Mersham Hatch, Ashford, which had been cut down. Maidstone. Bull (1864a, 1864b, 1907) refers to a report from Mr Dickson in 1817 of mistletoe growing on an oak some 4 miles (7 km) from Maidstone by the side of the Medway. A herbarium sheet with mistletoe stems and leaves in BM is annotated “May 20 [28?] 1818. Found in company with Mr Rishon the Mistletoe growing upon the oak about four mills [sic] from Maidstone Kent by the side of the Medway. James Dickson.” [James Dickson was one of the founders of the Horticultural Society of London and of the Linnean Society]. Sheldwich. Anon. (1873b) reports mistletoe on oak at Lord Sondes’ Park, Lees Court. Lees Court is at Sheldwich, just south of Faversham. Sir George Sondes, Earl of Faversham, was the owner in the 17th century and ownership continued in the family until late 19th century (Canterbury Library and Faversham Library, pers. comms., December 1996). Lees Court Mansion House and surrounding gardens were sold in 1971; there is no knowledge of any relevant records neither for the formal gardens nor for the parkland (Lees Court Estate, pers. comm., April 1997). SURREY (V.C. 17) Dunsfold. Bull (1864a, 1864b, 1907) refers to a report of a 150 year old oak with one bunch of mistletoe growing alone in a pasture at Burningfold Farm, Dunsfold. Salmon (1931) refers to the report in Bull (1864b) and adds that “I saw it there in 1917 - oak is Q. pendunculata”’. There is an herbarium sheet for mistletoe at SLBI which includes both a specimen of mistletoe from an apple as well as a twig with oak leaves which are clearly Q. robur (= Q. penduculata) with the annotation “Lower leaves from Oak at Burningfold, Surrey with a twig of the Mistletoe Oak. 28 August 1921. Coll. RWR.” This herbarium sheet was originally part of the London Natural History Society herbarium and R. W. Robbins, of Limpsfield, Surrey, was Recorder of the Botanical Section of the London N.H.S. (Peter Holland, Herbarium Curator, pers. comm., October 1996). Lousley (1976) reports a failure to refind mistletoe on oak in Surrey, including the famous mistletoe-oak at Dunsfold. The present owner of Burningfold Manor, which is the farmhouse, has never seen mistletoe on the mature oaks (M. J. Simmonds, pers. comm., March 1996) Farnham. Bull (1907) in a postscript refers to a report of mistletoe on an oak at the Bishop’s Park at Farnham. Farnham Park was the private deer-park attached to Farnham Castle, home of Bishop of Winchester, and now belongs to Waverley Borough Council. The Ranger has no knowledge of mistletoe on oak (Ron Hills, pers. comm., April 1996). Godalming. Jesse (1844) records part of an oak branch with mistletoe growing on it being sent to him. Bull (1864a, 1864b) refers to this record but was not able to find out about this tree. Norwood. Aubrey (1719; vol. II, p. 34) reports mistletoe on an oak which was felled about 1657. Richmond. Maxwell (1900) notes a report of mistletoe on an oak in Richmond Park which he could not confirm. There is no current knowledge of mistletoe on oak (Jayne Bragham, Assistant Manager, Richmond Park, and Bill Cathcart, Crown Estate Office, Windsor Great Park, pers. comms, June 1996). ESSEX (V.CC. 18 & 19) Britten (1870) reports Turner, without giving the source, as saying “I never sawe more plentye of righte Oke miscel, then Hugh Morgan shewed me in London. It was sent to him oute of Essex: where there is more plentye then in anye other place in Englande that I have ben in.” Epping Forest. Jermyn (1974) notes that mistletoe still persists on oak in Epping Forest. The only mistletoe currently present in Epping Forest is growing on apple, hawthorn and poplar at the north end just outside the Forest boundary (Jeremy Dagley, ecologist with the City of London Corporation which owns Epping Forest, pers. comm., April 1996). Great Dunmow. Jermyn (1974) notes that current records include an oak at Marks Hill, Great Dunmow [the records for this flora no longer exist]. A Phase I habitat survey of this wood in June 1993 found no mistletoe (Adrian Knowles, Essex Wildlife Trust, pers. comm., April 1996) and the owner of this wood has no knowledge of a mistletoe-oak (R. Wallis, pers. comm., April 1996). Hatfield Forest. Record dated 1970 in the Biological Records Centre. The recorder has revisited Hatfield Forest but cannot locate the tree (P. C. Devlin, pers. comm., January 1998). 244 J.D. BOX Woodford/Laughton. Warner (1771) reports that mistletoe is “Found on an oak, between Woodford Row and The Bald Faced Stag, near The Ten Mile Stone: and on an apple-tree in an Orchard in Loughton: and on several trees, many of them oaks, between that place and Mr Conyers’s, Copped Hall.” Woodford, Loughton and Buckhurst Hill (Bald Faced Stag) are adjacent to Epping Forest but are now part of the urban fringe around London and the mistletoe-oaks are long gone (Dr K. J. Adams, B.S.B.I. Recorder, pers. comm., April 1996). BERKSHIRE (V.C. 22) Windsor. Maxwell (1900) notes a report of mistletoe on oak in Windsor Forest which he was not able to confirm. There is a record of one bunch of mistletoe on red oak (Q. rubra, syn. Q. borealis) growing as a roadside tree in Windsor Great Park; probably male as no berries present in January 1997 or November 1998 (Bill Cathcart, Crown Estate Office, Windsor Great Park, pers. comm.). Visit on 12/6/97. Mature Q. borealis growing on light sandy soil with acid grassland and heather nearby; tree is in good condition with a girth of 2:54 m at 1-3 m; one globose bunch of mistletoe hangs down from a lateral branch at some distance from the main trunk and c. 5-6 m above the ground. SUFFOLK (V.CC. 25 & 26) Drinkstone. Record dated 1971 for Rookery Farm, Drinkstone in the Biological Records Centre. There is no local knowledge of a mistletoe-oak at Rookery Farm (D. & H. Donaghy, pers. comm., July 1997). East Bergholt. Record dated 1971 in the Biological Records Centre. The recorder has stated that mistletoe was artificially planted on an oak in the 1930s where it grew until the late 1970s (F. R. Dodgson, pers. comm., June 1997). NORFOLK, EAST (V.C. 27) Aldeby, Woodbastwick & Shotesham. Trimmer (1866) reports mistletoe “On oak at Aldeby and Woodbastwick,...” and ““..on the oak at Shotesham.” CAMBRIDGESHIRE (V.C. 29) Borley Wood. Evans (1939) reports that mistletoe is very scarce in the county, but seems to be native on the oak at Borley Wood. The Ancient Woodland Inventory for Cambridgeshire (1987) indicates that 48 ha of this 50 ha wood have been felled and replanted; the wood has been visited by recorders for the Cambridgeshire Flora Project who have not recorded mistletoe (N. P. Millar, Cambridgeshire Biological Information Service, pers. comm., December 1996). NORTHAMPTONSHIRE (V.C. 32) Rockingham Forest. Bull (1864a, 1864b, 1907) reports that mistletoe formerly grew on an oak near Moorshay Lawn in Rockingham Forest. The current spelling is “Morehay Lawn”. It is a relic from the grassy launds of the medieval compartmented Forest, and there is no mistletoe-oak (Jeff Best, pers. comm., July 1996). GLOUCESTERSHIRE (V.CC. 33 & 34) Riddelsdell, Hedley & Price (1948) report mistletoe on various trees including oak around Cheltenham, on oak at Frampton , on oak at Sedbury Park, and on oak at Drybrook and Aylburton. Holland (1986) notes that the county was well surveyed during the B.S.B.I. Winter Mistletoe Count of the 1970s; refers to the existence of the Frampton oak; but reports that the four other oak trees mentioned by Riddelsdell et al. (1948) have not been located. Cheltenham. Anon. (1873b) reports an account of mistletoe on oak some two miles (3 km) from Cheltenham. Elwes & Henry (1907) were unable to get any confirmation of this report. Chepstow. Bull (1864a, 1864b, 1907) refers to a report of an 85 year old oak on the northern mound of Badams Court, Sedbury Park with one bunch of mistletoe with berries. A visit to Badams Court in October 1996 found an oak (Q. robur, c. 200 years old) on the raised ground to the north of the house, but no mistletoe; the father-in-law of the present owner moved down to Badams Court seventy years ago and has never seen mistletoe on the oak; there is abundant mistletoe (both sexes) on an apple tree next to the house and two bunches on a nearby aspen. 7 MISTLETOE ON OAKS IN BRITAIN 245 Frampton-on-Severn. Lees (1851a, 1851b) reports a first-hand account by Professor Buckman of Cirencester Agricultural College of a century old oak with mistletoe. Bull (1864a, 1864b) refers to a report by Mr Clifford of Frampton Court [the landowner] that the mistletoe was dead and the branch decayed. Subsequently, Bull (1907) refers to a report that the mistletoe had sprouted from the trunk of the oak. Elwes & Henry (1907) report that the mistletoe-oak was still living in 1904 on the authority of H. Clifford [the landowner]. Holland (1986) records that the ancient hedgerow oak (Q. robur) on farmland near Frampton-on-Severn was still standing in 1974 with one large clump of mistletoe near the trunk. Visit in February 1997. Mature Q. robur growing in a hawthorn hedge with pasture on both sides; appears to be an old pollard that has long grown out; in good condition with some dead branches and a major split in the trunk (? lightning) which has healed; girth is 4-6 m at 1-3 m. One straggly/globose bunch of mistletoe growing at base of a main branch on the north side of the tree at c. 6 m above the ground. There were at least five insertions of mistletoe - two on underside of branch, three on east side of branch; berries were definitely present on one of the insertions but it was too late in the season to be certain about the others. MONMOUTHSHIRE (V.C. 35) Goitre. Jesse (1844) reports mistletoe on oak at Penporthleuny in the parish of Goitre. Bull (1864a, 1864b, 1870) refers to this report but notes that the mistletoe had been cut down some 25 years previously. Gwehelog Common. The only mistletoe-oak currently known in v.c. 35 is at Gwehelog Common (Trevor Evans, B.S.B.I. Recorder, pers. comm., May 1996); this is the Q. robur reported without a specific location in Evans (1995). Visit on 15/2/97. Mature Q. robur growing with young hazel, holly, elder and bramble, near other oaks and ash; the tree is on a slope in a small triangular area which appears to have once been common land but is now fenced; the tree is some 50 m north of the fork in the track on the eastern side of the eastern lane; girth is 3-05 m (at nght-angles to main axis of tree) at 1-3 m (measured from ground level on upper side of tree) . One globose bunch of mistletoe with at least five insertions on north side of main trunk at c. 10 m above ground level; probably male as no berries present. Absence of berries confirmed in November 1998 (Derek Parks, pers. comm.). Llangattock Lingoed. Bull (1870) includes a sketch and a description of the mistletoe-oak; the sketch shows an oak with spreading branches growing in a wood. The accompanying text records Q. sessiliflora [= Q. petraea] of around 80 years old with a girth of 4’6” at 5’ (1-37 m at 1-5 m) growing in the hedgerow of a narrow coppice at the Hendre; it had one bunch growing from the underside of a branch. Elwes & Henry (1907) report that this tree is not known to exist at the Hendre. There is an oak (c. 150-200 years old) growing on the east side of the lane some 200 m south of entrance to The Hendre; it has been high pollarded but looks dead and is smothered with ivy; there are two other oaks within 100 m, but both look less than 100 years old. The oak used to be part of a small copse which was cleared for arable land; the family of the current landowner can still remember this being called Gipsy Patch because gipsies would stop under the tree and collect mistletoe up to 20 years ago (Sue Llewelyn, pers. comm., November 1996). St Dials. Jesse (1844) includes St Dials in a list of mistletoe hosts and locations. Bull (1864a, 1864b) refers to this report but notes that the tree had been cut down about 12 years previously. Usk. Mistletoe oak was reported near Usk by Jesse (1844). Bull (1864a, 1864b) refers to this report but notes that the tree had been cut down some five years previously. HEREFORDSHIRE (V.C. 36) Brampton Bryan. Mistletoe was recorded on scarlet oak, Q. coccinea, in Brampton Bryan Park by Geoffrey Messer, a local botanist, in an unpublished MS dated 1988 (Tom Wall, English Nature Site Manager, pers. comm., June 1996). Visit on 30/12/96. A group of three scarlet oaks (Q. coccinea) planted around 1910. The mistletoe is on the northernmost oak; the girth is 2-04 m at 1-3 m. A well developed mistletoe plant growing on the north side of the main trunk at the junction where a side branch has broken off (c. 7 m above ground level); straggly not globose; probably male as no berries present. Absence of berries confirmed in November 1998 (Victoria Harley, pers. comm.). 246 JeDEBOX Bredwardine. Anon. (1870) records the discovery of mistletoe on Q. pedunculata [= Q. robur| about half a mile west of Moccas Park by Sir George Cornewall in conjunction with a photograph and a description of the mistletoe-oak as one of the “The Remarkable Trees of Herefordshire.” The photograph shows the oak as a mature tree with spreading branches which is on the edge of a field: the oak is reported as having a girth of 11°6” at 5’ (3-51 m at 1-5 m), being in a hedgerow, and having mistletoe growing in fifteen different places on the tree. Elwes & Henry (1907) record H. J. Elwes being shown this mistletoe-oak by Sir George Cornewall in 1902. A record dated 1971 in the Biological Records Centre states “Now has | large bunch - 7 small bunches - some berries”. Two bunches of mistletoe were reported on the oak in the 1980s (Stephanie Thomson, B.S.B.I. Recorder, pers. comm., April 1996). The mistletoe-oak is covered by the County of Hereford Tree Preservation Order No. 7 made in 1953. Visit on 31/12/96. Mature Q. robur with many staghorn and dead branches. It is one of a pair of oaks on a fenceline at the top of a steep bank between two fields; the other oak shows signs of pollarding and is not staghorned. Girth of mistletoe-oak is 5-82 m at 1-3 m above ground level on the upper side of the bank. Two bunches of mistletoe were seen on the south-east side of the main trunk at c. 9-5-10 m above the top of the bank; both bunches had berries and neither was globose. Brinsop. Mistletoe on an oak was first seen in the 1990s (David Lovelace, pers. comm., June 1996); an oak with two bunches of mistletoe at the same place was seen in 1993 (Tom Wall, pers. comm., March 1996); and one good bunch of mistletoe was seen on Q. robur by a small stream (Stephanie Thomson, B.S.B.I. Recorder, pers. comm., April 1996). Visit on 31/12/96. Three bunches of mistletoe on mature Q. robur growing on bank of a stream; girth is 3-35 m at 1-3 m. One large globose bunch of mistletoe at c. 9-5 m above ground level which may be more than one plant and is inserted either on the main trunk, or on a small branch which it surrounds, or both; berries present. One of the upper bunches appears dead as the foliage is very brown; there is a third bunch at c. 15 m above ground level which has berries. Deerfold Forest. Bull (1869a, 1869b, 1907) includes a sketch and a description of the mistletoe- oak; the sketch shows an oak with spreading branches growing in a hedge. The accompanying text records one bunch of mistletoe growing on Q. sessiliflora [= Q. petraea] which was in a hedge and was some 50-60 years old with a girth of 5’8” at 5’ (1-73 m at 1-5 m). Anon. (1930) refers to a visit by the Woolhope Field Naturalists’ Club to the mistletoe-oak at Haven Farm which had a circumference of 9’ 9 1/2” at 5’ [2-98 m at 1-5 m]. The oak was estimated to be 130 years old; the mistletoe was observed to be growing in the same spot on the same branch as it was in 1870 although it was no longer near the top of the tree [the reference is to Bull (1869) which was issued in 1870]. A record in the Biological Records Centre states “Extinct c. 1963”. Tonkin (1984) describes a visit to the oak in March 1984, notes that the mistletoe was no longer there, and speculates that the branch with mistletoe was possibly blown off in the gale on 2 January 1976; the girth was measured as 13’ 9” at 5’ (4-19 m at 1-5 m). Visit on 31/12/96. Mature oak (which appears to be a hybrid between Q. robur and Q. petraea) with spreading branches in a hedgerow with hawthorn, blackthorn, honeysuckle, Rosa sp.; the oak has no sign of staghorn branches or disease; girth is 4-47 m at 1-3 m. One bunch of mistletoe comprising two major plants and some minor ones at c. 10-5 m above ground level: berries present. The mistletoe is on the west side of the western of the two main trunks (which divide around 6-5 m above ground level). This is different from the position shown in the 1869 sketch (Bull 1869) but a major branch was lost prior to 1984 (Tonkin 1984). Eastnor. Beaton (1837) records being given a shoot of oak with two mistletoe plants on it from an oak on Earl Somers’ estate at Eastnor Castle. It is reported that there are several mistletoe plants, one of which is of great age and nearly 5’ (1-5 m) in diameter. The editor of The Gardener’s Magazine was sent the oak branch with the letter from D. Beaton and notes that the mistletoe appears to be male and the oak appears to be Q. sessiliflora [= Q. petraea]. Loudon (1838) reports an oak growing near Eastnor Castle with several plants of mistletoe, one of which was of great age and vigour with its branches being nearly 5 feet (1-6 m) in diameter. A large and handsome specimen of this mistletoe was sent to him by Mr Beaton in March 1837 and exhibited on 4 April 1837 at the meetings of the Horticultural Society and the Linnean Society. The former is confirmed by Anon. (1837) who reports that mistletoe on oak from Mr Loudon, to whom it was sent by Mr D. Beaton of Haffield, near Ledbury, was exhibited on 4 April 1837 at the London Horticultural Society. The latter is recorded in the General Minute Book of the Linnean Society for MISTLETOE ON OAKS IN BRITAIN 247 1830-1837, pp. 348-349. Lees (1842, 1851a) records that in June 1837 he saw a young oak, about 70-80 years old, with four fine bushes of mistletoe in Earl Somers’s park, at Eastnor, near Ledbury, on the side of the Ridgeway; enquiries and a careful examination of 300 acres (120 ha) of oak wood revealed no others. Bull (1864a, 1864b, 1907) records a visit to the 80-90 year old oak on the side of the drive leading from the Park up the Ridgeway hill towards Malvern; both male and female mistletoe was present and it grew in three large bunches and in four other places on the tree. Anon. (1866a) includes a photograph and description of the mistletoe-oak as one of a series of “The Remarkable Trees of Herefordshire’; the photograph shows the oak growing in very open woodland (possibly recently thinned) with a yew tree immediately adjacent to it. The accompanying text refers to mistletoe growing in seven places on Q. pedunculata [= Q. robur| which is more than 100 years old and has a circumference of 5’ 3” at 5’ [1-6 mat 1-5 m]. A herbarium specimen of mistletoe held at K is annotated “From the oak in Eastnor Park, Ledbury. Two large plants grow on this oak, both are on one branch and run together. The oak was in young leaf... J. D. Hooker, April 20, 1878”. Hadfield (1974) refers to the mistletoe-oak reported by Loudon (1838) and confirms that it was still present in 1973. Visit on 31/12/96. Mature Q. robur growing in mixed deciduous woodland (ash, yew, oak, field maple and hazel over dog’s mercury) on limestone; the woodland comprises c. 100 year old oak standards over hazel/field maple regrowth; the girth of the oak is 2-67 m at 1-3 m. Luxuriant growth of mistletoe, with many plants inserted into the top and the bottom of branches as well as on the main trunk at c. 8 m above ground level on north/north-east side; berries present. The yew tree next to the oak has a divided main stem and appears to be very similar to the yew in photograph of the mistletoe-oak in Anon. (1866a). Hereford. Anon. (1933, p. xv) reports mistletoe growing on oak in the garden of Vanham House, Pengrove, within the city boundary. It has not been possible to locate Vanham House nor to trace the name in contemporary reference sources and maps. Ledbury. Beaton (1837) writing from Haffield, near Ledbury, notes a report of mistletoe on an oak close to a willow loaded with mistletoe on a farm near Ledbury; the oak had been cut down six years previously. Leintwardine. Report of a mistletoe-oak in the churchyard of St Mary Magdalene, Leintwardine (Paul Hand, pers. comm., December 1996). Visit in December 1996. Red oak (Q. borealis) with a girth of 1-98 m at 1-3 m. One globular bunch of mistletoe made up of a number of separate main stems fused together; inserted on the top, the side and the bottom of an east pointing branch, c. 2 m from the main trunk and c. 2 m above ground level; berries present. Llangarron. Purchas & Ley (1889) report a large oak with mistletoe at Llangunnock Farm, Llangarren (sic) discovered by Mr T. T. Mayos. The owners of Llangunnock Farm for 30 years now live in a new house nearby and have not seen mistletoe on the oaks in the area (Mr & Mrs Jenkin, pers. comm., September 1996). Moccas. There is a report of a tree near Moccas House which was doing well in 1970 with a dozen bunches of mistletoe (Stephanie Thomson, B.S.B.I. Recorder, pers. comm., April 1996). Subsequent inspection of the tree found no mistletoe; but Ivor Saunders, the gardener at Moccas Court, reports that there had been mistletoe on a branch which was blown off in a gale two or three years ago (Stephanie Thomson, pers. comm., February 1997). Morehampton. Bull (1864a, 1864b, 1907) refers to a report of a mistletoe on oak somewhere in the parish of St Margaret’s, Moorhampton (sic). A farmer at Morehampton Park Farm had no knowledge of such a tree (pers. comm., May 1996) neither had the vicar of St Margaret’s, Morehampton (F. E. Rodgers, pers. comm., September 1996). Putley. Record of one bunch on red oak (Q. borealis) by roadside (Stephanie Thomson, B.S.B.I. Recorder, pers. comm., April 1996). Visit on 31/12/96. Healthy, mature red oak (Q. borealis ) with a spreading crown growing by the roadside in a row of six mature red oaks with hawthorn, hazel, ash and Prunus sp.; the girth is 2-31 m at 1-3 m. One globose bunch of mistletoe growing at the base of the main branch on east side of tree at c. 8 m above ground level; inserted on top of the branch; berries present. 248 J.D. BOX Stretton Sugwas. Anon. (1928, page lvi) gives a brief report without details. A herbarium specimen of mistletoe at K is annotated “30.vu1.35. Stretton Sugwas. Viscum “from oak’. Coll. 30.7.35. Recd. August 1935. Dr Durham” [mistletoe on oak at Stretton Sugwas is mentioned without any details by Durham (1935)]. There is a record dated 1971 in the Biological Records Centre (see Table 2). In 1996 the tree is much as it appears in a photograph from the 1930s which shows a stag’s-horn tree (Stephanie Thomson, B.S.B.I. Recorder, pers. comm., April 1996). Visit on 31/12/96. Q. robur growing in a hedgerow; tree is a very old pollard covered in brambles and ivy; upper branches fairly bare, lower ones showed good leaf growth; approx. girth 5-64 m at 1-3 m measured on the field side; one globose bunch of mistletoe on a lower branch at c. 3 m above ground level; many insertions on top and bottom of branch; berries present. Tedstone Delamere. Lees (1851b) refers to a report from Rev. Canon Cradock of two young oaks in the parish of Tedstone-de-la-Mere on banks of Sapey Brook near Knightsford’s Bridge. Bull (1864a, 1864b, 1907) refers to the discovery in 1853 of a mistletoe-oak by Dr Cradock (Principal of Brasenose College); he reports that the oak is 60 years old, has one bunch of female mistletoe, and has been known about by the woodman for some 30 years. Anon. (1866b) includes a photograph and description of the mistletoe-oak as one of the series “The Remarkable Trees of Herefordshire’; the text refers to mistletoe growing in one bunch of five stems some 40 feet (c. 12 m) above the ground on Q. pedunculata (= Q. robur) in the corner of a wood on Primrose Hill; the oak was recorded by Dr Cradock in 1851 and a woodman had known of it for more than 20 years before; a Mr Evans (presumably the landowner) is recorded as cutting the ivy from the tree and promising to remove the trees around the mistletoe-oak. Anon. (1867b) reports a visit to the mistletoe-oak and correspondence with the landowner, Mr Bickerton Evans of Whitbourne Hall, concerning the removal of ivy from the tree. The parish is now incorporated into the parish of Greater Whitbourne. The oaks on the west of the Sapey Brook were felled in the 1950s/1960s (Richard Belville, pers. comm., February 1996); the owners of the woodlands on the other side of the Sapey Brook have looked at the oaks around Primrose Cottage, but found no mistletoe (pers. comm., Julia & Bill Evans, April 1996). Tretire. There is a herbarium specimen of mistletoe in BRIST held in BRISTM; annotated “Viscum album on large oak tree in meadow on Tredunnock Farm, Tretire, Herefordshire. Sept. 29, 1927. E. Armitage” and “I saw the mistletoe on this tree many years ago, & there are a large number of branches scattered about in the tree. E. A.” [Eleanora Armitage studied the flora of Herefordshire and Gloucestershire]. The current owner of Tredunnock Farm has no knowledge of any oaks with mistletoe; there may have been a fairly old oak in the hedgerows which were removed to make way for the airstrip at the back of the farm (Julie Smith, pers. comm., March 1996). Woodbury Hill. Cornewall (1903, pp. 104-105) records that he discovered a mistletoe-oak in January 1903 on the west side of Woodbury Hill on the Moccas Estate. He estimated the age as 220 years old and noted that the tree was picturesque and gnarled; it was in company with a row of large limes which had been felled. There has been woodland thinning and conifer planting in the area, and the oaks on Windy Ridge are only around 75 years old (Simon Quarn, farm manager at Wilmaston Farm, pers. comm., June 1996). WORCESTERSHIRE (V.C. 37) Bewdley. Rea (1910, p. 272) records a mistletoe-oak growing on the county boundary at Dowles parish. Salter (1928) records one from the grounds of Mr J. Steele Elliot, Dowles Manor, Bewdley. A record by L. C. Lloyd is reported in Dallman (1939) as “A flourishing plant of some years’ establishment on an Oak tree (Q. robur L.) in Mr J. Steele Elliott’s grounds at Dowles Manor, just inside the Shropshire border (Lloyd). The only instance of its growth on oak so far recorded in the county” [the oak is in fact just in Worcestershire - see below]. John Robinson (English Nature Site Manager, Wyre Forest N.N.R.) was shown this oak by Steele Elliot’s daughter, Mrs P. Tristram-Ede (pers. comm., December 1996). She can recall one very large bunch of mistletoe on it in the 1940s & 1950s which was too high to pick. This was on a branch which is now dead and has ropes for a swing hanging from it. Visit in December 1996. It is a 180/200 year old oak which is a hybrid between Q. robur and Q. petraea, although tending towards Q. petraea. The oak grows at the top of the north bank of the MISTLETOE ON OAKS IN BRITAIN 249 Dowles Brook at a point which is just in Worcestershire (the county boundary deviates from the brook at this point). The oak is in a stream corridor and near to a field. Bromyard. Anon. (1878) records two oak trees with mistletoe in Buckenhill Coppice, near Bromyard. Droitwich. Report of Q. robur which used to have a single bunch of mistletoe, but the branch with the mistletoe dropped off a couple of years ago; the oak was in the hedgerow which ran along Green Lane, Hadzor, just to the east of the M5 at Droitwich (SO/910.622) (Andrew Fraser, pers. comm., February 1996). Site visits in March and September 1996 found two oaks in the hedge on the west of the lane, but no sign of mistletoe, although there is an abundance of mistletoe on limes in the vicinity. Knightsford. An herbarium specimen of mistletoe at K is annotated “Viscum album from oak. Knightsford Worcester May 1875. ex Herbario Joannis Fraser MD. Presented by the Trustees of the Shrewsbury Public Library, June 1954.” The bridge on the main road over the River Teme at Knightwick is Knightsford Bridge. It is probable that this herbarium specimen came from the Knightwick oak (see below). Knightwick. Anon. (1871) records mistletoe on oak near Knightwick Church and reports that this is a recent discovery. A local resident describes an oak between Knightwick and Suckley which has always been known as the “mistletoe oak’; this tree is some 400 m from the old church which was replaced by a mortuary chapel about 100 years ago; she does not remember mistletoe growing on it and neither did her parents who were both born in the parish in 1885 (Mary Walker, pers. comm., January 1997). A visit in February 1997 found a mature Q. robur (c. 200 years old) in a hedgerow on east side of the road from Knightwick to Suckley, approx. halfway between Knightwick Manor and the main road. There was no sign of mistletoe on the oak and there were no other mature oaks around. Lindridge. Anon. (1872) confirms a report of mistletoe on oak. Stoulton. Anon. (1884) records mistletoe on oak; this was still extant in 1901/1902 (Rea 1903). STAFFORDSHIRE (V.C. 39) Stafford. An herbarium sheet with a stem and leaves of mistletoe in the Morisonian collection, OXF, is annotated “Viscum quercinum ex agro Staffordiens, donum D. Thurstin, Orielens. 1690” [Translation: Mistletoe of oak from a field near Stafford, given by D. Thurstin of Oriel College]. SHROPSHIRE (V.C. 40) Sinker et al. (1985) indicate that 2% of the mistletoe records for the county are associated with oaks without giving further details [in fact the data came from the 1969/70 B.S.B.I. mistletoe survey (lan Trueman, pers. comm., January 1996)]. Neen Savage. Records dated 1967 and 1976 are in the Biological Records Centre. The branch with mistletoe fell off some two years after 1976 record was sent in (Mrs Heywood-Waddington, pers. comm., June 1997). A visit to the oak in June 1997 found no mistletoe. WEST GLAMORGAN (V.C. 41) Vale of Neath. Bull (1864a, 1864b, 1907) refers to a report of an oak which was 2 miles (3 km) below Aberpergwm House but which was blown down in a storm. CARMARTHENSHIRE (V.C. 44) There are reports of mistletoe on several oaks in the remains of the old forest along the courses of the Towy and Cothi rivers (Webster 1885). May (1967) cites this report in an appendix of species presumed lost or extinct in Carmarthenshire. There are no recent records of mistletoe from the Tywi or Cothi valleys nor any current records of mistletoe on oaks in Carmarthenshire (Pryce 1996). CARDIGANSHIRE (V.C. 46) Gogerddan. Salter records a mistletoe-oak on 26/12/27 in his unpublished diaries (National Library of Wales MS 14445B) and records that the mistletoe-oak was adjacent to a tall poplar which also had mistletoe (Salter 1928). Subsequently, he mentions the poplar as being felled and refers to mistletoe growing on lime, but makes no reference to the oak (Salter 1935). Currently, there is no mistletoe present at this site (A. O. Chater, B.S.B.I. Recorder, pers. comm., September 1996). 250 J.D. BOX MONTGOMERYSHIRE (V.C. 47) Trueman et al. (1995) report that a recent survey found mistletoe on oak, but no details are given. The records came from David Orton who undertook a mistletoe survey in 1993/94 (Marjorie Wainwright, B.S.B.I. Recorder, pers. comm., January 1996). Abermule. Report from 1993/94 of a very small mistletoe plant on Q. robur at Penybryn; the mistletoe is not growing there now (David Orton, pers. comm., June 1996). CAERNARVONSHIRE (V.C. 49) Anglesey. Lees (1842, 1851a) notes that Mr J. F. Dovaston reported in Loudon’s Magazine of Natural History (vol. 5, p. 203) that he saw mistletoe growing on oak and “what is more singular, hanging almost over a very grand druidical cromlech” in the Marquis of Anglesea’s park at Plas Newydd, on the island of Anglesey. It has not been possible to find the original report by Dovaston. Prior to 1842, there are two complete series of Loudon’s Magazine of Natural History but there is no reference to Dovaston, nor to mistletoe, in any of the volumes. There is no reference to Dovaston writing about mistletoe in the Catalogue of Scientific Papers 1800-1863 compiled by the Royal Society of London (G. A. Knowles, Manager, Current Information Services, The Royal Society, pers. comm., October 1996). There are no longer any trees near the cromlech at Plas Newydd; most of the trees are believed to have been felled around the turn of the century and the last by the 1950s (National Trust, pers. comm., March 1996). LEICESTERSHIRE (V.C. 55) Leicester. Report by Pulteney (1795, p. clxxix) “In Quorn Wood. On the oak, and other trees. On apple-trees in old orchards.” Pulteney (1795) contains a lot of records from Buddon Wood; the entry for Rubus idaeus (p. clxxx1) states that it is found in Quorn or Buddon Wood and the names were evidently interchangeable. Buddon Wood lies just to the south of the village of Quorn. It is a large ancient wood which was clear-felled soon after the Second World War and mistletoe has not been seen since Pulteney’s day (Michael Jeeves, B.S.B.I. Recorder, pers. comm., September 1996). NOTTINGHAMSHIRE (V.C. 56) Ordnoyo (1807) notes that mistletoe occurs on various hosts including oak but gives no details. Sherwood Forest. Maxwell (1900) includes a report which he could not confirm. There is no current knowledge of mistletoe on oak (Martin Freeman, Sherwood Forest Visitor Centre, pers. comm., February 1997). DERBYSHIRE (V.C. 57) Staveley. Arnold (1887a) reports, without giving a source for the quotations, that “in one of Colepeper’s MSS. at the British Museum, in a curious notice of Sir Peter Freschville’s house at Staveley, Derbyshire, is this passage:- ‘Heare my Lord Freschville did live, and heare grows the famous Mistletoe tree, the only oake in England that bears Mistletoe.’ And to this tree the following letter, written between 1663 and 1682, from the Countess of Danby to Mrs Colepeper, probably refers: - ‘Dear Cozen, -Pray if you have any of the mistleto of yo" father’s oke, oblige me so far as to send sum of it to yo" most affectionat servant, Bridget Danby.’”’ The British Museum has a large collection of letters to and between Colepeper and his wife, including the undated note of Lady Danby to Mrs Colepeper: “Deare Cosen / pray if you haue any of the Miselto of yo' fathers oke oblidge me so fare as to send sume of it to y' most / affectionat sarviant / Bridgett Danby” (folio 46, Harley MS 7005). Colepeper’s lengthy notes relating to Staveley occupy folios 139-289 of Harley MS 7602, the volume covering the letter ‘S’ of his large collection of miscellaneous notes and observations. A rapid search of the volumes covering both ‘M’ for mistletoe and ‘S’ for Staveley has not revealed any mention of mistletoe (W. H. Kelliher, Curator of Manuscripts, The British Library, pers. comm., December 1996). [See record for Sheffield under Yorkshire, south west (V.C. 63)]. YORKSHIRE, SOUTH WEST (V.C. 63) OR DERBYSHIRE (V.C. 57) Sheffield. A report of an oak with mistletoe near Sheffield in the the 2nd edition of Ray’s Catalogus Plantarum Angliae (Ray 1677) contains details which were omitted from the Ist edition MISTLETOE ON OAKS IN BRITAIN 251 (Ray 1670). Ray (1677) has “Siquis viscum in Quercu crescentem videre desiderat, Sheffeldiam adeat, a qua non procul inveniri eum a D. Fr. Jessop certior factus sum” [Translation: Anyone who wants to see mistletoe growing on an oak should go to Sheffield, not far from where I have been informed by D. Fr. Jessop, it can be found] whereas Ray (1670) only has “Siquis viscum in Quercu crescentem videre desiderat’”’. Francis Jessop (1638-1691) of Broomhall, Sheffield, met John Ray in 1668 and Ray published some of Jessop’s plant records (Desmond 1977). It seems probable that this is the same tree as that referred to in the 17th century accounts concerning the mistletoe-oak at Staveley in Derbyshire (Arnold 1887; Folio 46, Harley MS 7005, British Museum); Staveley is only some 14 km from the centre of Sheffield. PERTHSHIRE, EAST (V.C. 89) Errol. Kanner (1939, p. 913) reports an old account of mistletoe growing in profusion on a vast old oak in the neighbourhood of Errol, not far from the Falcon stone; the fate of the family of Hay was reputed to be linked to the continued existence of this tree [see also Gurney (1848, pp. 576- 577) and Melville (1935, p. 156)]. Kanner (1939) also notes that the 13th century Scottish poet, Thomas Rymour of Ercildowne (Thomas the Rhymer, 1220-1297), is credited with the authorship of a poem which deals with the connection between this mistletoe-oak and the fate of the Hay family: While the mistletoe bats on Errol’s aik, And that aik stands fast, The Hays shall flourish, and their good grey hawk Shall nocht flinch before the blast. But when the root of the aik decays And the mistletoe dwines on its withered breast The grass shall grow on Errol’s hearthstane, And the corbie roup [raven croak] in the falcon’s nest. The estate was sold by the Hay family in the 1630s to cover the debts of the 10th Earl (Jeremy Duncan, Local Studies Librarian, A. K. Bell Library, Perth, pers. comm., April 1997). UNKNOWN LOCATION Report of three plants of mistletoe on oak seen by a Mr Knowlton in August 1765 on the estate of White Esq. at Watling Wells (Dillwyn 1843). Knowlton worked at Londesborough Hall in the East Riding of Yorkshire (John Edmondson, pers. comm., October 1998), but it has not been possible to trace Watling Wells despite the assistance of the Yorkshire Museum, the English Place-Name Society and the Ordnance Survey Library. DISCUSSION The earliest known report of mistletoe on oak in Britain is the poem attributed to the 13th century Scottish poet, Thomas the Rhymer, describing the mistletoe-oak at Errol in Perthshire (Kanner 1939; see also Gurney 1848 and Melville 1935). There are three references to mistletoe on oak in the 17th century: the mistletoe-oak at Norwood (Surrey) felled in 1657 (Aubrey 1719); mistletoe on oak near Sheffield (Yorkshire) reported by Ray (1670, 1677); and the herbarium specimen of mistletoe from an oak from Stafford (Staffordshire) dated 1690. There was considerable interest in mistletoe and its host species from the 19th century until the early part of the 20th century and various accounts contain details of mistletoe-oaks in Britain (Lees 1842, 1851a; Bull 1864a, 1864b; Anon. 1873a; Webster 1885; Arnold 1887; Purchas & Ley 1889; Bull 1907; Elwes & Henry 1907; Somerville 1914; Tubuef 1923; Nicholson 1932; Durham 1935). Bull (1864a, 1864b) built on the initial observations of Lees (1842, 1851a) and gives details of six records/ authenticated reports of mistletoe-oaks. Footnotes are given in the later reprint of this paper (Bull 1907) which extended the list of records/authenticated reports to eleven trees. Later lists of mistletoe-oaks appear to be derived from earlier accounts, often combined with additional second-hand reports without any first-hand evidence. Nevertheless, the various lists compiled between the middle of the 19th century and the first part of the 20th century usually describe between ten and twenty mistletoe-oaks. 252 J/DY BOX Currently, eleven oaks with mistletoe growing on them have been located and verified in Britain (Table 1 and Figure 1D). The majority of the mistletoe plants are female and the mistletoe-oaks are concentrated in Herefordshire with single examples in Berkshire, Gloucestershire and Gwent. The records cover ten separate 10-km squares with the Brinsop and Stretton Sugwas oaks being in the same 10 km square. This figure is similar to the twelve 10 km squares for which mistletoe-oaks were reported in the 1969-1972 B.S.B.I. survey (Perring 1973). Written records go back at least a century for the existing mistletoe-oaks at Bredwardine, Deerfold Forest, Eastnor and Frampton-on-Severn. Of these, the mistletoe-oak at Eastnor has the longest recorded history being originally reported by Beaton (1837). Only four of the eleven muistletoe-oaks reported by Bull (1907) are still in existence: Bredwardine, Deerfold Forest, Eastnor (all in Herefordshire) and Frampton-on-Severn (Gloucestershire). Surprisingly, the very old mistletoe-oak at Stretton Sugwas (Herefordshire) was not known to Bull and was not reported in the literature until 1928. The other six existing mistletoe-oaks have not been recorded/reported in the published literature. These results, together with the lists of between ten and twenty mistletoe-oaks in the 19th and early 20th centuries, suggest a population of mistletoe-oaks which is apparently fairly stable in number, but in which there is a considerable turnover as losses are compensated by the discovery of new sites and the parasitism of new trees. Pedunculate oak, Q, robur, is the most frequent host amongst these existing mistletoe-oaks. The presence of mistletoe on red oak, Q. borealis, at three separate sites, as well as on a scarlet oak, Q. coccinea, is notable as the B.S.B.L survey recorded pedunculate oak as the only host species of oak (Perring 1973). It is interesting to note that Somerville (1914) includes a report of mistletoe on Q. rubra (syn. Q. borealis) in Worcestershire although no details are given. Mistletoe, however, is found on “red oaks” (Q. rubra, Q. coccinea, Q. palustris) in other parts of Europe, particularly in France (Timbal et al. 1994). Pedunculate oak is also the most frequent species of oak identified in past records/reports of mistletoe-oaks which now either no longer exist or where mistletoe is no longer present on the host tree. These records/reports identify pedunculate oak, Q. robur, at four sites [Tedstone Delamere (Herefordshire), Abermule (Montgomeryshire), Dunsfold (Surrey), and Bewdley (Worcestershire - although direct observation of leaves from this oak indicates that the tree has characteristics which are more those of a hybrid oak than of a pedunculate oak)]; sessile oak, Q. petraea, at one site (Llangattock Lingoed, Gwent); and hybrid oak, Q. robur x Q. petraea, at one site (Bristol, Somerset). The estimated ages of the existing oak hosts range from 90-400 years. The hosts include mature trees, but the hosts are not mostly very old, dying trees as the B.S.B.I. survey suggested (Perring 1973). Indeed, a photograph of the Stretton Sugwas oak (Herefordshire) taken in the 1930s shows a stag-headed tree very similar to the present tree. The existing mistletoe-oaks do not occur in woodland but prefer more open situations such as woodland edges, hedges, parks and even a churchyard. Timbal et al. (1994) report a similar finding with mistletoe on “red oaks” in France. Records and reports of other oaks with mistletoe confirm that the trees occur in open situations; locations in woodland tend to be on the edge of the woodland, for example “in the corner of a wood” (Tedstone Delamere, Herefordshire), or in “a fringe of ancient woodland” beside the Avon Gorge (Bristol, Somerset), or near a stream in the Wyre Forest (Bewdley, Worcestershire). It seems likely that the distribution of early (pre-1800) records and reports of mistletoe-oaks (Figure 1A) reflects the places visited by early botanists and the chance survival of records. There has been no significant change in thé distribution of mistletoe-oaks recorded and reported in the 19th century (Figure 1B) compared to that for the 20th century (Figure 1C). The pattern of distribution for all records and reports of mistletoe-oaks (Figure 1D) is similar to the distribution for all mistletoe records given in Perring (1973) and Briggs (1995). However, the range of existing confirmed mistletoe-oaks (Figure 1D, solid circles) is significantly more restricted than the historical range [Figure 1D, all circles], but coincides with the concentration of mistletoe records in Herefordshire and the surrounding counties (Perring 1973; Briggs 1995). In conclusion, the verification of only eleven mistletoe-oaks confirms the rarity of mistletoe growing on oak in Britain. Past records, reports and general accounts suggest that this rarity has not changed significantly since the 17th century when it was commented on by Ray (1670, 1677) MISTLETOE ON OAKS IN BRITAIN 253 FIGURE 1. A. Distribution of records and reports of mistletoe-oaks pre-1800. B. Distribution of records and reports of mistletoe-oaks in the 19th century. C. Distribution of records and reports of mistletoe-oaks in the 20th century. D. Distribution of records and reports of mistletoe-oaks (O) and existing mistletoe-oaks (@). 254 J.D. BOX and Evelyn (1664). Indeed, mistletoe from the oak at Eastnor was considered sufficiently noteworthy to be exhibited in 1837 at both the Linnean Society and the London Horticultural Society. Mistletoe (Viscum album) growing on native oaks (Q. robur and Q. petraea) is a rare association throughout western Europe (various sources are cited by Tubuef 1923 and Grazi & Urech 1983), although there are at least 150 sites for mistletoe on “red oaks” in France (Timbal et al. 1994). The present range of verified mistletoe-oaks is, however, less than the historical range based on past records and reports. This may be due to the clearance of woods and hedges resulting from the intensification of agriculture and the extension of urban areas over the past 50 years. Within the present range, the existing mistletoe-oaks are concentrated in Herefordshire which coincides with the core of the present distribution of mistletoe in Britain (Perring 1973; Briggs 1995). ACKNOWLEDGMENTS The assistance of B.S.B.I. recorders throughout Britain is very gratefully acknowledged as is the help given by the staff of local Biological Record Centres in England and Wales and the staff of various herbaria and libraries. Particular assistance was given by Gill Gent (B.S.B.I. recorder for Northamptonshire), Stephanie Thomson (B.S.B.I. recorder for Herefordshire), Paul Hand (Shropshire), Sue Liptrot and her colleagues (Shropshire Libraries), Gina Douglas (librarian and archivist at the Linnean Society of London) and Glenys Knowles and Mary Sampson (library of the Royal Society). The owners and occupiers of the sites of mistletoe-oaks, both past and present, have been particularly helpful both in correspondence and through arranging site visits in winter. Jane Croft and Henry Arnold at the Biological Records Centre of the Institute of Terrestrial Ecology at Monks Wood have given much help by computerising the records and reports and generating the distribution maps using Alan Morton’s DMAP programme. Lastly, I must acknowledge the assistance given by Franklyn Perring and Jonathan Briggs whose involvement in surveys of mistletoe has added significantly to our knowledge of this species. REFERENCES ANON. (1837). The Gardener’s magazine 13: 333-334. ANON. (1866a). The remarkable trees of Herefordshire: the mistletoe-oak at Eastnor. Transactions of the Woolhope Naturalists’ Field Club, illustration facing p. 149, published 1867. ANON. (1866b). The remarkable trees of Herefordshire: the mistletoe-oak at Tedstone Delamere. Transactions of the Woolhope Naturalist’s Field Club, illustration facing p. 165, published 1867. ANON. (1867b). Transactions of the Woolhope Naturalists’ Field Club, p. 11, published 1868. ANON. (1870). The remarkable trees of Herefordshire: the mistletoe-oak of Bredwardine. Transactions of the Woolhope Naturalists’ Field Club 1870, illustration facing p. 288 & pp. 317-318, published 1871 [in an article ‘Incidental Notes on Remarkable Trees in Herefordshire’ by a Commissioner from the Woolhope Club, pp. 288-321]. ANON. (1871). Crown East, Cotheridge, Broadwas, Doddenham Dingle, Devil’s Leap, Nipple Well, Ankerdine and Knightsford Bridge: Thursday, 31st August, 1871. Transactions of the Worcestershire Naturalists’ Club 1847-1896, pp. 175-179. ANON. (1872). Shrawley Wood: Tuesday, 22nd October, 1872. Transactions of the Worcestershire Naturalists’ Club, 1847-1896, pp. 192-195. ANON. (1873a). Mistletoe. The Leisure hour, No. 1101, 1 February 1873, pp. 71-72. ANON. (1873b). Mistletoe oaks. The Leisure hour, No. 1110, 5 April 1873, p. 224. ANON. (1878). Bromyard and Avenbury: Friday, 31st May, 1878. Transactions of the Worcestershire Naturalists’ Club, 1847-1896, pp. 246-253. ANON. (1884). Kempsey, Croome, Besford, Pershore and Stoulton: Friday, 13th June, 1884. Transactions of the Worcestershire Naturalists’ Club 1847-1896, pp. 313-317. ANON. (1928). Spring annual meeting: Thursday, April 19th, 1928. Transactions of the Woolhope Naturalists Field Club 1928, pp. xliv—lvi, published 1931. ANON. (1930). Second field meeting, Tuesday, June 24th, 1930: Deerfold Forest and Wigmore. Transactions of the Woolhope Naturalists Field Club 1930-1932, pp. xvii—xx, published 1933. ANON. (1933). First field meeting, Thursday, June 24th, 1930: Monnington and Moccas. Transactions of the Woolhope Naturalists Field Club 1933-1935, pp. xi—xvi, issued 1936. MISTLETOE ON OAKS IN BRITAIN 295 ARNOLD, F. H. (1887a). Notes on the mistletoe. The Naturalists’ monthly 1(4): 65-67. ARNOLD, F. H. (1887b). Flora of Sussex. Hamilton, Adams & Co., Chichester. AUBREY, J. (1719). Natural history and antiquities of Surrey. E. Curll, London. BEATON, D. (1837). Instances of the mistletoe being found on the oak; with remarks on grafting and budding the mistletoe. The Gardener’s magazine 13: 206-208. Box, J. (1995). The festive ecology of holly, ivy and mistletoe. British wildlife 7: 69-74. BRIGGS, J. (1995). Mistletoe - distribution, biology and the National Survey. British wildlife 7: 75-82. BRITTEN, J. (1870). Mistletoe on the oak. Journal of botany 8: 87. BULL, H. G. (1864a). Mistletoe in Herefordshire. Transactions of the Woolhope Naturalists Field Club 5: 59- 108. [This was reissued in 1907 with footnotes and a postscript as part of a bound volume covering the years 1852-—1865.] BULL, H. G. (1864b). The mistletoe (Viscum album, L.) in Herefordshire. Journal of botany 2: 361-385. [Note on paper that it is abridged from the Transactions of the Woolhope Naturalists’ Field Club with corrections by the author; the text appears to be almost the same as Bull (1864a)]. BULL, H. G. (1869). Remarkable plants in Deerfold Forest. Transactions of the Woolhope Naturalists Field Club 1869, illustration facing p. 15 & pp. 15—16, 1869, published 1870. BULL, H. G. (1870). The mistletoe-oak at Llangattock Lingoed. Transactions of the Woolhope Naturalists Field Club 1870, pp. 68-69 & illustration facing p. 68, published 1871. BULL, H. G. (1907). The mistletoe in Herefordshire. Transactions of the Woolhope Naturalists Field Club, 1852-1865, pp. 312-347. [A reprint of the early volumes of the Transactions with footnotes and postscript added to the paper originally published in 1864]. CORNEWALL, G. (1903). Early annual meeting, Tuesday, March 31st, 1903: Presidential Address. Transactions of the Woolhope Field Naturalists’ Club, 1902-1904, pp. 97-106, published 1905. DALLMAN, A. A. (1939). Recorder’s report on botany. Record of bare facts for the year 1939, No. 49, pp. 4— 13, issued May 1940. Caradoc & Severn Valley Field Club, Shrewsbury. DESMOND, R. (1977). Dictionary of British & Irish botanists and horticulturists. Taylor & Francis Ltd and The Natural History Museum, London. DILLWYN, L. (1843). Hortus Collisonianus: an account of the plants cultivated by the late Peter Collinson Esq. FRS. Privately published, Swansea. DURHAM, H. E. (1935). On mistletoe. Transactions of the Woolhope Naturalists’ Field Club 1933-1935, pp. 140-153, issued 1938. ELWES, H. J. & HENRY, A. (1907). The trees of Great Britain and Ireland 2: 333-334. Privately published. EVANS, A. H. (1939). A Flora of Cambridgeshire. Gurney & Jackson, London. EVANS, T. G. (1995). The Mistletoe Survey 1994—96 in Gwent. BSBI News 70: 27. EVELYN, J. (1664). Sylva, or a discourse of forest-trees, 1st ed. London. FRAZER, J. G. (1922). The Golden Bough. Macmillan Press, London. GISSING, T. W. (1855). A letter to the editor, ‘Botanical Notes, Notices, and Queries’ section, The Phytologist, New Series, 1: 191-192. GoopD, R. (1984). A concise Flora of Dorset. The Dorset Natural History and Archaeological Society, Dorchester. GRAZI, G. & URECH, K. (1983). La susceptibilité des chénes, des ormes et des mélézes au gui (Viscum album L.). Revue Scientifique du Bourbonnais pp. 6-12. GURNEY, D. (1848). The record of the House of Gournay, Part II. Privately printed by John Bowyer Nichols and John Gough Nichols, London. HADFIELD, M. (1974) The oak and its legends, in MORRIS, M. G. & PERRING, F. H., eds. The British Oak: its history and natural history. E. W. Classey (Faringdon, Berkshire) for the Botanical Society of the British Isles. HAMILTON, G. J. (1975). Forest mensuration. Forestry Commission Booklet No. 39. Her Majesties Stationery Office, London. HOLLAND, S. C., ed. (1986). Supplement to the Flora of Gloucestershire. Grenfell Publications, Bristol. JERMYN, S. T. (1974). Flora of Essex. Essex Naturalists’ Trust, Colchester. JESSE, E. (1844). Scenes and tales of country life; with recollections of natural history. John Murray, London. KANNER, L. (1939). Mistletoe, magic and medecine. Bulletin of the history of medicine 7: 875-936. LEES, E. (1842). The Botanical Looker-out among the wild flowers of the fields, woods, and mountains, of England and Wales, pp. 6—23. Tilt and Bogue, London and H. Davies, Cheltenham. LEES, E. (1851a). The Botanical Looker-out among the wild flowers of England and Wales, 2nd ed. Hamilton, Adams & Co, London. LEES, E. (1851b). New localities for mistletoe on the oak; with some remarks in reference to a paper on the mistletoe in “The Naturalist’ for September, by Mr McIntosh. The phytologist 4: 357-360. LOUDON, J. C. (1838). Arboretum et Fruticetum Britannicum 3: 1831. Longman, Orme, Brown, Green & Longman, London. LOUSLEY, J. E. (1976). Flora of Surrey. David & Charles, Newton Abbot. 256 J.D. BOX MAXWELL, H. (1900). Memories of the Months. Second series, pp. 284-285. Edward Arnold, London. MAY, R. F. (1967) A list of flowering plants and ferns of Carmarthenshire. West Wales Naturalists Trust, Havorfordwest. MELVILLE, L. (1935). Errol: its legends, lands and people. Hunter, Perth; reprinted in facsimile, 1985. Morris, D. (1917). The mistletoe. Its life history and associations with primitive religion, folk lore and supersitions. The Bournemouth Guardian, 22 December 1917. NICHOLSON, C. (1932). The mistleto and its hosts. The Gardeners’ Chronicle, pp. 102-104, 145-146. ORDONYO, T. (1807). Flora Nottinghamiensis. B. Crosby & Co., London. PERRING, F. (1973). Mistletoe, in: GREEN, P. S., ed. Plants wild and cultivated, pp. 139-145. Botanical Society of the British Isles. PRYCE, R. (1996). Monitoring mistletoe in Carmarthenshire. Llanelli Naturalists newsletter No. 60: 9-10. PULTENEY, R. (1795). A catalogue of some of the more rare plants found in the neighbourhood of Leicester, Loughborough and in Charley Forest, in NICHOLS, J., ed. The History and Antiquities of the County of Leicester , vol. 1, part 1, pp. clxxvii — cxc. John Nichols, London. PURCHAS, W. H. & LEY, A., eds. (1889). A Flora of Herefordshire. Jakeman & Carver, Hereford. RAY, J. (1670). Catalogus Plantarum Angliae, et Insularum Adjacentium, 1st ed. p. 319. John Martyn, London. RAY, J. (1677). Catalogus Plantarum Angliae, et Insularum Adjacentium, 2nd ed. p. 307. John Martyn, London. RAYNER, J. F. (1929). A supplement to the Flora of Hampshire. Privately published by the author. REA, C. (1903). Discursive notes and observations for the seasons 1901 and 1902. Transactions of the Worcestershire Naturalists’ Club, Vol. 3, 1899-1906, pp. 121-135. REA, C. (1910). Additions to the botany of Worcestershire. Transactions of the Worcestershire Naturalists’ Club, Vol. 4, 1907-1910, pp. 270-293. RIDDELSDELL, H. J., HEDLEY, G. W. & PRICE, W. R. (1948). Flora of Gloucestershire. The Cotteswold Naturalists’ Field Club, Cheltenham. ROE, R. G. B. (1981). The Flora of Somerset. Somerset Archaeological and Natural History Society, Taunton. ROPER, I. M. (1916). Mistletoe on the oak in Somerset. Journal of botany 54: 88. RopER, I. M. (1918). Presidential Address: Mistletoe. Annual report and proceedings of the Bristol Naturalists’ Society, Fourth Series, 4(3): 175-185, issued for 1915-16. SALMON, C. E. (1931). Flora of Surrey. G. Bell & Sons, London. SALTER, J. H. (1928). Mistletoe on oak. North Western Naturalist 3: 39. SALTER, J. H. (1935). The flowering plants and ferns of Cardiganshire. University Press Board, Cardiff. SHOOLBRED, W. A. (1920). The Flora of Chepstow. Taylor & Francis, London. SINKER, C. A. et al. (1985). The ecological Flora of the Shropshire region. Shropshire Trust for Nature Conservation, Shrewsbury. SOMERVILLE, W. (1914). The mistletoe in England. Quarterly journal of forestry 8: 20-25. SWETE, E. H. (1854). Flora Bristoliensis. J. van Voorst, London. TIMBAL, J., KREMER, A., LE GOFF, N. & NEPEVEAU, G. (1994). Le Chéne Rouge d’Amérique. Institut National de la Recherche Agronomique, Paris. TONKIN, M. (1984). The Wigmore Inclosure Act and Award, 1810-1828. Transactions of the Woolhope Naturalists Field Club 54: 283-300. TOWNSEND, F. (1883). Flora of Hampshire, 1st ed. L. Reeve & Co., London. TOWNSEND, F. (1904). Flora of Hampshire, 2nd ed. L. Reeve and Co. Ltd., London. TRIMMER, K. (1866). Flora of Norfolk. Hamilton, Adams & Co, London. TRIMMER, K. (1885). Supplement to the Flora of Norfolk. Jarrold & Sons, London. TRUEMAN, I. C., MORTON, A. & WAINWRIGHT, M. (1995). The Flora of Montgomeryshire. Montgomery Field Society/Montgomery Wildlife Trust, Welshpool. TUBUEF, K. VON (1923). Monographie der Mistel. Verlag Oldenbourg, Miinchen u. Berlin. WARNER, R. (1771). Plantae Woodfordiensis: a catalogue of the more perfect plants growing spontaneously about Woodford in the county of Essex . Privately printed. WEBSTER, A. D. (1885). Parasitical flowering plants. Hardwicke’s Science-Gossip 21: 172-175 [No 248, August 1885]. WHITE, J. (1994). Estimating the age of large trees in Britain. Research Information Note 250. The Forestry Authority, Edinburgh. WHITE, J. W. (1912). The Flora of Bristol. John Wright & Sons Ltd, Bristol. WITHERING, W. (1796). An arrangement of British Plants, 3rd ed. M. Swinney, Birmingham. WOLLEY-Dop, A. H. (1937). Flora of Sussex. Kenneth Saville, Hastings. (Accepted January 2000) Watsonia 23: 257-267 (2000) 257. A morphological comparison between some British Orobanche species (Orobanchaceae) and their closely-related non-British counterparts from continental Europe: Orobanche reticulata Wallr. s. L. Ma YorFOEEY Department of Biological Sciences, Institute of Environmental and Natural Sciences, University of Lancaster, Lancaster, LA] 4YQ ABSTRACT A morphological comparison has been made between lowland British plants and those of the European mountains, both referable to O. reticulata s.l.. A brief summary is given of the previous taxonomic treatments which have been applied to this group. It is concluded that two separate but closely-related taxa exist, and that these are best treated at subspecific level, the British plant being referred to subsp. procera (Koch) Dostal. A possible third taxon, perhaps no more than a pale colour form of one or both of the above, has been described at various ranks under the epithet ‘pallidiflora’. No direct support for its continued recognition has been established. KEYWORDS: Taxonomy, root parasites, broomrape. INTRODUCTION BACKGROUND Orobanche is a genus of obligate root parasites, which completely lack chlorophyll and are therefore unable to photosynthesise and so independently sustain themselves. To do so they parasitise a wide range of (usually) dicotyledonous, and very often herbaceous, hosts. The genus is distributed throughout much of the world’s northern temperate latitudes and has a northern limit in southern Scandinavia. Only relatively few species occur in the southern hemisphere and some of these are probably introductions. In comparison to continental Europe, the British Isles have a very limited Orobanche flora, only ten species being recognised by Stace (1991). This is in contrast to mainland Europe where even the conservative treatment in Flora Europaea (Chater & Webb 1972) describes 45 at species level. Apart from the relatively widespread O. minor Sm., which in fact may have been largely introduced into this country, all the other British species are apparently native and several are very rare. Four of the latter have been accorded British Red Data Book status (Wigginton 1999): O. artemisiae-campestris Gaudin (= O. loricata Rchb.), O. caryophyllacea Sm., O. purpurea Jacq., and O. reticulata Wallr. In Europe, however, these and many others are much more frequent, and the abundance and species diversity increases appreciably as one moves south, with the Mediterranean area having an especially rich Orobanche flora. Many species show close morphological similarities one to another and can therefore be difficult to separate taxonomically. In this paper, the relationship is examined between British plants referable to O. reticulata Wallr. and those from continental Europe. In doing so, much reliance has been placed on the morphometric assessment of living populations of plants, with additional information obtained from the study of preserved herbarium material. PROBLEMS ENCOUNTERED IN THE STUDY OF THE GENUS This is acknowledged to be a taxonomically difficult genus (e.g. Chater & Webb 1972, Pignatti 1982, Rumsey & Jury 1991) and meaningful work, whether in the field or in the herbarium, is difficult to achieve. For instance, these morphologically very similar taxa possess few readily 258 Nia YOROLEY, quantifiable characters. The above-ground plant comprises little more than an inflorescence, the stem of which bears simple, sessile, scale-like leaves which rapidly become brittle soon after emergence. This lack of normal leaves is therefore a major loss of potential taxonomic information and the inflorescence is the only part of the living plant which it is feasible to study. Being fleshy and lacking chlorophyll, Orobanche plants make poor herbarium specimens, sometimes forming a rather amorphous mass so that important characters are lost. This is especially true of corolla and stigma colours and subtleties of corolla shape. However, from well- pressed specimens with accompanying field colour notes, useful information can be gained. Unfortunately, a large proportion of specimens are not preserved to such a high standard, and so it is better to study the genus as living plants. There are also other difficulties. Many of the original authors’ descriptions are very brief and inadequate or even ambiguous. In such cases, recognition of the taxon to which the description is applied may be almost impossible to achieve. Very often descriptions refer to non-diagnostic characters, such as leaf and stem, whilst information on much more important characters is generalised or even omitted. Type specimens are rarely indicated by these authors and when they are known, they may be very old and by now seriously fragmented, and so taxonomically almost worthless. OROBANCHE RETICULATA s. |. O. reticulata s. I. is widespread but local throughout much of the montane/subalpine region of Europe and is also recorded from lowland habitats elsewhere on the continent. However, it was not until shortly after the start of the twentieth century that plants apparently corresponding with the description of continental O. reticulata were recognised in Britain. These occurred (and still do so) in a very limited area of Yorkshire, as lowland plants, usually on the magnesian limestone where they mainly parasitise Cirsium arvense, growing on rough calcareous grassland and scrub, and especially on river margins and flood plains. Since there must be many similarly suitable habitats throughout the rest of the country, this very restricted British range is difficult to explain. The discovery in Britain in 1907 is of some interest and has been summarised elsewhere (Foley, 1993). I have examined both the British plant and those of the European mountains on several occasions in the field and certain morphological differences between the two are apparent which suggest the existence of separate taxa. BASIC TAXA RELATED TO O. RETICULATA AND SUBSEQUENT TREATMENTS O. reticulata was first described by Wallroth (1825) from plants collected near Toulouse, France. His description is quite vague and does not especially differentiate it from other similar taxa. He described the calyx as divided into simple, linear-lanceolate teeth, shorter than the tubular, inflated, 20 mm long corolla. The lobes of the lower lip were rounded and equal in size and the filaments and style were more or less glabrous. His most significant statement referred to the purple coloration of the corolla caused by the reticulate veining of the upper lip, and this appears to be a consistent character of O. reticulata s. l. as seen in the field. The type specimen (Toulouse, 1807, Fliigge) is held in LE(!) and, although comprising only the upper part of the inflorescence, equates satisfactorily with living plants seen in mountain areas of southern France, but much less so with those from Britain. Shortly after Wallroth’s description of O. reticulata, Wimmer & Grabowski (1829) described O. pallidiflora from a specimen collected in Poland. This plant was pubescent with flowers in a lax spike. The calyx segments were described as ovate-acuminate and the corolla as curved, tubular- ventricose, with glands and purple veins distally. The corolla lobes were crenulate margined and the filaments pilose at the base and + glabrous above. There is very little information here to separate it from Wallroth’s O. reticulata. Type material of O. pallidiflora has not been located but specimens in several herbaria are referred to this species. Another worker who described similar plants from Germany was Koch. His description of O. procera (Koch 1833) stated that the calyx segments were uniformly divided or entire, the corolla yellow-brown, campanulate, inflated towards the base and curved, and the lobes of the lips denticulate and ciliate. The filaments, inserted at the base of the corolla, were sparsely hairy below with a few glandular hairs above. In Koch’s opinion the plant lay taxonomically close to OROBANCHE RETICULATA 259 O. rapum-genistae but the corolla and filament characters given clearly differentiate the two. The plants upon which he provided his description apparently grew close to the Rhine and, from his comments, it is clear that Koch knew the living plant from this same area. At the same time as he published this description, he also described O. scabiosae (Koch 1833) based on plants from a mountain area near Berchtesgaden, Germany, growing at an altitude of 5500 feet and apparently parasitising Scabiosa columbaria. His diagnosis described the calyx segments as entire or divided into teeth of equal length; the corolla campanulate, slightly curved, glandular-hairy with the margins of the lobes wrinkled, the filaments inserted low in the corolla with scattered hairs in their lower part but almost glabrous above. He made interesting comments on the corollas being basically bright ochre-yellow but tinged reddish-orange on the lip margins, violet in the dorsal area and also violet-veined on the lips. Most subsequent workers have treated O. scabiosae as a synonym of O. reticulata Wallr.. This seems justified in view of Koch’s description, especially of the corolla colour and its montane habitat. It is clear, however, that Koch considered it to be quite separate from his O. procera which is described almost immediately before it in the same paper (Koch 1833). From this it might be assumed that O. reticulata and O. procera are themselves morphologically quite separate. No type specimens of either of the two species described by Koch have been located and none of either is indicated as having been seen by Beck (1930). Beck (1890) recognised three varieties within O. reticulata: the type (var. typica), and vars. pallidiflora and procera. Later, he largely adhered to this treatment (Beck 1930), retaining the three taxa at varietal level and providing a key to their identification where var. procera was separated from the others chiefly by the shape of its broad, tubular-campanulate, completely curved corollas, 15—18 mm in length. In the other varieties he considered the corollas to be tubular or campanulate, longer than wide, with the dorsal line straight or only slightly curved, 20-23 mm long or occasionally shorter. He separated var. pallidiflora from the type mainly by its paler corolla, especially the degree of venation and presence of violet glandular hairs, and also in its slightly more hairy filaments and choice of host (Cirsium and Carduus spp., rather than members of the Dipsacaceae or other members of the Asteraceae). In their French Flora, Rouy & Foucaud (1909) placed these taxa under O. platystigma Rchb. based on a specimen of Reichenbach f. The first, var. scabiosae (based on Koch’s O. scabiosae) had a tubular corolla 20-22 mm long, yellow at the base, then amethyst or purple distally and covered with coloured glandular hairs. Another variety, var. procera, had a much broader, subcampanulate, shorter corolla (16-18 mm long), weakly veined violet or purple but only on the upper lip and generally with fewer, less strongly coloured glandular hairs. Gilli (1965), working on central European taxa, recognised two subspecies of O. reticulata: the type and subsp. pallidiflora. He described subsp. pallidiflora as having corollas which were whitish or yellowish but pale lilac distally with few dark-coloured glandular hairs. In their upper part, the filaments were sparsely glandular hairy to glabrous. He considered this taxon to occur only in low-lying areas in Germany and Austria north of the Alps, where it parasitised species of Cirsium and Carduus. Under subsp. reticulata he described the corolla as usually longer than broad (20-23 mm long) and the dorsal line straight or only slightly curved. Within this he included var. procera (Koch) Beck, characterised by a quite different corolla shape: broader, tubular- campanulate and 15—18 mm long with the dorsal line completely curved. Gilli described the type (subsp. reticulata) as differing from subsp. pallidiflora mainly in corolla colour, which was yellow only at the base and otherwise intensely coloured violet to purple with dark veins and glandular hairs. In Flora Europaea, Chater & Webb (1972) described only one species, O. reticulata Wall., placing O. pallidiflora under synonymy; there was no mention of O. procera. However Pignatti (1982) recognised two taxa in Italy: subsp. reticulata from the alpine area and subsp. pallidiflora from lower altitudes. He separated the latter from the typical form by its paler corolla colour and less glandular-pilose filaments. This was largely supported by Uhlich et al. (1995) when considering European taxa as a whole but Hartvig (1991), who distinguished the two taxa in a somewhat similar way, considered that for the montane areas of Greece most plants belong to var. pallidiflora rather than var. reticulata. More recently, Kreutz (1995) recognised two taxa at species level in Europe: O. reticulata from the alpine areas and O. pallidiflora from lower altitudes. He placed the British plant under the latter whilst including var. procera within O. reticulata. 260 Me J. Y. FOLEY From the varied treatments outlined above it is apparent that much confusion exists regarding the taxonomy of O. reticulata s.l., but that it includes at least two taxa which differ mainly in their corolla shapes. These are the montane/subalpine O. reticulata and the lower-altitude O. procera. Plants with a paler corolla often appear to be referred to O. pallidiflora. METHODS A fundamental approach in this work has been to carry out detailed field examinations of the morphology of living plants in their natural habitats. From personal field experience on the genus and also from that of other workers (e.g. Beck 1890, 1930; Guimaraes 1904; Wilkomm & Lange 1870; Wilkomm 1893; Coutinho 1913; Schiman-Czeika 1964; Gilli 1965, 1982; Chater & Webb 1972; Feinbrun-Dothan 1978; Pignatti 1982; Meikle 1985; Valdés et al. 1987; Jones 1989; Rumsey & Jury 1991), the characters described below and listed in Table 1 appear to be the most useful in circumscribing closely related Orobanche taxa. Many taxa are best described by using several characters in combination. The methods used were as follows. A random selection was made of ten plants within a population, all of which were in full flower. By “full flower” was meant plants in which the lowest third of the inflorescence had fully open corollas, one of which was utilised for the morphological evaluation of floral characters. For this purpose, only freshly-opened corollas were used. This approach is similar to that previously employed for the assessment of taxa within other genera (e.g. Roberts & Gilbert 1963; Roberts 1966, 1988; Bateman & Denholm 1983; Foley 1990) and has also been recently applied to populations of British Orobanche taxa by Jones (1989). With practice, selection bias can be avoided and a representative sample of plants assessed. The raw data obtained from each of the populations studied are given in Foley (1998). Measurements were achieved in two ways. Dividers used in conjunction with a ruler calibrated to 0-5 mm were found convenient and very satisfactory for dimensions upwards of 20 mm. For smaller values, a Leitz Wetzlar measuring lens possessing a built-in scale calibrated to 0-1 mm was used. For the very largest dimensions such as plant height and inflorescence length it was often not possible to measure reliably to less than 5 mm due to slight vagaries in stem contortions etc.. All values were taken in a direct line and therefore lengths of curvature (for instance of corollas) were ignored in such measurements. Quantifiable numerical data obtained from the morphometric measurements of many of the characters shown in Table | were also subjected to Principal Components Analysis (PCA) using the Clustan 4 computer programme (Wishart 1987). By this technique, each specimen measured is located in a multidimensional array, in which the number of dimensions is equal to the number of characters measured, and where the most similar specimens are placed closest together. This technique extracts the axes of greatest variation from the multidimensional space, thereby simplifying the space to a few (usually 2 or 3) dimensions, so allowing the location of each specimen to be visualised. Qualitative characters (34-45) and those of a binary nature could not be used in PCA. Several of these are known to be important diagnostic characters for Orobanche taxa and, of course, were fully considered alongside the PCA data and in the subsequent description and delimitation of the individual taxa. Of especial significance are corolla colour and shape, calyx shape and stigma lobe colour. Binary characters which clearly separate sections or groups within the genus (e.g. presence or absence of stem branching (character 7) or presence or absence of bracteoles (character 22), were not analysed further. The following characters (enumerated in Table 1) were employed in the PCA: 1-6, 8-15, 18-20, 23-29b, 33. In the diagrams, the component accounting for the greatest variance (PC1) is represented by the x-axis, that of next significance (PC2) by the y-axis. Photographic records were also made of plants within each population studied; many of these are illustrated in Foley (1998). Whenever conservation considerations allowed, a typical specimen was pressed for the herbarium (E, LANC) and inflorescences preserved in 70% alcohol (now in LANC). Most major British herbaria likely to contain material of the genus were visited during this work: and a great number of loans were also obtained from European institutions. The internal flower structures of preserved specimens were examined by detaching individual corollas and gently boiling them for c. 10 minutes in water to which a small amount (<0-01%) of surfactant had been —— OROBANCHE RETICULATA 261 TABLE 1. CHARACTERS AND SCORING SYSTEM UTILISED IN THE MORPHOMETRIC ANALYSIS (ALL MEASUREMENTS IN MM) 1 PLANT HEIGHT: measured from ground level to the uppermost tip of the inflorescence. 2 STEM WIDTH: measured immediately below the lowest corolla in the inflorescence. 3 STEM PUBESCENCE: estimated on the scale shown below *. 4 INFLORESCENCE LENGTH: measured from the point of attachment of the lowest corolla to the uppermost tip of the inflorescence. 5 INFLORESCENCE DENSITY: estimated on the scale:- 0 = lax; 1 = fairly lax; 2 = fairly dense; 3 = very dense. 6 COROLLA ATTITUDE: estimated on the scale:- 0 = almost erect to 3 = almost level, patent. 7 STEM HABIT: 0 = simple, unbranched; | = pedicellate at least below; 2 = branched. 8 © MAXIMUM LEAF WIDTH: as measured. 9 LEAF PUBESCENCE: estimated on the scale shown below *. 10 COROLLA LENGTH: measured in a direct line from base of the corolla to the extremity of the upper lip. 11 COROLLA VERTICAL HEIGHT: measured distally at a point just prior to reflexing of the lips. 12 DORSAL LINE CURVATURE: estimated on the scale:- 1 = almost straight; 2 = slightly curved; 3 = strongly curved; 4 = genuflexed. 13. DIVISION OF UPPER LIP: estimated on the scale:- 0 = entire; 1 = notched; 2 = slightly bilobed; 3 = strongly bilobed. 14. LOWER LIP CILIATE OR OTHERWISE: 0 = not; 1 = yes. i LOWER LIP DENTICULATE OR OTHERWISE: 0 = not; 1 = yes. 16 & ATTITUDE OF UPPER AND LOWER LIPS OF COROLLA: estimated on the scale:- -1 = reflexed; | 0 = upright; 1 = intermediately forwards; 2 = forward. 18 RELATIVE SIZE OF THE CENTRAL LOBE OF THE LOWER LIP OF COROLLA IN RELATION TO THE LATERAL LOBES: ratio estimated. 19 SHAPE OF THE LOBES OF THE LOWER LIP OF COROLLA: estimated on the scale:- 0 = rounded, distinctly obtuse; 1 = more or less acute. 20 COROLLA PUBESCENCE: estimated on the scale shown below *. 21 COROLLA CONSTRICTED DISTALLY: estimated as:- 0 = not; 1 = yes. 22 BRACTEOLE: 0 = absent; | = present. 23. BRACT LENGTH: measured from the point of attachment to the tip. 24 BRACT PUBESCENCE: estimated on the scale shown below *. 25 CALYX - LENGTH OF SEGMENTS; if divided, includes the longest tooth. 26 CALYX - SEGMENTS DIVIDED INTO SEPARATE TEETH OR OTHERWISE: Degree of division estimated on the scale:- 1 = equally divided; 2 = unequally divided; 3 = very unequally divided; 4 = entire. 27 CALYX PUBESCENCE: estimated on the scale shown below *. 28 TYPICAL HEIGHT OF INSERTION OF THE FILAMENTS ABOVE THE COROLLA BASE: measured from the point of attachment of the lowest 29a FILAMENT PILOSITY (UPPER PORTION): estimated on the scale shown below *. 29b FILAMENT PILOSITY (LOWER PORTION): estimated on the scale shown below *. 30 ANTHERS OR STYLES EXSERTED: estimated on the scale shown below *. 31 ANTHERS PILOSE: estimated on the scale shown below *. 32 STYLE PILOSITY: estimated on the scale shown below. * 33. STIGMA LOBE SEPARATION: estimated on the scale:- 0 = fused to 3 = well separated. GENERAL SCALE: 0 = not, absent; 1 = slightly; 2 = appreciably; 3 = very, densely, excessively. (Values intermediate between those given above may, in some cases, be arrived at). NOTE: Additional characters (e.g. leaf length, bract width, maximum width of corolla tube, maximum distance between the extremities of the lateral lobes of the lower lip of the corolla) were also measured during the early stages of this work. However, some of these were later abandoned when the character states were found to yield information of little value. The following additional characters were recorded descriptively or photographically since they are not readily quantifiable. They were recorded from fresh material in full flower. 34. STYLE COLOUR, 35. STIGMA LOBE COLOUR, 36. COROLLA SHAPE, 37. COROLLA COLOUR, 38. LEAF SHAPE, 39. LEAF COLOUR 40. BRACT SHAPE, 41. BRACT COLOUR, 42. CALYX SHAPE, 43. CALYX COLOUR, 44. FILAMENT BASE SHAPE, 45. STEM COLOUR. 262 M. J. Y. FOLEY added. Formerly dry, brittle material could then be softened and more easily handled and dissected. It was therefore possible to measure the height of filament insertion and the degree, type, and distribution of pilosity along their length. Also, if the stigma and style were first softened by a similar process and carefully pressed with clean filter paper, an indication of stigma lobe colour could sometimes be obtained. RESULTS AND DISCUSSION A COMPARISON BETWEEN BRITISH POPULATIONS AND PLANTS FROM MONTANE/SUBALPINE HABITATS IN EUROPE A morphological evaluation was carried out on plants referred to O. reticulata from three British populations within the plant’s very restricted British range in Yorkshire (see Table 2 for locality details). These populations were compared one to another as well as to plants from montane localities in eastern France. In the latter area, however, although several populations were located and examined, only one population was found (Petit Mont Rond, Jura) which had plants in sufficiently large numbers and in a suitable condition for a morphological evaluation to be made. This was despite searches on several field visits over three seasons (1995-1997) and was mainly due to their unpredictable flowering behaviour at higher altitudes. Mean values and standard errors of the relevant characters are summarised in Table 2 and the quantifiable numerical elements of these data have been subjected to Principal Components Analysis (PCA). Examination of PCA data (Figure |) for the three British populations shows good overlap on the PC1 axis but with some separation for the Aberford population on the PC2. This appears to be mainly due to their less robust habit (reflected in reduced height, inflorescence length, stem width, and leaf, bract and calyx dimensions (Table 2)). On the other hand there is very clear separation on the PC] axis between the British and the French plants and an examination of 4, pPc2 |S + Oo a4 2 4 e|° ® one | @ © O Sie. ee ao J 0. @ ‘ea® PCI e @ ° os Se 2 o. S + O 2 O Ace! 5 3 7 1 3 5 45; Aberford (Yorkshire) © Ripon (Yorkshire) © Oxclose (Yorkshire) @® Petit Mond Rond, Jura (France) FIGURE |. First two axes of Principal Components Analysis for individual plants from populations referred to Orobanche reticulata from Yorkshire, England and France. PC] & PC2 jointly account for 43-6% of the total variance. OROBANCHE RETICULATA 263 TABLE 2. COMPARISON OF MORPHOLOGICAL CHARACTERS OF POPULATIONS OF BRITISH AND ALPINE O. RETICULATA S.L. Locality Number Date Grid Reference/Coordinates Plant height Stem Inflorescence Leaf Bract Corolla Upper lip Lower lip width pubescence colour length density width pubescence shape colour length shape pubescence colour length height approximate shape attitude dorsal line pubescence colour divided ciliate denticulate shape of lobes relative size (central to lateral) Bracteole Calyx Filaments Anthers Stigma lobes Apparent host length segment division pubescence shape of teeth colour insertion height pilosity above pilosity below pilose separation colour BRITAIN Aberford Yorkshire 10 2.vii.1994 SE/43.35 407 + 16-5 8-0 + 0-4 2:5+0 fawn-violet 130-3 + 10-1 2:1+0-1 4:5+0-1 1-540 linear-lanceolate cream 17:9+0-8 linear-lanceolate 2:1+0-1 cream-purple 20-4 + 0-4 6-8 + 0-3 + campanulate 1:5+0 3:0 + 0 1:6+0-1 cream-purple 1:7+0-1 0 0 0 1-:1+0 0 11-4+0-4 2:5 + 0-3 2-0 +0 tapering mauve-purple 3:-2+0-1 0-1+0 0-2 + 0-1 0 1:7+0-1 mauve-purple Cirsium arvense Mean values and standard errors (in mm) BRITAIN Oxclose Yorkshire 10 9.vii. 1994 SE/36.46 529 + 33-2 9-2+0-5 2:0+0 cream-mauve 181 + 17-7 1-9+0-1 5-2+0-3 1-6+0 linear-lanceolate mauve-cream 19-6 + 0:8 linear-lanceolate 2:0 + 0 cream 21-7+0:-5 8-1+03 + campanulate 1-5 +0 cream-purple 1-8+0-1 0 0 0 10+0 0 12-6+0-5 2:9+0-3 1-6+0-1 narrow acuminate cream-violet 3-2+0-1 0 0-4+0-1 0 1:7+0-1 brown-purple Cirsium arvense BRITAIN Ripon Yorkshire 5 13.vii.1994 SE/31.73 481 +51-0 10-4 + 0-7 2:0+0 deep violet 156 + 19-6 2-0+0-1 4-7 + 0-2 b5'+ 0 linear-lanceolate violet-fawn 21-9+0-8 linear-lanceolate 2:0+0-1 mauve-violet 21-1+0-4 6:9 + (0-2 + campanulate 1-5+0 2:6+0-1 1-520 cream-purple 1-6 + 0-2 0 0 0) 1-:0+0 0 15-1 +0-6 3-3 +0-3 1-8+0-1 broadly tapering mauve-violet 3-4+0-2 0) 0) 0 1-9+0 brown-purple Cirsium arvense FRANCE Petit Mont Rond Jura 10 6.vill. 1995 46 20’N; 6 02’E 324 + 19-7 6:9 + 0-4 2:0+0 cream-mauve 155 + 14-5 2:5+0 4-0+0-1 15+0 linear-lanceolate yellow-cream 16-6 + 0-9 linear-lanceolate 2:0+0 cream 18-5 + 0-4 6:7 + 0-3 tubular-tubular- campanulate 10+0 2:2+0-4 1:-7+0-1 cream-yellow-purple 13+0-1 2:0+0 narrow, tapering cream-yellow 2:-4+0-1 0-2 +0-1 0-3 +0-1 0 2:4+0-1 brown-purple Carduus defloratus 264 MJY. gEOLEY: Table 2 shows where the main differences lie. Other than in the different degree of robustness mentioned above, the three British populations show close agreement in almost all important characters (Table 2 and Figure 1). This is especially notable for those related to the corolla (shape, size, height of filament insertion, filament pilosity and stigma lobe separation). Corolla shape, being relatively broad in relation to length and strongly curved, is especially characteristic and constant for the British plants and differs appreciably from the more tubular, much less curved corollas of those from mountain areas of France. Being qualitative, however, this character is not included in the PCA. The French plants also differ from the British in their lower level of filament insertion, their wider stigma lobe separation and in the presence of ciliate, denticulate lobes to the lower lip of the corolla. Corolla length in this population is also less than for British plants, but this may not be significant as it was found to be somewhat longer in other French populations seen but which were not morphologically assessed in detail. Broad similarities in corolla height of all four populations did not reflect the characteristic corolla shape apparent in the British plants, i.e. relatively broad in relation to length. During the course of this work and earlier, 44 populations of the British plant have been examined (Foley 1992a). All appeared to be morphologically similar but showed appreciable variation in colour, especially that of the stem, calyx and corolla. A few populations had individuals with very atypical coloration (Foley 1992b), some being uniformly rich yellow or occasionally pure white. Apart from these exceptional plants, corolla colour differences were usually due to the varying presence of mauve-purple glandular hairs and purple corolla venation. This could vary widely and sometimes partially obscured the basic cream to greyish base colour of the corollas. Plants seen in French montane/subalpine populations were similarly, but less variably, veined and embellished with mauve-purple glandular hairs, usually over a deeper, almost yellow, background. The above observations suggest that the French montane/subalpine and British plants belong to O. reticulata s. s. and O. procera respectively. Whether O. reticulata s. s. is a plant wholly restricted to higher altitudes may be open to question since Wallroth’s type specimen was collected from near Toulouse (France), some distance north of the Pyrenees, but no locality details are indicated. Although not in an ideal condition, the specimen conforms satisfactorily to plants seen by myself in the French mountains and, in corolla shape especially, it differs from the British plant in ways similar to those mentioned above. The British plants on the other hand largely agree with the description given by Koch (1833) for O. procera, which occurs in lowland habitats in Germany and elsewhere. This applies especially to corolla shape which is campanulate distally, inflated towards the base and curved; his description is much closer to the British plant than to that from the Jura. A similar comment regarding the distinctive corolla shape of O. procera is given by Rouy & Foucaud (1909) under O. platystigma var. procera: “corolle plus large, subcampanulée, et plus petite’, i.e. broader and shorter; this in comparison to other similar taxa within the group. Koch’s O. procera grew in fields close to the River Rhine; the British plant is also most frequently found near river margins and flood plains (Foley 1993) where it usually parasitises Cirsium arvense. Despite similarities existing between the lowland British and the European montane/subalpine plant, they can be readily distinguished by several clear-cut character differences. Due to this and the fact that they appear to occupy quite distinct habitats, often parasitise a different host genus and have a somewhat different geographical range, their appropriate taxonomic rank appears to be at the subspecific level. The lowland plant of Britain and Europe therefore becomes O. reticulata subsp. procera (Koch) Dostal and the montane/subalpine plant subsp. reticulata. The status of O. pallidiflora Wiramer & Grabowski, however, remains uncertain. From its original description it is unclear whether its affinities lie closer to subsp. reticulata or to subsp. procera. By its specific epithet (= pale-coloured flowers) and the authors’ brief reference to corolla colour, it would appear that this is its most distinctive character. If so, it may well be that light- coloured variants of subsp. reticulata (and even subsp. procera) are routinely referred to it. Although I have seen an appreciable number of herbarium specimens referred to O. pallidiflora (placed at that or lower ranks), most of these are closer to subsp. reticulata than to subsp. procera. In this genus, however, subtlety of colour is one of the first characters lost on preservation and it is one reason why investigations into this taxon have not been pursued here. It is possible that taxa referred to O. pallidiflora are mere colour forms of one or both subspecies of O. reticulata. No living populations were discovered during this work which could be reliably referred to this plant. OROBANCHE RETICULATA 265 CONCLUSIONS There are two distinctive, yet closely allied taxa: (a) O. reticulata subsp. reticulata, a plant of montane-subalpine occurrence, usually found at altitudes between 1000 and 2500 metres and apparently widespread but local in suitable areas of continental Europe, where its main host is Carduus defloratus, and (b) O. reticulata subsp. procera, a plant of lowland Europe (including Britain), often found in agricultural areas and in the vicinity of major rivers or flood plains where seed may dispersed by water-borne means. There are also other superficially similar lowland European plants which have been referred to O. pallidiflora but which are somewhat problematic. It is possible that these are simply colour variants and that such morphological differences as exist are at least partly determined by the different hosts parasitised. J. Zazvorka (litt. comm.), based on his experience in (the then) Czechoslovakia, goes even further and considers that all taxa embraced by O. reticulata s. |. are really a single taxon whose differing morphology is host-determined. There is no clear evidence for this, but the varying degree of robustness is one character which may well be host-determined. TAXONOMIC DESCRIPTION OROBANCHE RETICULATA WALL. SUBSP. RETICULATA (syn.: O. scabiosae Koch; O. pallidiflora auct., O. platystigma Rchb.; O. carlinoides Miégeville; O. cardui Sauter; O. scabiosifixa St Lager). Described from southern France “Jn agro Tolotono”. Type collected at Toulouse in 1807 by D. Fliigge (LE!). Typically 300-350 mm tall, with a fairly dense, many-flowered inflorescence. Stem: 6-8 mm wide immediately below the inflorescence, distinctly glandular-pubescent, cream-mauve. Leaves: maximum width c. 4 mm, linear-lanceolate, yellow-cream, glandular-pubescent. Bracts: 12-18 mm long, linear-lanceolate, cream, rapidly darkening with age. Calyx: c. 10-12 mm long, segments usually unequally divided, sometimes entire, teeth narrow, tapering, cream-yellow, glandular-pubescent. Corolla: 18-22 mm long, suberect, tubular to tubular-campanulate, yellow- deep cream with mauve-purple veins and varying amounts of purple glandular hairs mainly distally on upper lip, dorsal line not strongly curved. Upper lip slightly bilobed; lobes of lower lip somewhat ciliate and denticulate, rounded, + similar in size. Filaments: inserted at 2-0-2-5 mm above the corolla base, very slightly glandular-pilose to glabrous above, + glabrous below but sometimes with a few hairs at the actual base. Stigma lobes: dark purple-brown with yellow undertones, distinctly separated. Parasitic mainly upon Carduus defloratus, but probably also on other taxa including Knautia spp. A plant mainly of montane to subalpine habitats over a range of substrates, perhaps descending to lower altitudes close to its main area of distribution. Flowers June-August (—September). Widespread but local in suitable mountain habitats of mainland Europe including the Alps, the Pyrenees and the Balkans. Its distribution outside this area is unclear at present. OROBANCHE RETICULATA SUBSP. PROCERA (KOCH) DOSTAL (syn.: O. pallidiflora auct.; O. cirsii Fries; O. mathesii Schultz: O. serotina Kirschleger). Described from Germany “auf lehmigen Aeckern in Gegend von Mannheim in der Rheinpfalz... [etc.]”. Specimens collected by Schimper but type not located. Similar to subsp. reticulata but more robust and taller (400-550 mm); corolla: broader and almost campanulate, quite strongly curved, lobes of lower lip not noticeably ciliate or denticulate; filaments: inserted higher (3-0-3-5 mm) above the corolla base; stigma lobes: less widely separated. Parasitic on Cirsium spp., especially Cirsium arvense and related genera. A plant mainly of lowland habitats, river banks and flood plains, field borders, ruderal habitats, roadsides, unimproved pastures, scrub, woodland clearings and perhaps occasionally reaching submontane grasslands. Flowers June-August. Widespread throughout much of central Europe north of the Alps, including Germany and Poland and with disjunct occurrences in eastern England (Yorkshire). 266 VAY FOLEY ACKNOWLEDGMENTS I am most grateful to the trustees of the Welch Bequest Fund of the Botanical Society of the British Isles and to the Botanical Research Fund (Kew) for financial assistance during this project. REFERENCES BATEMAN, R. M, & DENHOLM, I. (1983). A reappraisal of the British and Irish dactylorchids, 1. The tetraploid marsh-orchids. Watsonia 14: 347-376. BECK, G. (1890). Monographie der Gattung Orobanche. Bibliotheca Botanica, 19: 1-267. Cassel. BECK, G. (1930). Orobanchaceae, in ENGLER, A., ed. Das Pflanzenreich IV. 261: 1-348. Willhelm Engelmann, Berlin. CHATER, A. O. & WEBB, D. A. (1972), Orobanche in TUTIN, T. G. et al., eds. Flora Europaea 3: 286-293. Cambridge University Press, Cambridge. COUTINHO, A. X. P. (1913). Flora de Portugal. Aillaud, Lisboa. FEINBRUN-DOTHAN, N. (1978). Flora Palaestina, 3. Israel Academy of Sciences and Humanities, Jerusalem. FOLEY, M. J. Y. (1990). An assessment of populations of Dactylorhiza traunsteineri (Sauter) So6 in the British Isles and a comparison with others in continental Europe. Watsonia 18: 153-172. FOLEY, M. J. Y. (1992a). Survey for Orobanche reticulata (Thistle Broomrape) in North Yorkshire. English Nature, unpublished. FOLEY, M. J. Y. (1992b). Some British Orobanche variants. BSBI News 60: 64. FOLEY, M. J. Y. (1993). Orobanche reticulata Wallr. populations in Yorkshire (north-east England). Watsonia 19: 247-257. FOLEY, M. J. Y. (1998). Taxonomic problems in European members of the genus Orobanche L. Ph.D. Thesis, University of Lancaster. GILLI, A. (1965). Orobanche, in HEGI, G., ed. Illustrierte Flora von Mittel-Europa, 6(1): 470-505. Carl Hanser Verlag, Miinchen. GILLI, A. (1982). Orobanchaceae, in DAVIS, P. H., ed. Flora of Turkey and the East Aegean Islands 7: 1-23. University Press, Edinburgh. GUIMARAES, J. d’A. (1904). Monographia das Orobanchaceas Portuguezas. Broteria 3: 5—207. Lisboa. HARTVIG, P. (1991). Orobanche in STRID, A & TAN, K., eds. Mountain Flora of Greece, 2: 263-276. Edinburgh University Press, Edinburgh. JONES, M. (1989). Taxonomic and ecological studies in the genus Orobanche L. in the British Isles. Ph. D. Thesis, University of Liverpool, Liverpool. Kocu, W. D. J. (1833). J. C. Rohlings Deutschlands Flora 4. Friedrich Wilmans, Frankfurt. KREUTZ, C. A. J. (1995). Orobanche: Die Sommerwurzarten Europas 1. Stichting Natuurpublicaties Limburg, Maastricht. MEIKLE, R. D. (1985). Flora of Cyprus 2. Bentham-Moxon Trust, Royal Botanic Gardens, Kew. PIGNATTI, S. (1982). Flora d'Italia 2. Edagricole, Bologna. ROBERTS, R. H. (1966). Studies on Welsh Orchids, 3. The coexistence of some tetraploid species of marsh- orchid. Watsonia 6: 260-267. ROBERTS, R. H. (1988). The occurrence of Dactylorhiza traunsteineri (Sauter) Sod in Britain and Ireland. Watsonia 17: 43-47. ROBERTS, R. H. & GILBERT, O. L. (1963). The status of Orchis latifolia var. eborensis Godfery in Yorkshire. Watsonia 5: 287-293. ; Rouy, G. & FOUCAUD, J. (1909). Flore de France 11. Les Fils d’Emile Deyrolle, Paris. RUMSEY, F. J. & JURY, S. L. (1991). An account of Orobanche L. in Britain and Ireland. Watsonia 18: 257- 29); SCHIMAN-CZEIKA, H (1964). Orobanchaceae, in RECHINGER, K. H. ed. Flora Iranica, 5: 1-25. Akademische Druck und Verlagsamstalt, Graz. STACE, C. A. (1991). New Flora of the British Isles. Cambridge University Press, Cambridge. UHLICH, H., PUSCH, J. & BARTHEL, K-J. (1995). Die Sommerwurzarten Europas: Gattung Orobanche. Die Neue Brehm-Biicherei Bd. 618 Westarp Wissenschaft, Magdeburg. VALDES, B., TALAVERA, S. & FERNANDEZ-GALIANO, E. (1987). Flora Vascular de Andalucia Occidental, 2. Ketres Editora S. A., Barcelona. WALLROTH, C. F. W., (1825). Orobanches generis diaskene. Fredaricum Wilmans, Frankfurt. WIGGINTON, M. J. (1999). British Red Data Books 1. Vascular Plants, 3rd ed. Joint Nature Conservation Committee, Peterborough. WILLKOMM, M. & LANGE, J. (1870). Prodromus Florae Hispanica, 2. Sumptibus E. Schweizerbart, Stuttgart. WILKOMM, M. (1893). Supplementum Prodromi Florae Hispanicae. Sumptibus E. Schweizerbart, Stuttgart. OROBANCHE RETICULATA 267 WIMMER, F. & GRABOWSKI, H. E. (1829). Flora Silesiae, 2. Guilielmum Theophilum Korn, Breslau. WISHART, D (1987). Clustan User Manual, 4th ed. Computing Laboratory, University of St. Andrews. (Accepted July 1999) FL Peo r,s de Rt Ie SE SORIA yo Soh MDE TCR em an REI a IRI TE OC RA LEE I LOE LOI I NE ORS AAI LOPLI IED AE AD DAYAR BO ELL TIS Ba {Se OPE EA mt Ie EUR As *, Watsonia 23: 269-278 (2000) 269 Viola rupestris and its hybrids in Britain B. JONSELL* Bergius Foundation, Royal Swedish Academy of Sciences, Box 50017, S-104 O05 Stockholm, Sweden I. NORDAL Department of Biology, University of Oslo, P.O. Box 1045, N-0318 Oslo, Norway and F. J. ROBERTS Eden Croft, 2 Wetheral Pasture, Carlisle CA4 SHU ABSTRACT Hybrids between Viola riviniana and V. rupestris were identified by means of isoenzymes from three of the four localities for V. rupestris known from Britain. Differentiating characters between V. riviniana and V. rupestris are primarily found in leaf shape and indumentum, flower colour, spur shape and capsule shape. The ecology of V. rupestris is presented and compared with that of the hybrid and V. riviniana. KEYWORDS: Viola riviniana x V. rupestris, isoenzymes, morphology, distribution, ecology. INTRODUCTION In Britain Viola rupestris F. W. Schmidt has a very restricted distribution with four locality groups in northern England. Hybrids with V. riviniana (V. x burnatii Gremli) have been recognized or suspected in those localities but the identification on morphological grounds has been regarded as provisional owing to lack of knowledge about variation in the parental species. Valentine (1975) still recognized the hybrid from only one of the V. rupestris localities, Widdybank Fell (Teesdale). The aim of this study is to identify British hybrids by means of isoenzymes and to discuss their morphology, ecology and distribution against that background. V. rupestris has a wide palaearctic distribution, seemingly more-or-less continuous from eastern Siberia to northern and central Europe. To the west it reaches the Pyrenees, south-eastern France, western Germany and south-eastern Norway and has isolated locality groups on the Dutch dune coast and in northern England. Montane populations in northernmost Scandinavia are distinguished as subsp. relicta Jalas (Jalas 1950; Nordal & Jonsell 1998); otherwise no subspecies have been described. V. rupestris has the chromosome number 2n = 20, which is the lowest number known in the section Viola, but evidence from isoenzyme patterns has revealed that the basic number is n = 5 (Marcussen & Nordal 1998). V. rupestris belongs to the subsect. Rostratae, which in Britain is represented also by V. persicifolia Schreber (2n = 20), V. reichenbachiana Jord. (2n = 20), V. canina L. (2n = 40), V. riviniana Reichenb. (2n = 37-46, according to Valentine 1949) and V. lactea Smith (2n = 58). Crossing experiments between a number of these species were performed by Moore & Harvey (1961), who made conclusions about the distribution and sharing of genomes among them. V. rupestris was, however, not included in their study. For Britain Valentine (1975) reported only V. riviniana x rupestris, while previous records of the hybrid with *Address for correspondence: e-mail bengtj @bergianska.se 270 B. JONSELL, I. NORDAL AND F. J. ROBERTS V. canina were discarded as being based on V. riviniana. Outside Britain V. rupestris hybrids have been reported with V. canina and V. riviniana (and even with V. hirta L., which seems highly doubtful). For combinations with the two first-mentioned evidence at least from Scandinavia seems substantiated (Nordal & Jonsell 1998). The progeny of crossings undertaken between V. canina and V. rupestris and between V. riviniana and V. rupestris have shown pollen fertility between 0 and 5% and 0 and 14%, respectively (Bethke 1882; Erdner 1907; Becker 1910; Schofer 1954; Schmidt 1961; Harvey 1966). In a number of species of Viola sect. Viola it has been shown that isoenzymes vary little if at all within species and therefore can be used effectively as markers for the distinguishing of hybrids (Marcussen & Nordal 1997; Nordal & Jonsell 1998). MATERIALS AND METHODS All known Viola rupestris localities in England (Fig. 1) were visited by the authors from 8-12 May 1998 (Table 1, sites 1-4, 8, 12-13). Permission was granted by English Nature and landowners for very restrained collection of material. The intrapopulational homogeneity known from Nordic representatives of the sect. Viola (subsect. Rostratae, cf. Nordal & Jonsell 1998) justified restricted collection, an important consideration due to the relative rarity of the species. Analyses were undertaken on an average of about four (in practice 1-7) individuals per site. In most cases the sampling was non-destructive with removal of only a single leaf from the plant. A few whole specimens were transferred to a greenhouse at the University of Oslo for comparative cultivation. V. riviniana was sampled from the same or close sites (Table 1, sites 1-5, 8-13). In addition a particular search was made for putative hybrids. The removed leaves were placed in plastic bags and kept cold with freezing elements for one week until they arrived at the laboratory. The transferred plants were immediately planted in pots and grown under greenhouse conditions with an 18 hour day at 16°C and a 6 hour night at 12°C. TABLE 1. LOCALITIES AND ACCESSION NUMBERS OF SAMPLED POPULATIONS Site Grid reference V. rupestris hybrid V. riviniana Ingleborough 1. Near Crummack Farm SDITT.T1 7650-2, 3 1652-15225 105 1=2, 3,453 7653-1 2. Long Scar SD/76.72 7654-1, 2, 3, 4,5 7655-1 3. Sulber Gate SD/77.72 7656-3, 5 7656-2, 4, 6 4. Near Crummack Farm SD/76.71 7659-1 7660-1, 2, 3 5. Near Crummack Farm SDT Tai 7664-1, 2, 3 6. Near Crummack Farm SD/77.71 37a 2.35 7. Long Scar SD/76.71 3773-1, 2 Teesdale 8. Widdybank Fell NY/81.29 7665-1, 2,4,7,8 7665-3 7666-2 7666-1, 3, 4 7667-15253 9. Widdybank Pasture NY/84.30 7668-1, 2, 3 10. High Force NY/88.238°". 767 141,2 7669-1, 2, 3 11. Cronkley Fell NY/84.28 7674-1 Long Fell (Brough) 12. Long Fell NY/76.19 7678-6, 7, 8 7676-1, 3 7676-2 7677-1, 2 7678-3, 4 Arnside Knott 13. Arnside Knott SD/45.77 1679-1, 2,354 7680-1, 2 VIOLA RUPESTRIS AND ITS HYBRIDS 2K & Durham é A Widdybank Fell Ee Long Fell at North é ; Yorkshire Cumbria 8 Arnside Knott @ Ingleborough Morecambe P 0 1 2 3 4 5 6 R4 8 9 0 FIGURE 1. 1-km distribution of Viola rupestris (solid circles) and V. rupestris + V. x burnatii (half-circles) in Britain. The field-sampled leaves and one from each of the transplanted specimens were crushed in a grinding buffer (Morden et al. 1987). The crude extracts were absorbed into paper wicks for subsequent horizontal starch gel electrophoresis (Wendel & Weeden 1990). Altogether, seven enzyme systems were analysed: IDH (isocitrate dehydrogenase), MDH (malate dehydrogenase), and PGM (phosphoglucomutase) were run on the so-called “D- system”; AAT (aspartate aminotransferase), AMP (aminopeptidase), GPI (gluco-6-phosphate isomerase), TPI (triose- phosphate isomerase) in the “A/B- system” (Selander & Yang 1969; Soltis et al. 1983, see also Nordal & Jonsell 1998). After the plants had been in cultivation under similar conditions from May to October 1998 leaves from the (surviving) plants were pressed for comparison. ISOENZYME RESULTS Three of the isoenzymes revealed patterns that gave different band phenotypes in Viola rupestris and V. riviniana and an additive pattern in the hybrids: GPI, PGM and TPI (Fig. 2). The pattern was totally consistent so that hybrid constitution in one isoenzyme system always inferred hybrid constitution in the other two. The enzymes AAT, IDH and MDH did not differentiate between the two species, and accordingly cannot be used in hybrid analyses. The results for AMP indicate intraspecific variation in V. rupestris, which needs to be further elucidated. The dimeric enzyme GPI showed activity in two regions (Fig. 2). In the anodal region two bands, always blurred, were displayed in both species and the putative hybrids. In the cathodal region a three-banded pattern was always found in V. rupestris, best interpreted as two allelic bands with an intermediate heterodimer, in a pattern of fixed heterozygosity, which may be expected in an allotetraploid. The octoploid V. riviniana always displayed a five-banded pattern, in 272 B. JONSELL, I. NORDAL AND F. J. ROBERTS GPI PGM TPI R; values FIGURE 2. Isoenzyme results for gluco-6-phosphate isomerase (GPI), phosphoglucomutase (PGM) and triose- phosphate isomerase (TPI) for Viola riviniana (RIV), V. x burnatii (x) and V. rupestris (RUP). which the second and the fourth bands probably constitute heterodimers. With this interpretation V. rupestris has the genotype BD and V. riviniana ACD in this complex “locus”. The hybrids displayed an ABCD pattern with most of the constituent heterodimers, except the CD heterodimer, present in the individuals. In PGM (a monomeric enzyme) activity was also revealed in two regions (Fig. 2). In the anodal area a fixed, identical two-banded pattern was shared by the two species. In the cathodal area V. riviniana most often showed two bands (A and B), whereas V. rupestris only revealed the B-band. In this case the hybrid pattern was identical with that of V. riviniana. In very few cases V. riviniana displayed a band in a more cathodal (C) position, and one hybrid (7665-3) displayed a triple-banded ABC pattern (not illustrated). TPI (a dimeric enzyme) revealed a fairly complicated multibanded pattern, difficult to break down into allelic interpretation. The bands may be referred to three regions. The most anodal displayed one band shared by both species. Clear variation between the two species was found in the next region, where V. riviniana had activity in front of V. rupestris (Fig. 2) and where the hybrid revealed additivity. V. riviniana displayed two distinct bands in this region and V. rupestris three. The latter is best interpreted as a fixed heterozygote (BD) with an intermediate heterodimer. V.riviniana might have a third indistinct band and if so an AC-genotype, or the two distinct bands may represent two monomorphic loci which do not interact. These alternative interpretations will not influence the testing of the hybrid hypothesis. In the hybrids, with a putative AB(C)D constitution, the bands tend to fuse, as visualised in Fig. 2. In the third region both species showed a fixed three-banded pattern, - much weaker, however, in V. riviniana than in V. rupestris and the hybrid. MORPHOLOGICAL OBSERVATIONS SEPARATION OF VIOLA RUPESTRIS FROM V. RIVINIANA Although a distinct species, Viola rupestris is sometimes difficult to distinguish from V. riviniana, especially when this species is represented by small-flowered forms (var. minor (Murbeck ex Gregory) Valent.) as is the case in most British localities where it grows with V. rupestris. The confusion is reinforced by the fact that V. rupestris is not consistently pubescent along petioles and peduncles and on capsules as is generally stated in current Floras (e.g. Stace 1997). VIOLA RUPESTRIS AND ITS HYBRIDS 273 The most reliable separating characters for British plants of V. rupestris from V. riviniana are as follows: i. Leaf shape: the laminas of V. rupestris are ‘shovel-’ or “scoop’-shaped, with a clearly blunt apex, and with a less strongly cordate base than in V. riviniana. In V. rupestris the sides of mature leaves are often curled upwards (hence the ‘scoop’), whereas in V. riviniana it is the basal lobes which are often upcurled. The areas of the lamina between the veins are somewhat raised in most forms of V. riviniana, giving a + dimpled effect to the blade; in V. rupestris the surface is flatter. The crenations at the leaf-edges are flat (i.e. within the plane of the leaf) in V. rupestris, but in V. riviniana are slightly ‘crimped’, and thus more obvious. In Fig. 3 leaf- silhouettes demonstrate, on limited material, clear differences particularly in the cordate shape of the blade, the rounded blade apex in V. rupestris, and the intermediate nature of the hybrid. Indumentum: the highly distinctive fine, very short indumentum of V. rupestris usually extends up the peduncles, and up the petioles onto the margins of the basal lobes, leaving most of the upper lamina essentially glabrous in the great majority of plants. (In some variants, the fine indumentum extends all over the upper and lower surfaces of the lamina.) Where dense, the indumentum produces the impression of a grey ‘fuzz’ to the naked eye. It should be noted that some V. rupestris individuals are glabrous even along petioles and peduncles. Such individuals were found, and confirmed by isoenzyme analysis, in all the localities except for Arnside Knott. Furthermore, plants may often be found with indumentum on some petioles but not on most. The forms of Viola riviniana which accompany V. rupestris, while lacking this fine indumentum of dense short hairs, almost invariably carry on the upper lamina sparse long bristly hairs which are about 250-330 um long, and thus typically about six to eight times longer than the indumentum ‘hairs’ of V. rupestris, which measure around 35-50 um. These sparse, long, colourless hairs of V. riviniana can be inconspicuous, and are best observed by looking with a lens across the upper surface of a leaf bent over the finger to create a convex surface. Flower colour: the corolla colour in V. rupestris is often paler and less blue than in V. riviniana, being a ‘delicate pinkish-mauve’ in Teesdale plants (M. E. Bradshaw, pers. comm. ), although the colour may vary in both species, and also alters with exposure, bleaching and ageing processes. Looking into the flower from the front, in Viola riviniana there 1s usually a narrow band of darker violet separating the colour on the lower lip from the white in the throat; this band is apparently always lacking in Viola rupestris. The flowers of the Arnside Knott colony, which have been sparsely produced in recent seasons, are always pure white, as first noted by Valentine & Harvey (1961). The plant is illustrated in Halliday (1997). Spur: the spur of Viola rupestris is short, conical to a round tip without obvious groove or projection. The spur in Viola riviniana is long, stout, and often vertically grooved at the end, and often with a small projection from its upperside near the tip. (The colour of the spur varies in both species.) Ripe capsule: the ripe capsule of V. rupestris is almost globular with truncate apex in side aspect, whereas that of V. riviniana is more elongated, and most often has an obtuse or even acute apex. V. rupestris produces rosettes on lateral branches, which are often hidden within the substrate. Such branches are non-rooting, so that the rosettes do not become independent plants, although often giving this appearance (Bradshaw & Doody 1978b). V. riviniana, however, has genuine vegetative propagation in that adventitious shoots (soboles) may develop on the root system (Valentine 1949). It is worth noting that the forms of V. riviniana which occur in exposed sites close to V. rupestris can approach that species. In comparison with V. riviniana forms in nearby, more sheltered, sites such as block scree and fissures in limestone pavement they have a deeper green colour overall, are more prone to developing purple tints, have a much more condensed form with short petioles producing a tight rosette, have more leathery leaf texture, and smaller flowers. 274 B. JONSELL, I. NORDAL AND F. J. ROBERTS eae Viola rupestris Viola riviniana x rupestris (V. x burnatit) Viola riviniana FIGURE 3. Leaf silhouettes of Viola rupestris, V. x burnatii and V. riviniana. Scale c. x 0-6. HYBRID MORPHOLOGY Detection of the generally rare hybrids in the often abundant and dense populations of the species is not straightforward, due to variation of the parent species in all of the above characters, and the low stature of the plants. In plants examined to date, hybrids may be suspected by an intermediate leaf-shape (Fig. 3) and by the presence both of the dense short indumentum of V. rupestris - albeit patchy - on petioles and stems, and of the sparse longer hairs of V. riviniana on the upper leaf lamina. Some hybrid plants can appear glabrous, as with a proportion of the V. rupestris parent; however, a few hairs, sometimes of intermediate length and form, will generally be found on upper laminas, particularly on the midrib or patchily towards the basal lobes. Where the indumentum is well-developed, it is most abundant on the upper part of peduncles, but is less reliably found on petioles. Chasmogamous (i.e. opening) flowers appear to be produced less frequently in most seasons than by the parents. In colour and shape they tend to bridge the rather narrow differences between the parents. Flower shape tends to approach V. rupestris rather than V. riviniana. The dark bar in the throat usually characteristic of V. riviniana is in the hybrid faint or more-or-less absent. The dark purple colour of the veins (honey-guides) varies in density, and where they run out onto the mauve of the lower petal the pigmentation becomes more diffuse, contributing to the effect of a bar. ; Aborted flowers and lack of fruits late in the season (when cleistogamous flowers have been produced) strongly indicate hybrid plants. The hybrid is thought never to ripen capsules, so that the presence of ripe capsules in late summer, often abundant on mature plants, reliably indicates the species. Hybrids can most easily be detected at this time when the plants attain their greatest stature and leafiness. The soboles of V. riviniana are apparently inherited by at least some hybrid plants, allowing indefinite survival of individual plants. In the Long Fell (Brough) and Widdybank Fell (Teesdale) localities, hybrid patches of considerable size (and perhaps age) are known, up to 1—2 metres in diameter or more, and giving the appearance of clonal derivation from rare hybridisation events. VIOLA RUPESTRIS AND ITS HYBRIDS 275 The production of adventitious shoots is particularly vigorous in the hybrid; on Widdybank Fell densities of rosettes were recorded on trial plots of 656/m? for the hybrid, 149/m? for V. rupestris, and 175/m? for V. riviniana (Bradshaw & Doody1978a). ECOLOGICAL OBSERVATIONS HABITATS OF VIOLA RUPESTRIS In all its four English locations, Viola rupestris grows on exposures of Carboniferous limestone. V. rupestris tends to occur in the more open and more exposed habitats available, on the thinnest soils. It has two typical favoured habitats: one is on the eroding tops and flanks of low limestone hummocks growing either in fine cracks in the bedrock, or in a ‘clitter’ of small stones lying over the bedrock or as gently sloping screes; the other is in fine screes (more-or-less consolidated) below low escarpments. It also occurs much more sparsely on the ‘clints’ (or flat tops) of limestone pavements, but only where the rock surfaces are sufficiently fissured or the rock is covered in ‘clitter’ - the loose stones resulting from degradation and fracture of the bedrock. On the sites south-east of Ingleborough, some of the largest colonies are also on gently- to steeply-sloping faces of drift deposits over limestone. Bare clayey soil, evidently calcareous, is a feature of this habitat, the open nature of the sites apparently being maintained over long periods by soil-creep and by stock trampling. The colonies lie between 320-425 m a.s.1. On Widdybank Fell the plant occupies exposures of granular partially-metamorphosed limestone (‘sugar limestone’), the rapid erosion of which creates zones of loose calcareous rubble and coarse calcite ‘sand’. Viola rupestris appears to be an early colonist of such bare soils, with little else besides a thin turf or occasional tufts of Festuca ovina L. and plants of Minuartia verna (L.) Hiern. The species persists in closed Festuca turf, although flowering and fruiting less vigorously, and only where some disturbance (such as by moles) maintains areas of open soil. The development of denser closed swards appears eventually to oust the violet. Plants occur from 490-510 m a.s.1. The Long Fell colonies lie at 575-610 m a.s.l. on a shelving S.S.W.-S.E.- facing escarpment, the plants growing on slopes above cliffs, with a few patches on narrower ledges below the uppermost low cliffs of the scarp. Eighteen colonies were located in 1998. The plants are consistently dwarf in comparison with those at the other three sites; it is unclear whether this dwarfing is a response to the greater exposure and altitude (Heaton 1999). The dwarf form appears to be retained in cultivation. Its most typical closest associates in the upland sites, whether on un-metamorphosed limestones or calcareous boulder clay, are Carex flacca Schreber, C. panicea L., Carlina vulgaris L., Euphrasia confusa Pugsley, Sesleria caerulea (L.) Ard., and Thymus polytrichus Borbas. It can be found in open turf of Festuca ovina and Sesleria caerulea, but rarely in dense turf of either species. At Arnside Knott there exists the only colony of Viola rupestris known off the ‘high’ limestone. This site is on an exposed escarpment, in a very limited area free of scrub at 150 m a.s.]. (Lists of associates in the Arnside, Long Fell, and Widdybank localities are given in Valentine & Harvey 1961). HABITATS OF VIOLA RIVINIANA Viola riviniana almost invariably occurs in the near vicinity at all four sites, but is seldom intermingled. It obviously survives in closed turf of Festuca ovina, where V. rupestris is rare or absent, but conversely it avoids open ‘clitter’ areas of small stones or fine scree, colonised by V. rupestris. In a very dwarf and condensed form it is widely present also in the closed Agrostis/ Festuca grasslands between the limestone outcroppings. As a much larger and laxer plant with large flowers it occurs, often abundantly, in the hollows between larger stones and in the ‘grykes’ (vertical fissures) of limestone pavement. HABITATS OF HYBRIDS The hybrid prefers deeper soils, and can tolerate denser turf, than V. rupestris, but has not yet been seen far from colonies of the rarer parent. On Widdybank Fell the habitat consists of semi-open or closed swards generally intermediate between those preferred by the parents. It also occurs with 276 B. JONSELL, I. NORDAL AND F. J. ROBERTS Calluna on calcareous brown-earth soils, where moles are active (Bradshaw & Doody 1978b). On Long Fell one patch of hybrid plants was on a gently-sloping south-facing patch of loose stones immersed in black humus-rich soil, over bedrock. The few plants on Ingleborough have been found in open patches on drift covered in loose stones, between areas of Festuca ovina turf, within a few centimetres of the parent species. POPULATION SIZES OF VIOLA RUPESTRIS AND THE HYBRID Viola rupestris occurs most typically in local, dense colonies of a few to very many rosettes within limited tracts of its preferred habitat, whereas V. riviniana occurs more dispersed, but over much more extensive turf habitats between outcroppings of limestone. Work on establishing the population sizes and spatial limits of V. rupestris is continuing. As previously remarked, due to the difficulty of establishing the origin of rosettes, it has been usual to regard each rosette as an individual for recording purposes. Surveys in 1998 on parts of the known distribution on the south- east spurs of Ingleborough suggest that the population of the whole will amount to tens or hundreds of thousands of individuals. The Widdybank sites are estimated to have thousands of individuals (M. E. Bradshaw, pers. comm.), whilst the population on Long Fell was estimated in 1998 at between 1000 and 2000 plants. Arnside Knott had 56 plants in 1998 (Heaton 1999), Hybrid numbers in ‘the high hundreds’ are known on Widdybank Fell, with ten dense colonies and scattered groups and singles (M. E. Bradshaw, pers. comm.). It appears to be more localised on Long Fell, although this area has been worked much less thoroughly than has Widdybank Fell; the patches known in 1998 had a few dozen rosettes. On the Moughton/Crummack/Norber sites south-east of Ingleborough, however, only a few individual hybrid plants have yet been located in over 50 colonies of V. rupestris examined. DISCUSSION In the presented isoenzyme patterns of GPI and TPI, Viola rupestris in England displays exactly the same band patterns as were found in Nordic populations of the same species (Nordal & Jonsell 1998). The A-band in PGM in the English plants is shared with plants belonging to the North Scandinavian V. rupestris subsp. relicta and is different from the widespread ssp. rupestris. This indicates an interesting connection which will be analysed in a wider geographical context (Jonsell & Nordal in prep.). The V. riviniana plants analysed from the English localities have different band patterns compared to Nordic material in all the isoenzymes presented. Our English V. riviniana material consistently showing this isoenzyme pattern includes both small-flowered forms which often grow together with V. rupestris and larger-flowered ones from sites in the vicinity. Based on enzymatic evidence hybridisation was revealed in ali but one of the four British areas for V. rupestris. The exception is Arnside Knott though V. riviniana is frequent both on and close to the spot. In two of them (Widdybank Fell and Long Fell) the hybrid was previously known, while the record from Crummockdale (Ingleborough) is new (for details see Table 1). Of five sites for V. rupestris sampled in the Ingleborough area, where V. riviniana is common throughout, the hybrid was revealed only in one. This suggests that hybridization is comparatively rare there, although one has to allow for the difficulties of distinguishing hybrids in the field. The hybrids analysed give the impression of being F1 individuals and we have seen no traces of back-crossing to parents. This is in contrast to the situation in parts of Scandinavia where, according to preliminary observations (Nordal & Jonsell 1998), there are trends towards genetic swamping of V. rupestris, where this species is rare. There, however, V. canina seems usually to be the other parental species, though hybridisation with V. riviniana has also been documented (Nordal & Jonsell 1998). In many seasons V. rupestris is sparsely-flowered - at least as regards chasmogamous flowers, which appear mainly during May. Thus, given the very exposed habitats in which it typically grows, with little insect activity in the early season, the chances of cross-pollination of its chasmogamous flowers with those of V. riviniana are clearly low. These factors may account for the observed rarity of hybrids in most sites. Although both species appear to regenerate efficiently from seed the vast majority of seeds are likely to result from selfing within cleistogamous flowers. VIOLA RUPESTRIS AND ITS HYBRIDS 277 Whilst the fell-field habitats are undoubtedly over-grazed by sheep at present, the resultant erosion helps to retain open unstable areas suitable for colonisation of V. rupestris, but less favourable for propagation of hybrid clones, which need more stable substrates. The sterility of any hybrids, and consequent apparent lack of back-crossing, coupled with the abundant seed-production of the pure species, dictate against any swamping of Viola rupestris populations in the foreseeable future. The close spatial juxtaposition of the habitats of the two species means that they may be found in close proximity. The apparent tolerance of V. rupestris for eroding habitats with their mineral soils, where it evades competition not only from V. riviniana but also the turf-forming grasses, appears to have allowed its local survival. An observation by F.J.R. requiring further investigation is that in other areas of the north of England where Carboniferous limestone is exposed in apparently similar situations and climate, V. riviniana appears to occupy the types of microhabitat preferred by V. rupestris in its few stations. ACKNOWLEDGMENTS We would like to thank the following for help in various ways: Dr M. E. Bradshaw, Erica Donnison (English Nature, Cumbria Team), Peter Corkhill (Ingleborough National Nature Reserve), Kristian Heaton (Edge Hill University College), Stephen Hewitt (Tullie House Museum, Carlisle), John Hooson (National Trust), Chris McCarty (Teesdale National Nature Reserve), Major Ken Miles (Range Officer, M.O.D. Warcop Training Range), and Jacqui Ogden (English Nature, Cumbria Team). The distribution map in Fig. 1 was prepared using DMAP software produced by Dr Alan Morton. REFERENCES BECKER, W. (1910). Violae Europeae. C. Heinrich. Dresden. BETHKE, A. (1882). Uber die Bastarde der Veilchen-Arten. Diss. K6Onigsberg. BRADSHAW, M. E. (1985). Studies on the flora of Teesdale. Naturalist 110: 3-21. BRADSHAW, M. E. & Doody, J. P. (1978a). Chapter 2: Population-dynamics and Biology, in CLAPHAM, A. R., ed. Upper Teesdale, the area and its Natural History. Collins, London. BRADSHAW, M. E. & Doopy, J. P. (1978b). Plant population studies and their relevance to nature conservation. Biological conservation 14: 223-242. CLAPHAM, A. R., ed. (1978) Upper Teesdale, the area and its natural history. Collins, London. ERDNER, E. 1907. Sind die Veilchenbastarde fruchtsbar oder nicht? Allg. Bot. Zeitschr. 7/8: 117-118. GRAHAM, G. G. (1988). The flora & vegetation of County Durham. The Durham Flora Committee & Durham County Conservation Trust. HALLIDAY, G. (1997). A Flora of Cumbria. Shanleys, Bolton. HARVEY, M. J. (1966). Cytotaxonomic relationships between the European and North American rostrate violets. New Phytol. 65: 469-476. HEATON, K. (1999). Morphological variation in the British populations of the Teesdale violet (Viola rupestris F.W.Schmidt), pl. M.Sc. dissertation, Edge Hill University College JALAS, J. (1950). Zur Kausalanalyse der Verbreitung einiger nordischen Sand- und Ospflanzen. Annales botanici societatis zoologicae botanicae Fennicae ‘Vanamo’ 24: 1-362. MARCUSSEN, T. & NORDAL, I. (1997). Viola suavis M. Bieb., a new species in the Nordic flora, with the analysis of the relation to other species in the subsection Viola. Nordic journal of botany 18: 221-238. Moore, D. M. & HARVEY, M. J. (1961). Cytogenetic relationships of Viola lactea Smith and other West European arosulate violets. New phytologist 60: 85-95. MORDEN, C. W., DOEBLEY, J. & SCHERTZ, K. F. (1987). A manual of techniques for starch gel electrophoresis of Sorghum isozymes. The Texas Agricultural Experimental Station. The Texas A and M Univ. System, College Station, Texas. NORDAL, I. & JONSELL, B. (1998). A phylogeographic analysis of Viola rupestris: three post-glacial immigration routes into the Nordic area? Botanical journal of the Linnean Society 128: 105-122. ROBERTS, F. J. (1977). Viola rupestris Schmidt and Juncus alpinus Vill. in Mid-W. Yorkshire. Watsonia 11: 385-386. SCHMIDT, A. (1961). Zytotaxonomische Untersuchungen an europdischen Viola-Arten der Sektion Nominium. Osterr. Bot. Zeitschr. 108: 20-88. SCHOFER, G. (1954). Untersuchungen tiber die Polymorphie einheimischer Veilchen. Planta 43: 537-565. SELANDER, R. K. & YANG, S. Y. (1969). Protein polymorphism and genetic heterozygosity in wild populations of house mouse (Mus musculus). Genetics 63: 653-657. 278 B. JONSELL, I. NORDAL AND F. J. ROBERTS SOLTIS, D. E., HAUFLER, C. H., DARROW, D. C. & GASTONY, G. J. (1983). Starch gel electrophoresis of ferns: A compilation of grinding buffers, gel and eletcrode buffers, and staining schedules. American fern journal 73: 9-27. STACE, C. A. (1997). New Flora of the British Isles, Ed. 2. Cambridge Univ. Press. VALENTINE, D. H. (1949). Vegetative and cytological variation in Viola riviniana Rchb., in WILMOTT, A. J., ed. British flowering plants and modern systematic methods (conference report). B.S.B.I. Pp. 48-53. VALENTINE, D. H. (1975). Viola L., in STACE, C. A., ed. Hybridization and the flora of the British Isles, pp. 154-163. Academic Press, London. VALENTINE, D. H. & HARVEY, M. J. (1961). Viola rupestris Schmidt in Britain. Proceedings of the Botanical Society of the British Isles 4: 429-435. WENDEL, J. F. & WEEDEN, N. F. (1990). Visualization and interpretation of plant isozymes, in SOLTIS, D. E. & SOLTIS, P. S., eds. Isozymes in plant biology, pp. 5-24. Chapman and Hall, London. (Accepted March 2000) Watsonia 23: 279-286 (2000) 79 Carex muricata L. subsp. muricata (Cyperaceae) - a review of its present status in Britain Me %oFOLEY* Department of Biological Sciences, Institute of Environmental and Natural Sciences, University of Lancaster, Lancaster LAI 4YQ and M. S. PORTER* 5, West Avenue, Wigton Cumbria CA7 9LG ABSTRACT The present status of the sedge, Carex muricata subsp. muricata, critically endangered in Britain, is reviewed and, where relevant, the causes of its decline identified. A brief history of the plant at its few British colonies is also given. KEYWORDS: Prickly sedge, distribution, rarity, threats, conservation. INTRODUCTION It was only following the work of E. Nelmes (Nelmes 1947) that Carex muricata subsp. muricata was recognised as a British plant, having been included within C. muricata s.|. by earlier authors. Nelmes’ findings were based on herbarium studies in which he identified the plant in earlier collections made at four widely scattered British sites in Gloucestershire, Denbighshire, Yorkshire and Berwickshire. All other British material which he examined within C. muricata s.]. was assigned by him to subsp. lamprocarpa (as C. pairaei). This plant, whilst locally frequent throughout much of Britain, decreases rapidly moving eastwards in Europe where it is largely replaced by subsp. muricata. A taxonomic description of the two subspecies in Britain has been given by David & Kelcey (1985). Important character differences between them include the more robust habit, rigid, erect flowering stems, the much earlier flowering of subsp. muricata and, when immature, dark glumes which are much shorter than the matt green utricles giving the inflorescence a chequered appearance (Fig. 1). This contrast disappears as the fruit ripens, both the glumes and utricles becoming dark brown or even blackish. On the other hand, in subsp. lamprocarpa the glumes are initially pale yellow-green and almost the same length as the similarly coloured utricles, so that the chequered contrasting pattern of subsp. muricata is initially lacking and not apparent until maturity, when the glumes fade to white and the utricles darken. There are also ecological differences. In Britain, subsp. muricata is a plant of calcareous habitats, occurring especially over warm, sloping limestone scree, whereas subsp. /amprocarpa prefers an acidic substrate and is more a plant of the lowlands (David & Kelcey 1985). The present authors, familiar with subsp. muricata at its few known localities in northern England, and having recently become concerned about the apparent decline of the populations and the degeneration of the habitat at some localities, decided to investigate further the current status and ecology of the plant. This work was later expanded and incorporated into a country-wide project subsequently sponsored by English Nature. The original investigations were augmented by an assessment of the local distribution and management. Vegetative material was also collected for “Addresses for correspondence: e-mail m.foley @lancaster.ac.uk; m.porter@ wigton.cablenet.co.uk 280 M. J. Y. FOLEY AND M. S. PORTER subsequent genetic studies involving DNA sequencing from individual plants at the extant populations, and seed for storage in the Millennium Seed Bank at Kew. Further details relating to this survey and to the location of all known populations are given in a confidential report to English Nature (Foley & Porter 1999). KNOWN BRITISH LOCALITIES Nelmes’ critical work at K led to the initial recognition of subsp. muricata in Britain. In 1942 he first identified as such, a specimen in Druce’s herbarium which had been collected in 1900 on oolitic limestone at Woodchester, v.c. 34. Subsequently (Nelmes 1947), he identified three other collections as the same taxon from the top of a limestone hill near Wrexham, Denbs., v.c. 50 (1840, Bowman), from the castle grounds at Lauder, v.c. 81 (1878, Brotherston), and from limestone slopes and scree at Gordale, v.c. 64 (1934, Milne-Redhead & Sandwith; 1937, Lousley). However, his work appeared to generate little interest, until R. W. David and J. G. Kelcey made attempts to refind the plant at the original localities. Initially this was unsuccessful due to the difficulty in locating such a rare, and somewhat insignificant plant in large areas of apparently suitable habitat. However, in 1974 B. M. Mack found an unidentified Carex at Nympsfield, Gloucestershire which she sent to David (David & Kelcey 1975) and which was subsequently identified as subsp. muricata. Nympsfield is only a short distance from Woodchester, Druce’s original locality, and may be the same. In 1978, at Minera, near Wrexham, the plant was discovered by A. Newton. Despite claims that this was a rediscovery at the original 1840 locality (Anon 1981), there can be no certainty that this was So, as there is much suitable habitat elsewhere in this area. Just prior to this in 1977, F. J. Roberts had found a small population growing on wooded limestone pavement in upper Ribblesdale, v.c. 64. In the following year plants were seen at the same locality and their identity confirmed by R. W. David but there was some doubt as to whether this was Roberts’ original population (David 1979). At the previously known Yorkshire locality at Gordale, searches were initially unsuccessful, but subsp. muricata was eventually refound there in 1984 by P. Jepson. His locality is probably slightly different to that of the earlier collections of both Milne-Redhead & Sandwith and of Lousley. Up to the present, the original 1930s Gordale localities have still not been refound and may now be extinct as this is a very popular area with walkers and climbers who may have inadvertently eliminated the plant. Nevertheless, Gordale has a large area of suitable, often inaccessible habitat where the plant may perhaps still remain undetected. When D. Millward was surveying for the Flora of Wensleydale in 1987, she discovered a population of an unusual Carex in lower Swaledale. The plants were subsequently confirmed as subsp. muricata by R. W. David and occurred in a fairly open habitat at the base of limestone scree, the locality being some appreciable distance from other known sites. In 1999, S. J. Whild and A. Lockton discovered another locality near Nantmawr, Shropshire, v.c. 40, in a rather similar habitat to that at Swaledale and almost 30 km to the south of the Minera population. Of Nelmes’ original four localities, it is only at Lauder where the plant has not been refound at least within the general vicinity. However, here the habitat is not calcareous and appears to be quite unsuitable. It is possible that Brotherston’s specimen may have been mis-labelled or, perhaps more likely, may represent an accidental or even deliberate introduction. This will be discussed later. HISTORY AND PRESENT STATUS IN BRITAIN In addition to Nelmes’ original localities, at least three others have been found relatively recently in separate geographical areas, whilst the populations at Nympsfield and Minera may also be new finds. The recent history of all of these and a description of the present habitat and status of the populations are discussed below, largely in the sequence in which the populations were refound. The information is also summarised in Table 1. It should be borne in mind that the plants often occur in “clumps” so that it is difficult to decide whether one or several plants are present. Such “clumps” are considered here to be a single plant. ZO CAREX MURICATA SUBSP. MURICATA OHS JUSIOIFIP APYSTS Be PEG] Ul UMOUY yey US JUaTajyIp e Apquqgoid - Opg] Ul WeYyXerM Ivou UMOUY 5.5. OUIS JUdtayJIp ev A[qeqold - QOS] Ut Aqreou UMOUY ,. (,oAeU “SYOIMIOG ouou IIN JOU) JOUTXx9 (O)0 rl puepyed = puepyied Ayureur ‘vore o1ptoe 8/81 Ig Jopne’y] sossels sulidg (Ayiadoid Ag UMOISI9AO —uunIny §=§‘qd OW) UMOISI9AO SUTUIODIQ Sjur[d Jo —_,E_*9 ‘adoTs Jas JUO}SOUIT] “SyIOX'MA'N sulzeis = T'S'S'§ Joye (6D)+L L AYVUISIA oYeIPOUUT = payeyasoa APYST] “SuLIeJ-MS ——-LB6T $9 a[ePaeMg [jeurs [JeAg] ‘JuouraAed "SYIOX “AA-PIA SUTIO}IUOUL Ajauta.n xo asueyo JUO}SOUN] Poyejas9A APYST] ayepsofqqny asojo INN ynq 9]qQeIs (61D)Z 4 yuoredde opty ‘uado Ayirey JO sovyins Joddn LL61 +9 Joddq, AlOJDVJSNeS ISTMIOYIO ING uddo astMiay10 SUTIO}IUOU ‘UMOISIOAO jnq posseis o09-Sr odoys sursey "SYIOX “A-PIA weynsor TT S'S'S Apystys (S819 Vv ATIAvoy Joys “§ ‘99198 poweplfosuod daajs_ xxx P86T —-H9 s[ep1oy SUTIO}UOU eIOLy Jepnsal ‘oyeo (Adoues punois pue snumy Jo uorsnjoxa Poso]o F) Aq paindsqo 9a.19s o07-ST ‘IQAOO 901} pousyeoiy) ‘Adoued pasopo adoys Sutovj-¢ ‘9010s ‘auO)SOUNTT *squoqd Jo [eAOUII IIN A[O1OA0S (6)E +(0Z ‘papoom ATIAvoY ‘ISOO] IDAO GNIDS JOZPY ye Q/GT OS vIOUTP SULIO}IUOUL Je[nso1 ‘SUIZRIS popoom ISI] ‘sd youryxo pure UMOISIDAO oO€—-ST ‘adoy]s du0}soUTT] “SONOTD “AA Jo sultesyo IIN pournsoid (O)0 IZ JEYMOWIOS — JTJITOO IOAO puRTPOOM paresfo —s_&-_ € 6 Iv praysdurAy SUIALIY] = (S9OUDOSIIOTJUT oOb-O€ ‘adoys SULIOWUOW = 9ATOSOI ‘yeory] = pospuny [e.1dAas) IIIS DUOJSOUIT] PIJLIISIA ‘doyes Je[nso1 oimjeu yeooy juoredde ou QOL? - posueryoun Anysty ‘uado ‘sursey-¢ 6661 Or IMVUTJUR NY 666] Ul A][PUISLIO juowaseueU snqeys (SooUddSaIOTJUT pur) poisossng uonRAsosuoD yeosy sjuv[d SuLIOMOT] JO ‘ON yeyiqey JUIsaIg yeyiqey [VUISIIQ PUNO] ISIN] “o°A AVI[RIO0'T IVAHaHL GNV SA.LV.LS “LVLIAVH - NIVLIIAE NI SHILITVOOT VIVOINAW ‘dSANS VIVOIMAW XAYVO 31 ATAVL 282 M. 3.-Y¥; FOLEY AND M. S-PORTER. Full grid references for all known sites have been lodged with English Nature. 1. NYMPSFIELD (V.C. 34) It is not clear whether the population discovered here in June 1973 by B. M. Mack is the same as that known to Druce in 1900 (described as Woodchester) but is quite possibly so. It was recorded at Dingle Wood, Nympsfield, on a warm slope (hanger) over oolitic limestone, when 21 plants appeared after woodland clearance and re-planting. However, in 1976, it was described as hard to find but was still there in 1979. The population declined further as the site became overgrown so that there were only four plants in 1982. Fruitless searches were made thereafter and the site was cleared in 1986 in the hope that plants would regenerate. None emerged during the following spring and so subsp. muricata was considered extinct. Later that year (1987), three plants were introduced to the site (grown at Cambridge Botanic Gardens from seed originating from Dingle Wood) and whilst these persisted for a short while they rapidly declined in vigour and eventually died. A single, caged plant survived until 1989 but the unattended cage became choked with vegetation and the Carex succumbed. Material thought to originate from this site was grown on in S. C. Holland’s garden and two plants were transferred to that of M. Kitchen after her death. Further reintroductions within the area have been considered but have not yet been carried out. The authors feel that the provenance of such material should be checked to ensure that it is of Nympsfield origin. 2. MINERA, NEAR WREXHAM (V.C. 50) When discovered in 1978, the habitat comprised light scrub with c.20 plants of subsp. muricata. Shortly afterwards more were found close by and observations made soon afterwards referred to good numbers occurring on loose limestone scree with associates which included Teucrium scorodonia, a plant often indicative of stony habitats. In 1987 over 40 plants were reported but this was reduced to 27 in 1992. Since then, the site has declined rapidly, the original Corylus scrub developing into closed-canopy woodland. In consequence, litter and humus have accumulated and the original loose scree is now completely obscured and a ground flora is present in which Geranium robertianum and Sanicula europaea dominate. However, Hedera helix, which at one time was a problem, is no longer present. The Carex population is much reduced in size and vigour, the few plants surviving being in heavy shade and mainly vegetative. They also have to compete with the gradually increasing ground flora. Flowering spikes are few and utricles rarely seem to reach maturity. Typical associates of the Carex, such as Teucrium scorodonia, which were present in the original, more open habitat, have also been eliminated. When surveyed (1999) only three flowering plants were found, these possessing a total of eleven inflorescences between them, most of which were weak and depauperate. A few probable (but unconfirmed) seedlings were also present but seem unlikely to reach maturity unless conditions are improved by a drastic reduction of the tree cover. At the time of the survey it was also noted that cattle had broken through the protective fence and trampling had occurred. It is remarkable how rapidly this site has deteriorated and this illustrates the need for regular monitoring and management. 3. UPPER RIBBLESDALE (V.C. 64) This locality on partly shaded, vegetated limestone pavement appears to be unique in Britain, all other localities being, or having been, on sloping limestone scree. David (1979) questioned whether the population from which he confirmed the plant in 1978 was the same as that first discovered by Roberts in the previous year. However, his subsequent searches revealed only two additional isolated plants and it is ‘more-or-less certain that the two records were for the same population. Initially there were only four plants, two in each of two populations in close proximity. An additional plant was found by David some few hundred metres to the south but has not been seen since. Meanwhile, the original colony was decreasing. The northernmost population lost one plant in 1979 and the second by 1984 although an additional seedling was found situated between the two populations; this could not be detected in 1987. The other two plants (probably those first found in 1977) still survive. One of these is partially caged but still suffers from grazing by rabbits which often remove the flowering spikes. An additional seedling was found very close to these in 1999. The plants occur in an open area within this relatively heavily wooded site, closely CAREX MURICATA SUBSP. MURICATA 283 associated with such species as: Anthoxanthum odoratum, Dactylis glomerata, Geranium robertianum, G. lucidum, Veronica chamaedrys, Filipendula ulmaria, Saxifraga hypnoides, Silene dioica, Mercuralis perennis, Asplenium scolopendrium and A. trichomanes. Such associates correspond closely to those seen by David in 1978, suggesting that conditions here have changed little in the intervening period, in contrast to what appears to have happened at some other sites, and especially at Minera. The upper Ribblesdale locality is situated within the Ingleborough N.N.R. 4. GORDALE (V.C. 64) The history of the plant at this site is a little confusing as the complex topography makes it difficult to establish the precise whereabouts of early records. In 1985, the year after it had been refound by P. Jepson, only four plants were recorded in an area of 4 x 4 m, two of which were flowering poorly. These were on a steep, rocky apron at the base of a limestone cliff but were not thought to be of the same population as those recorded in the 1930s which were apparently above the waterfall. Subsequently, rumours circulated that the plants had gone from Gordale, but they were refound by Jepson in 1997. In 1999 the present authors confirmed a thriving population on the “apron”. Here there were at least 25 plants bearing a total of 128 flowering spikes, whilst a short distance to the south, on the sides of a steep buttress, a further eleven plants (57 flowering spikes) were found. The 45-60° sloping apron, whilst of an open, sunny aspect, was heavily grassed, mainly with Sesleria caerulea, Dactylis glomerata, Brachypodium sylvaticum and Festuca ovina, although the sedge still appeared to be flourishing there. Other close associates included Glechoma hederacea, Torilis japonica, Hypericum hirsutum and Centaurea nigra with a little Thymus, Teucrium scorodonia, Geranium robertianum and Campanula rotundifolia. On the nearby steep buttress there was less competing vegetation and the sedge grew in close proximity to plants typical of a more open habitat such as the last four mentioned above. On the “apron”, the grasses are apparently becoming more dominant since there is little evidence of grazing with only the occasional sheep roaming the area. There is, however, a great deal of other apparently suitable habitat at Gordale where subsp. muricata may occur, but most of it is difficult and even dangerous to explore. Past and present localities benefit from S.S.S.I. protection. 5. SWALEDALE (V.C. 65) Carex muricata subsp. muricata was first found here in 1987 near the foot of a S.W. facing, 30- 35° vegetated scree slope below steep limestone crags. Initially there were about 15 plants, seven of which had fruiting spikes, all occurring in an area of only a few square metres. The general habitat was, and still is, open and sunny with scattered Sambucus shrubs, but latterly invasive grasses strongly compete with the Carex with the result that, in 1999, the authors were able to find only seven plants. A little distance above these plants, the vegetation of the slope quickly changes and it becomes clothed in a dwarf, grazed, herb-rich community with Thymus, Pilosella officinarum, Leontodon hispidus and Linum catharticum the dominant species. Here a few small flowering plants of subsp. muricata are regenerating, apparently as a result of a reintroduction, despite the fact that at the time (1994) the existing population was not under threat. Unfortunately, it also appears that local plants were not used for this purpose and as a result the genetic make-up of this isolated population will ultimately be affected. The dwarf, herb-rich community described above would appear to be a more ideal habitat for subsp. muricata and was probably similar to that present when the original, currently overgrown, population was first discovered. An associated plant list given for 1990 appears to confirm this, as it includes such species of well grazed habitats as Thymus and Galium sterneri, both of which are now absent from the immediate vicinity of the native population. The increased competition here, probably resulting from a sharp reduction in grazing in an area now enclosed for the sedge’s protection, gives some cause for concern. This locality, which is situated on M.O.D. property, has S.S.S.I. protection. 6. NANTMAWR (V.C. 40) This population was identified during the Spring of 1999 during a routine survey of an existing local nature reserve. It had been noted as early as 1982 but had been wrongly referred to subsp. lamprocarpa. The plant is locally very abundant here on open, sunny, lightly vegetated, limestone scree with scattered Taxus and covers an area of approximately 20 x 30 m. The angle of the slope is c.30-40°. Plants occur both singly and in relatively large clumps in minimal competition. They 284 M. J. Y. FOLEY AND M. S. PORTER appear to be thriving and free flowering with abundant fruit set. It is very difficult to assess the number of plants present since many grow close together, but there may be up to 100 which produce several hundred inflorescences so making this the largest known British population. Typical close associates at this lightly vegetated habitat include Teucrium scorodonia, Campanula rotundifolia, Helianthemum, Thymus, Plantago lanceolata, Galium verum and Festuca ovina. Very close by, in a similar open habitat, is a small population of the locally rare Galeopsis angustifolia, a plant of loose stony substrates. 7. LAUDER (V.C. 81) An examination of the parkland habitat at, and in the vicinity of, Lauder Castle confirmed that there is no suitable, calcareous habitat present and that the 1878 record by Brotherston is either a result of an error in specimen labelling or, if genuine, represents a plant or plants deliberately introduced into the castle grounds. The latter would seem to be quite likely as other species, including sedges, which are outside their normal area of distribution in Britain, have been recorded at the castle (pers. comm. M. Braithwaite). The planting of aliens and exotics in estate woodlands was fashionable in the past and it is likely that from such a source has come the Lauder record. It should be considered only as a deliberate introduction here unless unequivocal confirmation can be established otherwise. CONCLUSIONS Examination of the plant at its few known British sites suggests that it thrives best on sunny, dry, limestone scree slopes in the presence of only minimum competition and shading. At the centre of its distribution (Scandinavia/continental Europe) it is able to tolerate a wider variety of soils as well as some shading (B. Jonsell, pers. comm.; B. G. Johannson, pers. comm.) but here in Britain, on the edge of its range it appears to be more exacting in its demands. Where ideal conditions occur such as at Nantmawr, robust clumps of plants exist which produce a large number of inflorescences, whose fruits ripen readily. Smaller, apparently younger, fruiting plants also occur on the open scree and on small ledges. Somewhat similar conditions occur at both Gordale and Swaledale, although there the habitat appears to be changing through invasion by coarser grasses. These populations should be examined annually in order to monitor for adverse change. In contrast, the Minera site has deteriorated rapidly in less than ten years and is now severely overgrown with Corylus woodland. In the area where the plants grow, sunlight is severely reduced and humus has accumulated over the scree so that a dense ground flora now competes with the sedge. In consequence, the few plants have become weak and appear unwilling to set much fruit. This site is in need of immediate attention. The other extant British locality is on limestone pavement at Upper Ribblesdale. This population has never been known to contain more than a few plants and is now reduced to just two flowering clumps together with a few seedlings. Rabbit grazing has been considered a problem elsewhere but is probably not the cause of loss here. Investigations into the plant’s autecology in this unique habit would be worthwhile. Reintroductions are currently being considered at Nympsfield and Upper Ribblesdale. At the former site, the authors feel that if the trees could be cleared from the original area it would then be suitable for the plant which might even regenerate naturally from dormant seed. At both sites, however, in order that genetic diversity is maintained, it is important that only seed or cultivated plants originating from the site in question is used when this is attempted. During this survey, seed from most of the extant sites was placed in the Kew Seed Bank, partly for this purpose. Leaf tissue for DNA analysis has also been collected. Unlike the commoner subsp. lamprocarpa, subsp. muricata is considered a very rare Red Data Book plant in Britain and is listed as critically endangered (Roberts 1999). However, its widely scattered, but very local, distribution (Gloucestershire to Yorkshire) suggests there may be undiscovered populations. Much apparently suitable ground occurs on the Craven limestones of Yorkshire, as well as in south Cumbria, Derbyshire, the Welsh borders and the Cotswolds and the plant may be overlooked. It may also wrongly recorded, perhaps as subsp. lamprocarpa or C. spicata, as this latter plant can also favour a similar habitat and grows with the Swaledale population. CAREX MURICATA SUBSP. MURICATA 285 Any plant of the Carex muricata aggregate found on warm, open limestone scree should be carefully examined in case it is this taxon. C. muricata subsp. muricata is best recognised by its relatively rigid, upright flowering spikes which appreciably exceed the leaves, and by the (initially) contrasting dark glumes and lighter utricles (see Introduction). On maturity the inflorescences may become decumbent and the fruiting head almost black. The utricles of subsp. muricata are both more flanged and smaller than in subsp. Jamprocarpa and C. spicata whilst the female spikelets are more rounded. To help in identification, typical plants and fruiting heads of both subspecies are shown in the accompanying drawing (Fig. 1). 4 Figure 1. Comparative sketches of Carex muricata subsp. muricata (above) and subsp. lamprocarpa (below): habit (x 1/5), inflorescences (x 2). A: Immature utricles and glumes (x 6) in June. B: Mature utricles and glumes (x 6) in August. 286 M. J. Y. FOLEY AND M. S. PORTER ACKNOWLEDGMENTS We wish to thank the relevant vice-county recorders for their help with information on the various sites and English Nature for allowing us to publish this paper. REFERENCES ANON (1981). Plant records. Watsonia 13: 342. DAVID, R. W. (1979). Another British locality for Carex muricata L. sensu stricto. Watsonia 12: 335. DAVID, R.W. & KELCEY, J. G. (1975). Carex muricata L. sensu Nelmes and Carex bullockiana Nelmes. Watsonia 10: 412-414. DAVID, R.W. & KELCEY, J. G. (1985). Biological Flora of the British Isles, 159. Carex muricata L. aggregate. Journal of ecology 73: 1021-1039. FOLEY, M. J. Y. & PORTER, M. S. (1999). A survey for Carex muricata subsp. muricata in Britain. Unpublished report to English Nature, Cumbria. NELMES, E. (1947). Two critical groups of British sedges. The Botanical Society and Exchange Club of the British Isles, Report for 1947, 13: 95-105. ROBERTS, F. J. (1999). Carex muricata L. subsp. muricata, in WIGGINTON, M. J. ed. British Red Data Book. 1. Vascular Plants, 3rd ed. J. N. C. C. Peterborough. (Accepted February 2000) Watsonia 23: 287—291 (2000) 287 Choice of surveyor is vital to the reliability of floristic change studies A. OREDSSON* Department of Systematic Botany, Lund University, O. Vallgatan 14-20, S-223 61 Lund, Sweden ABSTRACT The flora of an area of 46 km? of farmland and wooded hills surrounding the lake of Striern, Sweden was surveyed. Fifty-six different habitats were selected from topographical and geological maps, within which 214 plots were chosen by random sampling. The plots, mostly 10 m x 10 m, were investigated in 1972 and 1998. Each plot was recorded by walking through on parallel lines 1 m apart. Six of the best botanists from the region carried out the field work and their recording capacity was estimated; apart from one individual, variation between surveyors was negligible. Over the 26 years, the mean number of species recorded per plot fell from 32-5 to 30-2; the downward trend probably reflects a true decline in the frequency of species. KEYWORDS: Change in flora, different surveyors, Sweden. INTRODUCTION Ever since the times of Linnaeus, Swedish botanists have maintained a high level of knowledge of the native vascular plants in Sweden. The tradition of writing Floras of provinces is flourishing, often with many amateurs taking part in the field work. Floras change. Since World War I, a marked increase in the standard of living in Sweden has taken place which has had a massive impact on the natural environment, through air pollution and indirectly by changes in life style, including migration from the countryside (Oredsson 1990). Unfortunately there are few historical studies which have quantified the abundance of vascular plants sufficiently accurately to allow a statistically reliable assessment of change. Consequently, Red Lists of threatened species are rather more subjective than objective comparisons with the past. One Swedish opportunity to assess change against quantitative historic data was by repeating studies carried out by Hans Goransson as part of his quantitative survey of the flora related to pollen rain in Ostergotland in 1972 in which I took part. As I had already demonstrated considerable changes in northern Scania (the southern-most province of Sweden) after 25 years (Oredsson 1990), it was assumed that any changes which had taken place in Ostergétland would also be detectable. In this connection, a problem hitherto little discussed in Sweden has arisen, namely the influence of individual botanists on the results of the field work. This paper deals with a case in which six botanists were involved in recording the same habitats 26 years apart. The field work area covers 46 km’ of farmland and wooded hills surrounding the lake of Striern, in southern Ostergétland, about 200 km S.W. of Stockholm. METHODS The 1972 survey was carried out by quaternary biologist Hans Géransson (HG) and the author (AO), while the 1998 survey was carried out by me and four assiduous, proficient, amateur botanists from Ostergétland (proof that Sweden is still teeming with Linnean pupils!) - Janne Andersson (JA), Johan Bergstedt (JB), Dan Lindmark (DL) and Rolf Wedding (RW). Before the 1972 survey, the study area was stratified using 1:50,000 topographical and geological maps. 56 habitats were established, 44 of which combined information from both maps “Address for correspondence: e-mail: Alf.Oredsson @sysbot.lu.se 288 A. OREDSSON (e.g. forest on V4xjO granite, or open terrain on clay), and the remaining twelve from the topographical map (e.g. church, or public roads in open terrain). Within each habitat, 1-10 plots were chosen by random sampling depending on the size of the habitat. In 144 cases the general position of the plot was located using 100 m co-ordinates, in 70 cases using 50 m co-ordinates. In the field, the final position was then determined using further random sampling by assistants without specialist botanical knowledge, and thus the final positions differed slightly between the years, although the map co-ordinates were identical. 157 of the plots were 10 m x 10 m. On shorelines (22 plots) a 10 m x 20 m rectangle was surveyed, half in water, half on land. Along roads and water courses (22 plots) the length was 10 m, whilst the breadth varied. These 201 plots were marked out with a cord, and at least the terrestrial part of the plot was walked through along parallel lines 1 m apart. In the remaining 13 plots which included urban areas, the boundary of the garden, grounds or churchyard provided the outer edge of the plot. In each plot, all identifiable vascular plant species were recorded, no matter what stage in the life cycle (e.g. seedling or withered) or number of individuals. Combining both years, 543 species were found excluding cultivated species. RESULTS If all six botanists had a similar capacity to find almost every species in the plots, it would be possible to make a reliable comparison in the flora between 1972 and 1998. The results for the botanists were therefore investigated in detail. 1972 SURVEY Hans Goransson surveyed 131 plots once between the end of June and early August and 22 plots twice, the first time in May and the second in July or August (Table 1). He found an overall average of 33-1 species per plot, but plots surveyed once had an average of 32-6 species and plots surveyed twice an average of 36-0 species, though this was not statistically significant (The Mann- Whitney U-test; p=0-14). I surveyed 61 plots in early August, and found an average of 31-0 species per plot. To make allowances for the different habitats surveyed, which affect the number of species, only habitats investigated by both HG and AO were compared (viz. 107 plots in 21 habitats). For each habitat, the median number of species was calculated, and then the number of plots which fell above, equalling or below these medians summed (Table 2). Using these medians, the differences in number of species recorded between HG and AO are not statistically different (Table 3). It is concluded that HG and AO would have found approximately the same number of species if they had investigated the same plot independently in 1972. TABLE 1. TOTAL NUMBER OF RECORDS, NUMBER OF PLOTS AND AVERAGE NUMBER OF SPECIES PER PLOT IN THE STRIERN AREA SURVEYED IN 1972 AND 1998. 1972 1998 Surveyors Ga HGb HGS AOI total | AO2 DL JA JIB ~—s RW _ total No of records 792 4265 5057 1892 6949 | 1991 453, 2188 1290 537 6459 No of plots 22 131 153 61 214 63 18 68 50 15 214 Average no species/plot 36-0 32:6 ei 31-0 32:5 31-6 25:2 32:2 25:8 35-8 30-2 Surveyors in 1972: Hans Goransson (HGa), twice (12-28 May, 6 July to 5 August) or (HGb), once (29 June to 11 August); and Alf Oredsson (AO1) (5-11 August). Surveyors in 1998, 6-17 July: Alf Oredsson (AO2); Dan Lindmark (DL); Janne Andersson (JA); Johan Bergstedt (JB); and Rolf Wedding (RW). SURVEYORS AND FLORISTIC CHANGE STUDIES 289 TABLE 2. NUMBER OF PLOTS RECORDED BY DIFFERENT BOTANISTS IN 1972 AND 1998 WITH ABOVE, EQUALLING OR BELOW MEDIAN NUMBER OF SPECIES PER PLOT FOR THE SAME HABITATS. SURVEYORS AS IN TABLE 1. 1972 1998 HGa HGb HG AOI AO2 DL JA JB RW Above median 1 23 24 22 24 2 Di ily 8 Median 0 12 12 4 1 If 6 7 1 Below median 1 28 29 16 15 15 29 16 5 1998 SURVEY In 1998, the 214 plots were surveyed between 6-17 July by myself (AO2), Dan Lindmark (DL), Janne Andersson (JA), Johan Bergstedt (JB) and Rolf Wedding (RW). The average number of species recorded per plot varied considerably; DL and JB recorded the least, AO and JA the middle numbers and RW the largest numbers per plot (Table 1). Again, comparisons between surveyors need to take into account the different habitats which each surveyed (Table 2). Median species counts in 174 plots in 35 habitats were again investigated. JA and JB recorded almost as many plots above as below the median, AO2 and RW recorded slightly more plots above the median, and DL more below the median. The Fisher Exact Probability Test indicated no significant differences in efficiency between the surveyors, except for DL (Table 3). AO and RW had previous experience of field work of this kind, and in a given plot might have observed one or two species more than the other botanists, whereas DL came in late and so was less prepared for the task. COMPARISON OF 1972 SURVEY WITH 1998 SURVEY If HG recorded in 1972 with the same degree of accuracy as did jointly DL, JA, JB and RW in 1998, and if my capacity for recording species had not changed, my results from the habitat comparisons in both surveys should have been alike, and they were. The number of plots which fell above, equalling or below the medians in 1972 were 22—4—16; and in 1998 were 24—-1-15 (Table 2). Equalling values excluded, The Fisher Exact Probability Test gives p=0-17 (Table 3) which is not statistically significant. Overall I came closer to the ultimate goal of finding all species present in a plot than did the average colleague. As I recorded almost the same number of plots in each year, this should not have affected the comparison between years species by species. Between 1972 and 1998, the average number of species recorded fell from 32-5 to 30-2 species per plot, a reduction of 7%. Despite the fact that six different botanists were involved in the field work, the downward trend probably represents a true decline in diversity. Details of which species changed will be published in another paper (Oredsson, in prep.). TABLE 3. ESTIMATES OF DIFFERENT SURVEYORS’ RECORDING CAPACITY IN 1972, 1998 AND ONE SURVEYOR BOTH YEARS. DATA FROM TABLE 2 (EXCLUSIVE OF MEDIAN) TESTED BY THE FISHER EXACT PROBABILITY TEST (SIEGEL 1956). SURVEYORS AS IN TABLE 1. LOFZISENO9S AOQ2 0-005 JB 0-13 0-005 0-17 0-006 0-17 All figures are probability values 290 A. OREDSSON DISCUSSION DIFFERENCES BETWEEN SURVEYORS In 1987-1988, the B.S.B.I. Monitoring Scheme engaged over 1600 botanists in a sample survey of 10 km squares in Britain and Ireland, during which time 2660 taxa were recorded (Rich & Woodruff 1990). The problems encountered during a comparison with data from the Atlas of the British flora prompted the harsh conclusion that ‘the widespread occurrence of recording bias suggests that information about recorders and their behaviour should be collected and analysed as carefully as information about the organisms themselves’ (Rich & Woodruff 1992) Four tetrads (2 km x 2 km each) in West Sussex were investigated by 29 volunteers with a range of botanical abilities representative of many national and county Flora projects in July 1992 (Rich & Smith 1996). They worked in pairs recording vascular plant species for 2:5 hours in each square. Individuals were then rotated to different squares and paired with a different botanist. In all, 634 species were recorded. From this experiment with different surveyors, Rich & Smith learnt that different botanists were the key factor causing variation between the surveys. In light of these two examples, it was obviously important to recruit the very best botanists available for the 1998 survey of the Striern area, or the differences between surveyors might have masked any floristic change. The results demonstrate it is possible to even out the differences between surveyors provided they are of sufficient quality. This proved to be the case for everybody involved in this project. SIGNIFICANCE OF INTENSITY From April to October 1982, six competent field botanists independently recorded the vascular plants in three British woods, each about 30 ha (Kirby et al. 1986). Two different, methods were used; a walk along 3—3-5 km long routes and quadrat survey in squares of 200 m’. It was found that differences between surveyors or seasons were much smaller than differences in methods or intensities of survey. In the Striern surveys, the standard plot was walked through along ten parallel lines only 1 m apart so few, if any, vascular plant species should have eluded a skilled botanist’s observation. However, Goransson found on an average 3-4 more species in plots investigated twice than in plots investigated once, and although this is not statistically significant it may indicate there is still some variation due to the season. ADVANTAGE OF STRATIFIED RANDOM SAMPLING In August 1964, two professional botanists investigated the flora of the large peninsula of Lake Toisvesi in central Finland, which was divided into 50 1 km squares. Each botanist tried to find as many species as possible in every square. Altogether 362 species were recognised, eleven of which were only found by one surveyor, and 14 only by the other. Moreover, the two surveyors were shown to preferentially select different habitats to record which accounted for some differences between lists for the same squares (Kyt6vuori & Suominen 1967). In the Striern area, the stratified random sampling technique prevented the surveyors being drawn towards any favourite habitats. Stratified random sampling is also relatively efficient. In a survey of mountain forest types in the Swiss Alps, it was found that stratified random sampling provided an accurate picture of the small scale vegetation pattern at low sampling effort in areas sized 10-50 km? (Godickemeier er al. 1997). It is also used successfully for the British Institute of Terrestrial Ecology Countryside Survey (Barr et al. 1993). A quarter of a century ago, I pointed out the advantages of stratified random sampling for regional mapping of the vegetation (Oredsson 1974). Unfortunately, this method is still far from being generally accepted. Sticking to vague ‘walk-about’ strategies is a horrible waste of human capital, humiliating a large number of amateur botanists willing to spend so much of their spare time on botanical research. ACKNOWLEDGMENTS Thomas Karlsson and Jimmy Persson, at that time colleagues at the Institute of Systematic Botany, SURVEYORS AND FLORISTIC CHANGE STUDIES 291 Lund University, took part in the development of the method applied in the survey of the flora in the Striern area. They both raised valuable objections to premature ideas of mine. Bj6rn Marklén and Joachim Zethraeus (+) marked out the plots in the fields on 5—11 August 1972. In 1998, they were followed by Romia Ahmed, Gustav Jonsson, Stefan Jonsson, and David Rovan. It was a great pleasure to work together with all of them. Grants were gratefully received from the Lund Botanical Society and the King Carl XVI Gustaf’s 50th birthday foundation for Science, Technology and the Environment. REFERENCES BARR, C. J. et al. (1993). Countryside survey 1990. Department of the Environment, London. GODICKEMEIER, [., WILDI, O. & KIENAST, F. (1997). Sampling for vegetation survey: Some properties of a GIS-based stratification compared to other statistical sampling methods. Coenoses 12 (1): 43-50. KIRBY, K. J. et al. (1986). Seasonal and observer differences in vascular plant records from British woodlands. Journal of ecology 74: 123-131. KYTOVUORI, I. & SUOMINEN, J. (1967). The flora of Ikkalanniemi (commune of Virrat, Central Finland), studied independently by two persons. Acta Botanica Fennica 74. OREDSSON, A. (1974). A new method of mapping vegetation (In Swedish). Forskning och Framsteg 1974 (3): 5-7. OREDSSON, A. (1990). Changes in the flora of Matteréd, a parish in southern Sweden, between 1964 and 1989 (In Swedish, summary in English). Svensk Botanisk Tidskrift 84: 293-311. PERRING, F. H. & WALTERS, S. M., eds. (1962). Atlas of the British flora. Thomas Nelson and Sons, London. RICH, T. C. G. & SMITH, P. A. (1996). Botanical recording, distribution maps and species frequency. Watsonia 21: 155-167. RICH, T. C. G. & WOODRUFF, E. R. (1990). The B.S.B.I. Monitoring Scheme 1987-88. Chief Scientist’s Directorate report no. 1265. Nature Conservancy Council, Peterborough. RICH, T. C. G. & WOODRUFF, E. R. (1992). Recording bias in botanical surveys. Watsonia 19: 73-95. SIEGEL, S. (1956). Nonparametric statistics for the behavioural sciences. McGraw-Hill Kogakusha, Ltd, Tokyo. (Accepted January 2000) at. “ Watsonia 23: 293-297 (2000) 293 Recording and the declines of the Nationally Scarce plants Ajuga pyramidalis L. and Melampyrum sylvaticum L. T. C. G. RICH 67 Heol Uchaf, Rhiwbina, Cardiff CF 14 6SR and C. SYDES Scottish Natural Heritage, 2 Anderson Place, Edinburgh EH6 5NP ABSTRACT An assessment of the reasons for decline of the Nationally Scarce plants Ajuga pyramidalis and Melampyrum sylvaticum was carried out by revisiting pre-1970 sites listed in the Scarce Plants Database. 17% of the records were erroneous. Inadequate original information probably accounted for failure to refind another 25% of the records. Excluding these records, 25% of Ajuga and 33% of Melampyrum pre-1970 sites were refound. Loss of habitat has probably accounted for declines in 25% of Ajuga sites and 66% of Melampyrum sites. In 50% of the Ajuga sites suitable habitat was still present but plants were not refound. Real decline has occurred in both species, but the degree of change was masked by the problems of recording. Keywords: conservation, botanical records, habitat loss, Lamiaceae, Scrophulariaceae. INTRODUCTION Nationally Scarce species are those which occur in 16-100 hectads (10-km x 10-km squares) in Britain, and are of concern to conservationists as the second rarest group of species in the British flora after the Red List species. Despite protection measures, local populations of rare and scarce species are still liable to extinction, but the greatest loss of biodiversity has occurred, and appears still to be occurring, amongst common but evidently vulnerable taxa (Sydes 1997). Between 1990 and 1993 a review of the Nationally Scarce vascular plant species was carried out (the Scarce Plants Project), with a main aim of creating a computer database (Stewart et al. 1994). The resulting Scarce Plants Database records were used to revise the list of Nationally Scarce species, update distribution maps and assess changes in the frequency of species, and the results were presented in Scarce Plants in Britain (Stewart et al. 1994). The Scarce Plants review indicated that a number of species are declining rapidly, as judged either from the high proportion of pre-1970 records shown on some maps, or the high percentage of the post-1970 records compared to the records on the maps in the Atlas of the British flora (Perring & Walters 1962). For instance, species such as Ajuga pyramidalis L., Betula nana L., Carex maritima Gunnerus, Melampyrum sylvaticum L., Mertensia maritima (L.) Gray, Meum athamanticum Jacq. and Saxifraga nivalis L. are species which are fairly widespread in Scotland but have significant numbers of pre-1970 records in Stewart et al. (1994). The causes of these declines are not always clear, especially where they occur in the remoter parts of Scotland, and might be attributable to natural metapopulation dynamics, under-recording, habitat change, or combinations of the three. It is valuable to have some estimate of the relative importance of these factors when assessing priorities for conservation, so as to concentrate on species whose habitats are declining and not waste resources on species simply under-recorded. To investigate the reasons for the large apparent declines in these seven species, in 1995 Scottish Natural Heritage (S.N.H.) commissioned field work to revisit pre-1970 sites and assess possible reasons for loss of sites. Of these species, the changes in Ajuga pyramidalis and Melampyrum 294 T. C.G. RICH AND C. SYDES sylvaticum were quantified in the most detail, including searches for historical herbarium and literature records to provide additional background information. They provide good examples of the balance between decline and under-recording of species which occur in remote areas (e.g. Ajuga) or are difficult to identify (e.g. Melampyrum). The aim of this paper is to summarise the results of the work; full details are given in Rich et al. (1996). METHODS For both species, ten pre-1970 sites with known 1-km square or 6-figure grid references were selected from the Scarce Plants Database to give a wide geographical spread in Scotland. Sites with known 1-km squares or 6-figure grid references were selected to maximise the chances of refinding the original site as many old records are vague and may not even be localised to a hectad. Ajuga pyramidalis is widespread on the Inner and Outer Hebrides from which recent data are particularly lacking, but these islands had to be excluded on the grounds of cost. Remote upland sites for M. sylvaticum were also excluded for safety reasons. As there are few detailed pre-1970 records for Melampyrum, a number of sites had to be taken as ‘best guess’ 1-km squares (e.g. ‘Falls of Clyde’ was taken to be NS/88.41). In addition, recorders visited three known post-1970 sites to familiarise themselves with the species and their habitats, selected from personal knowledge or where there were 6-figure references in the Scarce Plants Database. A few additional sites were also investigated for both species. The sites were visited during late June and July 1995, and the full 1-km square searched as far as was practicable, concentrating on the most appropriate habitats. A wider area was searched where the previous record was not clearly defined. Where the plants could not be refound, the field workers assessed whether the plant was likely to have gone or not from the presence or absence of suitable habitat. Most Melampyrum sites were surveyed by R. FitzGerald, often with the help of the B.S.B.I. v.c. Recorders; the Ben Hiel site was surveyed by P. A. Smith. G. M. Kay surveyed Ajuga pyramidalis sites. The areas searched and locations of extant populations were indicated on 1:25,000 maps. Where populations were not refound, notes were made on the likely causes for loss or lack of re-discovery, and further enquiries were made about the locality from the original recorder, if they could be traced. Historical data from specimens in BM, BRISTM, E, GL, GLAM, K, LIV, MANCH and OXF (abbreviations follow Kent & Allen 1984) were abstracted. A literature search of records in the floras and journals held in the libraries at Kew and the Natural History Museum was carried out, with additional references traced in BSBI Abstracts and Simpson (1960). RESULTS The results of the field survey are given in Tables 1 and 2, and are summarised in Table 3. Erroneous records in the Scarce Plants Database, where either the species or grid reference was wrong, are believed to be responsible for failure to find one Ajuga (9%) and three Melampyrum (23%) pre-1970 sites (17% overall). In addition, insufficiently detailed original information was thought to account for lack of discovery in another two of the Ajuga (18%) and four Melampyrum (30%) pre-1970 sites (25% overall). Excluding these sites, plants were refound in two of the pre-1970 Ajuga sites (25%), and two of the pre-1970 Melampyrum (33%) sites (29% overall). Habitat change probably accounted for losses at two Ajuga (22%) and four Melampyrum (50%) pre-1970 sites (35% overall). One of the Ajuga sites appeared still suitable but plants could not be found after an extensive search indicating they may have disappeared perhaps by chance demographic events, but some were found in a new site nearby. Plants were not refound in the other pre-1970 sites either because they have disappeared or because they were over-looked. One post-1970 site with 6-figure grid reference was not refound for one of these same reasons. 166 (41%) historical records additional to the Scarce Plants Database were found for both species (Table 4), which provided valuable additional information about distribution and ecology, especially for M. sylvaticum where a number of specimens had been wrongly named (data not presented; see Rich 1997; Rich et al. 1998; 1999). DECLINES OF AJUGA PYRAMIDALIS AND MELAMPYRUM SYLVATICUM 295 TABLE 1. RESULTS OF 1995 FIELD SURVEY FOR AJUGA PYRAMIDALIS Site Loch Call an Uidhean Torr Achilty SW of Clais nan Cruineachd Traigh an T-Straithain Creag na Faollinn Aird a’ Chuilinn Rhigolter/An Fhireach Sanna Bay Creag Clais nan Cruineachd Clais nan Cruineachd Cnoc na Stri Deadh Choimhead Clashnessie Loch Buine Moire Thurso Sanna Bay Grid reference Last record Refound in 1995? Comments 6-figure 6-figure 6-figure 1 km 1 km 1 km 1km 6-figure 6-figure 1 km 6-figure 1 km 1992 1992 199] 1988 1967 1964 1964 1964 1961 1961 1961 1961 1954 1952 1916 New site Wrong grid reference in database Not refound, possibly still present Wrong grid reference in database Not refound, possibly still present Not refound, possibly gone Not refound Possibly a different site Inadequate information, later found to be inaccessible site on cliff Inadequate information, later found to be inaccessible site on cliff Two sites present Possible gorse invasion caused loss Not refound, possible gorse invasion or agricultural improvement Near old record TABLE 2. RESULTS OF 1995 FIELD SURVEY FOR MELAMPYRUM SYLVATICUM Site Birks of Aberfeldy Allt Doe Glen Buck Loch Ossian Linn of Corriemulzie Ballater Bridge Reekie Linn Struan, River Garry Roy Bridge Blackhall Woods Den of Airlee Glen of Reichip Ben Hiel - Ben Loyal Falls of Frenich Falls of Clyde Rowardennan 6-figure 6-figure 6-figure 1 km 1 km 6-figure 1 km? 6-figure 1 km? 1 km 1 km 1 km? 6-figure 1 km? 1 km? 1 km 1992 1984 LOT 1958 1957 1955 1950 1949 1896 1871 1846 1821 Undated Undated Grid reference Last record Refound in 1995? Comments Confused information, actual site not visited Biggest population found Grazing damage Now densely shaded Dense woodland and agricultural runoff may have caused loss Forestry caused habitat change Possibly inadequate information Forestry caused habitat change, but imprecise original record Imprecise original site, now densely shaded Possibly inadequate information, site now densely shaded Probable error for M. pratense Possibly inadequate information Error for M. pratense Error for M. pratense 296 T. C. G. RICH AND C. SYDES TABLE 3. CLASSIFICATION OF RESULTS INTO CATEGORIES RELATED TO RECORDING Melampyrum sylvaticum Ajuga pyramidalis Pre-1970 sites Sites refound in 1995 2 Sites not refound but possibly still present 2 Probable loss due to habitat change s) Incorrect/inadequate information 3 Erroneous record 3 Post-1970 sites Sites refound in 1995 2 Sites not refound 0 1 0 mere NO NbN Incorrect information Re ee LL New sites Total sites investigated 16 16 TABLE 4. NUMBERS OF RECORDS IN SCARCE PLANTS DATABASE AND ADDITIONAL RECORDS TRACED AFTER CORRECTION FOR DUPLICATES AND ERRORS Species Scarce Plants Database Additional records traced Total records Ajuga pyramidalis Total no. records 330 Bil 361 Pre-1970 hectads 37 2 39 Post-1970 hectads 53 0 53 Melampyrum sylvaticum Total no. records 70 135 205 Pre-1970 hectads 22 26 48 Post-1970 hectads pH | i 22 DISCUSSION Assessing the causes of the declines in Ajuga and Melampyrum was complicated by the problems of recording and lack of detailed information. 13% of the records in the Scarce Plants Database were found to be erroneous. Inadequate original information probably accounted for failure to refind another 17% of the records. Where we were able to trace the original recorders, they kindly provided valuable information about the records which would have been useful in the field, such as occurrence on cliffs or the accuracy of the grid references. Similarly, the work on historical records in herbaria and the literature provided valuable information. It is recommended for similar projects in the future that this original information be obtained prior to commencing field work. Ajuga pyramidalis and M. sylvaticum were refound in 29% of the pre-1970 1-km squares overall (Table 3). Suitable habitat was still present in a further 29% of the squares overall, and we may have over-looked the species in some of these sites. Searching the pre-1970 sites thoroughly is very difficult, especially when they are as large an area as a 1-km square, and both species can be difficult to find. A. pyramidalis tends to occur in very small, scattered populations, and localities on cliffs are difficult to search adequately or safely. M. sylvaticum is also usually found in small, discrete populations, often in steep, wooded ravines where access is difficult. It is always difficult to be certain that a species has gone from a site. Nevertheless, ecological theory suggests that populations can be expected to become extinct on some sites for a variety of reasons including natural metapopulation dynamics. Habitat change (e.g. scrub invasion, forestry) has probably accounted for loss of 22% of the Ajuga pre-1970 1-km squares (Table 1), and 50% of the M. sylvaticum squares in eastern Scotland (Table 2). Further discussion of the causes of the declines in these species are given in Rich ef al. (1999) and Rich et al. (1998) respectively. It is concluded that both species have declined, but the true extent of decline was masked by the problems of recording. DECLINES OF AJUGA PYRAMIDALIS AND MELAMPYRUM SYLVATICUM 297 The data recorded for rare species have become more detailed in recent years. The range of species recorded in this way should be broadened and the quality of information enhanced so that conservation resources do not have to be diverted into checking data when the Scarce Plants in turn become rare. The resource of historic records in Britain’s herbaria is under-utilised and could greatly extend our knowledge of past change. ACKNOWLEDGMENTS We would like to thank those who gave access to land or helped with the field work or information: Penny Angold, Elaine Bullard, Kathy Fallowfield, Ro FitzGerald, Sam Hallet, Barbara Hogarth, Graeme Kay, Peter Macpherson, Douglas McKean, Henry Noltie, Fred Rumsey, Alex Scott, Ros Smith, Edna Stewart, Neale Taylor, Agnes Walker, Keith Watson, Phyllis White and Peter Wortham. We would like to thank the Keepers of BM, BRISTM, E, GL, GLAM, K, LIV, MANCH and OXF for access to the herbaria and libraries, and an anonymous referee for suggesting improvements to the manuscript. The work was funded by Scottish Natural Heritage, contract SNH/001G/95 IBB. REFERENCES KENT, D. H. & ALLEN, D. E. (1984). British and Irish herbaria. Botanical Society of the British Isles, London. PERRING, F. H. & WALTERS, S. M. (1962). Atlas of the British flora. Thomas Nelson, London. RICH, T. C. G. (1997). Using botanical records to interpret changes in frequency of British plants. The Biology Curator 10: 8-12. RICH, T. C. G., FITZGERALD, R. & KAY, G. M. (1996). Review and survey of scarce vascular plants. 3 volumes. Unpublished contract report to Scottish Natural Heritage. 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. RICH, T. C. G., KAy, G. M., & SYDEs, C. (1999). Distribution and ecology of Pyramidal Bugle (Ajuga pyramidalis L.; Lamiaceae) in the British Isles. Botanical journal of Scotland 51: 181-193. SIMPSON, N. D. (1960). A bibliographical index of the British flora. Privately published. STEWART, A., PRESTON, C. D. & PEARMAN, D. A., eds. (1994). Scarce plants in Britain. Joint Nature Conservation Committee, Peterborough. SYDES, C. (1997). Vascular plant biodiversity in Scotland, in FLEMING, U. V. & USHER, M. B., eds. Biodiversity in Scotland: status, trends and initiatives, pp. 89-104. The Stationery Office, Edinburgh. (Accepted January 2000) y +e Watsonia 23: 299-303 (2000) 299 Dactylorhiza incarnata (L.) So6 subsp. ochroleuca (Wiistnei ex Boll) P. F. Hunt and Summerh. (Orchidaceae): A comparison of British and European plants Mid: Y. FOLEY. Department of Biological Sciences, Institute of Environmental and Natural Science, University of Lancaster, Lancaster LAI 4YQ ABSTRACT A brief summary is given of the occurrence of Dactylohiza incarnata subsp. ochroleuca in Britain and continental Europe and examples of the very few remaining native British plants have been compared to those at three European localities. Although the sample size at the British locality was inevitably extremely small, it is tentatively concluded that plants at all the localities are conspecific. KEYWORDS: morphometrics, diagnostic characters. INTRODUCTION Known in Britain under the name Dactylorhiza incarnata subsp. ochroleuca, this plant has been placed at differing taxonomic levels by many authors. Relatively rare and widely scattered from central Europe to Scandinavia, the only genuine British records are from eastern England (East Anglia) where it now appears to be reduced to one small population. As the subspecific epithet implies, the flower colour is pale yellow and both in Britain and on the continent, it has been over- recorded through confusion with yellow flowered, anthocyanin-lacking forms of other subspecies of D. incarnata, especially subsp. pulchella. Flower colour alone, therefore, is not sufficient to distinguish it and previous authors (e.g. Pugsley 1939; Nannfeldt 1944; Heslop-Harrison 1956; Rajchel 1964; Lundqvist 1967; Bateman & Denholm 1983, 1985) have variously used a combination of characters in its taxonomic delimitation. These include its relatively robust habit, tall, broad stem, large leaves and bracts, and relatively large, pale yellow, unmarked and trilobed labella, frequently having notched lateral lobes. It is a plant with a precise habitat requirement, growing in the drier parts of rich fens where it occupies a distinct ecological niche. As Orchis incarnata var. ochroleuca, it was first described by Boll (1860) from plants seen by Wiistnei at Sternberg (in Mecklenburg), Germany in 1854. From his record (Wiistnei 1854), it is not clear in exactly what type of habitat they grew and, because of this, reservations have been expressed (Bateman & Denholm 1985) regarding the true identity of the plants upon which the name ochroleuca is based. However, an examination of preserved specimens collected by Wiistnei (Mecklenburg, Wiistnei, s.d. (W!)), similar to those on which Boll based his original description, confirms that at least one of them is referable to this subspecies. This is the central specimen of the three mature specimens on the sheet and bears the label “Orchis incarnata L. var. floribus ochroleucis......In Mecklenburg. K Wiistnei”’. The first definite British record was made in 1935 from the upper Ouse/Waveney valley area (Blo Norton Fen, 23.vi.1935, Lousley (BM!)) and shortly afterwards it was also found at various nearby fens within the same general area of East Anglia. Previous to this, however, the Stephensons had tentatively recorded it from both Kidwelly, South Wales as well as from an unlocalised site in East Anglia (Stephenson & Stephenson 1923). Although no preserved specimens collected by the Stephensons from the latter area have been traced and the Kidwelly locality has not been re-found, a Stephensons’ specimen, labelled ‘“(b)? v. ochroleuca’, does exist (Kidwelly, 27.vi.1919 (TOR!)). In general habit this does show some affinity to less robust plants of subsp. ochroleuca, but its labella are line-marked and almost entire and these are characters not found in that plant. So, other than for the few known East Anglian localities, the plant has not been 300 M. J. Y FOLEY confirmed elsewhere in Britain although erroneous records proliferate due to confusion with similar coloured forms of other subspecies of D. incarnata. Mainly due to habitat drainage, subsp. ochroleuca is now a highly endangered British plant. There appear to have been no recent records at Blo Norton/Thelnetham and it may now be extinct there. Further east at Redgrave there were reports of a few plants until the late 1980s but none have been seen recently, although a single specimen was recorded at Market Weston Fen in June 1995 (M. Sanford, pers. comm., 2000). At Chippenham, perhaps now the only extant British site, there were as many as 50 flowering plants in the early 1980s but now, even in a good year, there are rarely more than five. Subsp. ochroleuca is also known from several parts of continental Europe, especially Germany (Lubs 1968; Fiiller 1983), Poland (Rajchel 1964) and the central Alps (Hegi 1939) and reaches as far north as southern Sweden (Nannfeldt 1944; Wiefelspiitz 1976) and Estonia (Kuusk 1991) and eastwards into Russia (L. Averyanov, pers. comm., 1990). Pale-yellow flowered plants from species-rich fens in the Picardy area of north-eastern France have sometimes been recorded as subsp. ochroleuca but these appear referable to pigment-lacking variants of the type or of subsp. pulchella. Further evidence for this was gained during a visit to the area in 1991. A MORPHOMETRIC COMPARISON OF BRITISH AND EUROPEAN PLANTS Whether the British plant is the same as that of continental Europe has sometimes been questioned (e.g. Stephenson & Stephenson 1923; Landwehr 1977). In order to examine this, the opportunity has been taken to compare the morphology of the few remaining examples of the only apparently extant British population with others occurring at localities in Scandinavia and mainland Europe. Due to its great scarcity, however, the number sampled in Britain was extremely small and, because of this, any conclusions drawn can only be tentative. To carry out this comparison, plants from three European populations - Storsund, Gotland (Sweden), Viidumae, Saaremaa (Estonia) and Murnau, Bayern (Germany), all apparently referable to subsp. ochroleuca, were morphologically compared to those at Chippenham (England) for those characters considered most useful in separating D. incarnata subsp. ochroleuca from closely allied taxa. Under precise conditions and in a similar manner to the method of Bateman & Denholm (1985), data were obtained in a non-destructive manner for the following characters: 1. plant TABLE 1. MORPHOMETRIC COMPARISON OF BRITISH AND EUROPEAN PLANTS OF DACTYLORHIZA INCARNATA SUBSP. OCHROLEUCA - POPULATION MEANS AND STANDARD ERRORS Chippenham Storsund, Gotland Viidumée, Saaremaa Murnau, Bayern (England) (Sweden) Island (Estonia) (Germany) No. of plants measured (n) =) 10 10 13 1. Plant height 338-3 + 23-8 235-1 + 20-3 321-8 + 19-1 291:5+7-9 2. Inflorescence length 42-3 +49 47-0 + 2-9 62:3 + 3-9 59-1 + 2:8 3. Stem width near base 6-1+0-6 6-8 + 0:5 7:54 0-4 5:7 40:3 4. Stem width below inflorescence 3-8+0-4 4:7+0-4 4:9 + 0-6 4-4+()-2 5. Longest leaf (length) LS 955 IZ7-1 2 7-7 117-6 + 6:3 108-5 + 3-7 6. Longest leaf (max. width) 17-8 + 0-4 220) & be] 22-4+ 1-0 22:9 + 1-1 7. Bract length 20:7:4'0:5* 33-9 + 2-8 29-4 + 1-3 26-1 + 1:0 8. Bract width 4-8 + 0-3* 6-8 + 0-6 6:5 + 0-3 4:5+0-2 9. Labellum width 6:5 +-0-1* 8-1 + 0-4 8:2 + 0:3 6-3 + 0-3 10. Labellum length 5-9 + 0-2* 6-4+0-1 6-4 + 0-2 6:3+0-1 11. Trilobed character 1-0 + 0-0* 0-9+0-1 0-64 0-1 0-7 + 0-1 12. Lateral lobe notching 1-0 + 0-0* 1:0 + 0-0 0:7 + 0-1 0-8 +0-1 (all measurements in mm) P= DACTYLORHIZA INCARNATA SUBSP. OCHROLEUCA 301 height, 2. inflorescence length, 3. stem width near base, 4. stem width below the inflorescence, 5. length of longest leaf, 6. maximum width of longest leaf, 7. length of lowest floral bract, 8. maximum width of lowest bract, 9. maximum width of labellum, 10. maximum length of labellum, 11. presence of distinct labellum sinuses (i.e. a measure of the degree of the trilobed character), 12. presence of notching on lateral lobes of labellum. The last two characters listed (11, 12) were scored as: 0 = character absent; 0-5 = present, but not highly developed; 1 = very noticeably developed whilst, in all cases, the flowers measured were those towards the base of the inflorescence. In addition all plants were checked for the presence/absence of line or dot markings on the labella and for leaf spotting. Wherever possible at least ten plants were assessed, chosen in as random a manner as possible so as to avoid selection bias. Unfortunately in Britain, where the population is now very small (five plants in that year), random selection could not be applied; furthermore, flowers of two of these could not be fully assessed due to partial damage, possibly by vermin. At all the localities the plants formed homogenous populations and were not interspersed with other subspecies of D. incarnata or different colour forms. At the British locality and at Storsund, populations were measured by the author. For the Estonian and German plants, vegetative characters were, under rigorous guide-lines, respectively measured by V. Kuusk (Tartu) and H. R. Reinhard (Zurich), whilst the floral characters were assessed by the author on material supplied on the same plants. For reference purposes, the same data (characters 1-12) were also obtained for a population of typical D. incarnata subsp. incarnata of normal flower colour (Hodbarrow, Cumbria, England). All these data were subjected to Principal Components Analysis (PCA) using the Clustan 4 computer programme (Wishart 1987) with the results shown in Figure 1. By this technique, each specimen measured is located in a multidimensional array, in which the number of dimensions 1s equal to the number of characters measured, and where the most similar specimens are placed closest together. The axes of greatest variation are extracted from the multidimensional space, thereby simplifying it to a few (usually two or three) dimensions and allowing the location of each specimen to be visualised. The respective mean and standard errors were calculated for each population (Table 1). +5 +3 +1 5 -3 -1 +1 +3 +5 FIGURE 1. PC1 : PC2 for individual plants of four populations of Dactylorhiza incarnata subsp. ochroleuca © Chippenham, Cambs. (England), LJ Vidumiae (Estonia), O Murnau (Germany), ® Storsund, Gotland (Sweden), compared to one of typical D. incarnata subsp. incarnata @ Hodbarrow, Cumbria (England). 302 Md. YeROLEY DISCUSSION In Fig. 1 the variance accounted for by the PC1 axis is 47-3%, that for the PC2 axis is 15-1%. In each case, characters making a significant contribution towards the variance include plant height and stem, inflorescence and labellum dimensions. By examining the plotted values in Fig. 1, it can be seen that the British plants lie especially close to those from Germany and plants of both these populations appreciably overlap with those from Sweden and Estonia. There is, therefore, no overall separation between the four populations on the PC1 axis but all are well separated from the control population of normal D. incarnata subsp. incarnata on this same axis. On the secondary axis (PC2) there is again broad overlap between the four subsp. ochroleuca populations with a slight shift and a greater spread for those from Sweden and Estonia. For the characters measured, therefore, the British plants lie morphologically very close to those from the three European populations. The character means of all four populations (Table 1) are also closely comparable. Perhaps surprisingly, the mean plant height was not as great as is often quoted. Higher values for labellum width have also been recorded (Bateman & Denholm 1985) but the British and German plants were similar in this respect. A fully trilobed labellum is also apparently not mandatory for every plant within a population but, despite this, it was most frequently present. Distinctly notched lateral lobes were also not evident in every case and it is possible that the presence of both of these characters may vary between the flowers on a single plant. However, most of the characters used to separate subsp. ochroleuca from closely-related taxa were found to occur in all of the plants sampled at the four populations. All of them were yellow-flowered and none had line-marked labella or spotted leaves. On the island of Oland (Sweden), plants of similar habit and with flowers of a deep violet coloration with unmarked, dark-centred labella, and of similar shape to subsp. ochroleuca occur, very occasionally, as isolated individuals amongst normal subsp. ochroleuca (Lundqvist 1967). The existence of a saturated reddish purple-based colour form in subsp. ochroleuca had been anticipated by Nannfeldt (1944) who predicted a coloration comparable to D. sambucina (L.) So6 rather than that typical of subsp. incarnata. So far, these plants appear to be known from only two localities to the east of Farjestaden and have been observed only very intermittently. According to Lundqvist (pers. comm.) and the colour illustration in Mossberg & Lundqvist (1994), they appear to be otherwise indistinguishable from normal subsp. ochroleuca. This suggests the presence of colour dimorphism in subsp. ochroleuca in a similar manner to that within other Dactylorhiza taxa (e.g. D. sambucina and D. sulphurea (Link) Franco) but with a very much smaller incidence of the violet form. SUMMARY Most populations of this quite rare plant are inevitably rather small. The three sampled in continental Europe contained only 20-50 plants whilst at Chippenham there were only five available. Admittedly, and especially for the latter, this is a very small number on which to base any conclusions and those given for the British plant can only be tentative. Nevertheless, for the diagnostic characters measured, the Chippenham plants were found to be morphologically very similar to those from the three European populations, and all of them could be defined by the characters attributable to D. incarnata subsp. ochroleuca. However, additional confirmation via molecular data would be useful in supporting this conclusion. ACKOWLEDGMENTS I am very grateful to H. R. Reinhard and V. Kuusk for their cooperation and help with the morphological measurements of their local populations; to M. S. Porter for assistance in the field and with translations; to L. Farrell and M. Wright for permission to examine the Chippenham plants; to the curators of the Natural History Museum (BM), Torquay (TOR) and Vienna (W) herbaria for allowing me to examine their specimens, and to D. M. Turner Ettlinger, R. Lidberg, A. Lundqvist, W. Wiefelspiitz and H. Wittzell for relevant information. DACTYLORHIZA INCARNATA SUBSP. OCHROLEUCA 303 REFERENCES BATEMAN, R. M. & DENHOLM, I. (1983). Dactylorhiza incarnata (L.) So6 subsp. ochroleuca (Boll) P. F. Hunt & Summerhayes. Watsonia 14: 410-411. BATEMAN, R. M. & DENHOLM, I. (1985). A reappraisal of the British & Irish dactylorchids, 2. The diploid marsh-orchids. Watsonia 15: 311-355. BOLL, E. (1860). Flora von Mecklenburg. Archiv des Verein der Freunde der Naturgeschichte in Mecklenburg 14: 307. FULLER, F. (1983). Die Gattungen Orchis und Dactylorhiza. Orchideen Mitteleuropas 3: 92-93. Die Neue Brehm-Biicherei, Wittenberg-Lutherstadt. HEGI, G. (1939). Illustrierte Flora von Mittel-Europa 2. Carl Hanser, Miinchen. HESLOP-HARRISON, J. (1956). Some observations on Dactylorchis incarnata (L.) Vermin. in the British Isles. Proceedings of the Linnean Society of London 166: 51-82. Kuusk, V. (1991). Dactylorhiza incarnata subsp. ochroleuca Eestissa. Orkidealehti 4: 98-99. LANDWEHR, J. (1977). Wilde orchideeén von Europa 1. Vereniging tot Behoud van Natuurmonumentem in Nederland, Amsterdam Luss, E. (1968). Vorkommen von Dactylorchis incarnata, var. ochroleuca im Kreisgebiet Neustrelitz - 1967. Naturschutzarbeit in Mecklenburg 11(1): 33-34. LUNDQVIST, A. (1967). Orchideen Olands. Svensk Botanisk Tidskrift 61: 463-493. MOSSBERG, B. & LUNDQVIST, A. (1994). Olandska angsnycklar. Svensk Botanisk Tidskrift 88: 85. NANNFELDT, J. A. (1944). Orchis strictifolia Opiz var. ochroleuca (Boll) Hyl., en fodrbissed svensk kalkkarrvaxt. Svedburg 1944: 601-622. PUGSLEY, H. W. (1939). Recent work on dactylorchids. Journal of botany, London 77: 50-56. RAJCHEL, R. (1964). Orchis incarnata L. subsp. ochroleuca (Wiistnei) O. Schwarz in Poland. Fragmenta Floristica et Geobotanica 10: 193-197. STEPHENSON, T. & STEPHENSON, T. A. (1923). The British forms of Orchis incarnata. Journal of botany, London 61: 273-278. WIEFELSPUTZ, W. (1976). Dactylorhiza incarnata subsp. ochroleuca (Boll) Hunt & Summerh. auf Gotland? Jahresbericht Naturwissenschaft Verein Wuppertal 29: 103-105. WISHART, D. (1987). Clustan user manual, 4th ed. Computing Laboratory, University of St. Andrews. WUSTNEI, K. (1854) in Koch, F., Bericht tiber die am 9 Juni unternommene Excursion nach der Umgegend von Sternberg. Archiv des Verein der Freunde der Naturgeschichte in Mecklenburg 8: 92-97. (Accepted January 2000) at Watsonia 23: 305-310 (2000) 305 Conservation of Britain’s biodiversity: Hieracitum cambricum (Asteraceae), Welsh Hawkweed ec. Ga RICH= Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP ABSTRACT Hieracium cambricum is a rare endemic hawkweed known from only three sites in Wales. It is a distinct, polycarpic perennial which flowers mainly May—June, and regenerates readily from wind-dispersed seed. It is probably a calcicole. Surveys of its sites were carried out in 1998. No plants were found at Graig-fawr (v.c. 41) where it may be extinct due to a rock fall. On the Great Orme (v.c. 49), where it is well known, 75 plants were found in a c. 25% sample of sites suggesting a population of c. 300 plants. Thirty-eight plants were found at Creigiau Eglwyseg (v.c. 50), the first time it has been recorded since 1907. Although rare, it does not seem to be significantly threatened, and both extant sites are Sites of Special Scientific Interest. KEYWORDS: Endemic, rare species, Wales. INTRODUCTION Hieracium cambricum (Baker) F. J. Hanbury, Welsh Hawkweed, is a very rare endemic plant, known from only three sites in Wales (Fig. 1.). Sell & West (1968) reported that it was frequent on the Great Orme, Caernarvon (v.c. 49) but had not been seen in its other two sites at Treorchy, Glamorgan (v.c. 41) and Llangollen, Denbighshire (v.c. 50) for over 60 years. Along with a number of other rare hawkweeds, it has recently been included in the 3rd edition of the Vascular Plant Red Data Book (Wigginton 1999), but nothing was known about its current status. Therefore in 1998 surveys were carried out to find out how many plants there were and determine its needs for conservation (Rich 1999); the purpose of this paper is to summarise the work. H. cambricum was first collected on the Great Orme by J. Ward in 1867, who sent material to F. J. Hanbury who thought it was a new species. In 1876, J. E. Griffith gave some specimens to W. H. Painter, who forwarded them to J. G. Baker for identification (Griffith 1895). This prompted Baker, who had known of the plant for many years but had not seen it, to describe it as H. caesium Fr. var. cambricum Baker (1879). Hanbury (1894) later upgraded it to a species, and placed it with H. vagense Ley in his group Vulgata Caulescentia. Linton (1905) transferred it to sect. Oreadea, its current position today. Pugsley (1948) included it in sect. Oreadea series Argentea. H. cambricum is a distinct hawkweed not closely allied to any other British species, and is morphologically constant in cultivation over decades (e.g. Hanbury 1894; Marshall 1901). Its outstanding characters are that it is virtually glabrous (even from the first growth), has a pale green-glaucous hue, narrow, sharply toothed or laciniate leaves with dull green or purplish spots (well-developed on the larger leaves, less so on the smaller ones), and strongly imbricated, obtuse and relatively hairless phyllaries (Pugsley 1948). The main source of confusion is likely to be with H. vagense, from which it can be distinguished by the more divided, subglabrous leaves, grey- green phyllaries, and lighter yellow ligules (leaves deeply toothed or sub-pinnatifid, pilose beneath, phyllaries dark green and ligules yellow in H. vagense). “Address for correspondence: e-mail: Timothy.Rich@nmgw.ac.uk 306 Date v.c. 41 29/6/1903 2/7/1906 v.c. 49 6/1867 1869 (pre 1889) 30/8/1876 1876 7/1879 7/1879 7/1880 1885 7/1886 7/1888 1888 29/5/1889 7/1890 7/1891 7/1691 25/6/1891 8/1891 6/1892 3/8/1892 a/1892 6/1892 24/6/1896 1898 1898 18/6/1901 21/6/1901 21/6/1901 9/7/1904 ZHI 90S 6/1905 26/6/1905 17/6/1907 18/6/1907 8/1908 Zi A92 1918 13/6/1919 20/5/1920 27/5/1920 t. Cy G, RICH TABLE 1. LIST OF RECORDS OF HIERACIUM CAMBRICUM Collector H. J. Riddelsdell Site Craig Fawr, Treorchy Sources BM, CGE, Riddelsdell (1907a) H. J. Riddelsdell Craig Fawr, Treorchy Riddelsdell (1907b) S. H. Bickham Llandudno CGE J. Ward Great Orme’s Head; cultivated, root collected K 1867 J. Ward Great Orme’s Head K J. Cryer Great Orme NMW J. Griffith Great Orme Griffith (1895) W. H. Painter Great Orme’s Head K; possible syntype? J. Gniffith Great Orme CGE; possible syntype? J. Griffith Great Orme CGE, K E. Starker Orme’s Head, cultivated Croydon A. Bennett OXF, K, NMW, OXF J. Griffith Great Orme BM, NMW, OXF J. Griffith Great Orme BM E. Starker Orme’s Head, cultivated Croydon A. Bennett OXF R. W. Scully Great Orme BM H. T. Mennell Great Orme BM J. Griffith Great Orme BM S.H.Bickham — Inravines amongst the rocks, Great Orme’s CGE Head J. L. Williams Great Orme NMW S. H. Bickham Great Orme BM E. F. Linton Great Orme’s Head, cultivated LIV; Set of British Hieracia (1896, etc.) no. 92. A. H. Wolley-Dod Great and Little Orme BM J. Griffith Great Orme BM, K J. L. Williams Great Orme BM E.S. Marshall Great Orme’s Head, cultivated NMW, OXF; Set of British Hieracia (1896, etc.) no. 92. E. F. Linton Great Orme’s Head, cultivated CGE, K, OXF E.S. Marshall Az Ley A. Ley E. F. & W. R. Linton A. Ley J. Griffith C. P. Hurst a ; = Soy {cox} aq a iq?) La} coy Pee = S S 2. ANAM D> > C S iq) en Q0 eS “< Great Orme’s Head, cultivated Great Orme’s Head South of Great Orme South of Great Orme North flank of Great Orme Highest corner of south west cliffs of Great Orme . Orme’s Head Great Orme North flank of Great Orme North flank of Great Orme Tyn-y-Coed, Great Orme Great Orme Great Orme, roots cultivated at Sedbury Great Orme, roots cultivated at Sedbury Great Orme Great Orme NMW Set of British Hieracia (1896, etc.) no. 92. LIV, K; BEC reports K, OXF, NMW;; Ley (1901); Set of British Hieracia (1896, etc.) no 165. Baker (1879). BM: Baker (1879). NMW BM, K, OXF; BEC reports HIERACIUM CAMBRICUM 307 TABLE 1. CONTINUED Date Collector Site Sources 31/7/1922 H.W. Pugsley Great Orme BM 8/6/1923 S. H. Bickham Great Orme, roots cultivated at Sedbury K, OXF 23/6/1928 T. J. Foggitt Great Orme BM 19/5/1946 J. A. Whellan Great Orme BM 12/6/1947 J. E. Lousley Cliffs near light house, Great Orme K 3/7/1949 __—*~P. D.. Sell SW side of Great Orme CGE 26/6/1952‘ P. D. Sell NE side of Great Orme CGE 11/7/1954 V. Gordan Great Orme, limestone cliffs NMW pHoM955 PF. Yeo Near Gogarth, Great Orme, limestone rocks CGE c. SH/763.831, c. 350’ 8/6/1964 U.K. Duncan Great Orme CGE 25/7/1969 W. Ramsden Great Orme’s Head on the plateau above the LIV marine drive near the light house 14/5/1988 J. Bevan Great Orme’s Head, west facing cliffs fide D. McCosh Hieracium database 29/5/1996 -V. Jones Great Orme’s Head, south facing cliff fide D. McCosh Hieracium database 11/9/1998 T.C.G.Rich& Great Orme, cliffs above marine drive from NMW W. McCarthy SH/767.842 to SH/756.835; 75 plants wc. 50 25/6/1901 A. Ley Cefn Fedw, Denbigh BM, CGE, K, OXF, NMW: Set of British Hieracia (1896, etc.) nowk65: 16/7/1903 A. Ley Limestone cliffs opposite Eglwyseg Church, BM Cefn Fedw 25/6/1907 A. Ley Cefn Fedw, opposite the new church, ridges CGE 10/9/1998 T.C.G. Rich Creigiau Eglwyseg SJ/22.45; 38 plants NMW DISTRIBUTION HISTORICAL INFORMATION Locality and habitat information was taken from the literature and herbarium sheets at BM, CGE, K, LIV, OXF and NMW. Information from the Countryside Council for Wales database of Red Data Book species was provided by R. A. Jones, and from the Hieracium database by D. McCosh. The records in Table 1 were used to direct field surveys in 1998. It is quite possible that it occurs in other currently unknown localities. Graig-fawr, Treorchy (SS/92.95 to SS/92.96), v.c. 41 H.. cambricum has only been collected once from Graig-fawr by H. J. Riddelsdell in 1903. Graig- fawr is a large cirque above Treorchy composed of Pennant Sandstone, most of which is very acidic but with occasional calcareous outcrops associated with coal and other seams. This site is so different to and so disjunct from the two known sites on Carboniferous Limestone in North Wales (see below) that the record looks erroneous, were it not for the consistency of Riddelsdell’s specimens and literature references. Graig-fawr is a huge site which is difficult to search thoroughly. It has been investigated for hawkweeds on at least four occasions since the 1950s (e.g. Wade et. al. 1994) but H. cambricum has not been refound. Riddelsdell (1907b) reported it growing with H. argenteum Fr., which is now confined to one gully at SS/927.956. The rocks in this gully are unstable, and potentially a small colony of H. cambricum could have been lost to a rock fall (one significant fall was noted). H. cambricum is probably extinct at this site. Great Orme (SH/7.8), v.c. 49 H. cambricum was first found by S. H. Bickham in 1867, and is still well known to local botanists. 308 T. C. G. RICH On 10 and 11 September 1998, a search was made by T.C.G.R. and W. McCarthy stopping at regular intervals along Marine Drive, and surveying sections of c. 50-100 m of cliffs, depending on ease of access. In ac. 25% sample of sites along 2:5 km of cliff, 75 plants were found scattered between SH/767.842 and SH/756.835. This suggests that the total population on the Orme is approximately 300 plants. Creigiau Eglwyseg (Cefn Fedw), Llangollen (SJ/22.45), v.c. 50 H. cambricum was found at this site in 1901 by A. Ley, and again in 1903 and 1907. There are no more recent records (one 1952 record held at the Biological Records Centre is a result of confusion with the Great Orme). On 10 September 1998, Creigiau Eglwyseg was visited specifically following up the localities on the herbarium sheets ‘opposite the new church’ and ‘Limestone cliffs opposite Eglwyseg Church’. Thirty-four plants were rediscovered scattered over c. 200 m of the major cliff between SJ/221.457 and SJ/222.459, at an altitude of c. 400 m. On the ridge above, another four plants were found at c. SJ/222.458, 430 m altitude. The plants were generally tucked into tiny crevices in the limestone on + vertical rock faces, with no competition from other species. There could be more plants elsewhere on this massive site. Unconfirmed sites One 1892 specimen by A. H. Wolley-Dod is labelled ‘Great and Little Orme’ (Table 1). There are no other records from the Little Orme, and the one label appears to have been used for all material collected from both sites by Wolley-Dod. Specimens distributed as the Set of British Hieracia no. 165 bear the label ‘south of Great Orme’. Whilst there are limestone outcrops to the south of the Orme (e.g. Bryn Maelgwyn SH/796.805, The Vardre SH/782.795), H. cambricum has not been reported when investigated for other hawkweeds. The locality could mean south of the Great Orme summit, which would relate to localities known on the Orme itself. A 1908 specimen collected by F. A. Lees from Tyn-y-Coed, Great Orme (c. SH/774.826; this distinguishes it from Tyn-y-Coed, a farmstead in the valley south of the Orme at SH/792.797). W. McCarthy lives in Tyn-y-Coed and has never seen the plant, but it is possible that it could have occurred there in the past. HABITATS Hieracium cambricum is a plant of dry rock ledges and crevices, with little competing vegetation in open, exposed, predominantly ungrazed situations. It appears to tolerate at least light shade. The soil in the rock crevices are impossible to extract, but on the limestones at least are likely to be immature rendzinas with high pH. Soil collected from the base of the cliff at Creigiau Eglwyseg with seedlings immediately below mature plants was a black, fine, organic rendzina, pH 7:8 (measured with a pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water). At the possible historical site on Pennant Sandstone in Graig-fawr, the soils were rankers derived from shale, pH 6:8. The vegetation of the limestone rocks at the Great Orme and Creigiau Eglwyseg is very similar, with typical limestone ledge communities with Asplenium ruta-muraria, Solidago virgaurea and Festuca ovina (referable to the OV39 Asplenium trichomanes - Asplenium ruta-muraria community of British Plant Communities; Rodwell et al. 1991 et seq.). On the Great Orme, a few plants were also found with Helianthemum canum in ledge communities transitional to the N.V.C. community CGI Festuca - Carlina grassland, and rarely in the lower MC4 Brassica oleracea maritime cliff community. At both sites, vegetative plants were seen in grazed CG10 Festuca- Agrostis-Thymus grassland. At Graig-fawr, the vegetation of the calcareous ledge where H. cambricum may have occurred is different to the acidic Calluna/Vaccinium-dominated vegetation of the remainder of the site but was not investigated. The more accessible plants had their inflorescences eaten by feral goats or rabbits on the Great Orme but the basal rosettes otherwise appeared tolerant of grazing. There was little sign of slug or snail herbivory in the wild, though seedlings are susceptible in cultivation. Plants at Creigiau Eglwyseg had severe leaf roll caused by aphids. HIERACIUM CAMBRICUM 309 FIGURE 1. National distribution map of Hieracium cambricum. @ 1998, O pre-1950. Plotted using DMAPW by Alan Morton. LIFE CYCLE Hieracium cambricum is a perennial which regenerates by seed. The main flowering period is June and early July, but plants at both Creigiau Eglwyseg and Great Orme had a second flowering period in the autumn (this often happens in Section Oreadea; Sell & West 1968). Plants at Creigiau Eglwyseg flowering in 1998 had some dead flowering stems from 1997 indicating that they are polycarpic. Hawkweeds are usually regarded as obligate apomicts but this has not been tested specifically in H. cambricum. The seeds are wind-dispersed, and tests in windless conditions show that seeds fall at a velocity of c. 150-200 cm per second, but can be kept afloat in the air by a breeze. There was ample evidence of regeneration from seed at the Great Orme, and some possible seedlings at Creigiau Eglwyseg. Seedlings have broader, nearly entire, sparsely hairy leaves. CONSERVATION Hieracium cambricum is not protected under Schedule 8 of the Wildlife and Countryside Act 1981, but is included in the Vascular Plant Red Data Book (Wigginton 1999), which should draw attention to its rarity and be sufficient to ensure its survival. All three sites also have statutory 310 T.€,G.RICH protection as S.S.S.I.s. At both Creigiau Eglwyseg and the Great Orme the populations grow on cliffs, where there is effectively no requirement for management (grasslands around both sites are grazed). H. cambricum does not appear to be under any immediate threat, though rock climbing, scrub invasion, spread of alien plants, rocks falls and rock safety work could potentially affect the plants in the longer term at both extant sites. It is suggested that the populations should be monitored every five years. Seed has been collected from Creigiau Eglwyseg and the Great Orme and deposited in the Millennium Seed Bank at the Royal Botanic Gardens, Wakehurst Place. ACKNOWLEDGMENTS This work was jointly funded by Countryside Council for Wales and National Museum & Gallery, Cardiff as part of a joint project into conservation and ecology of critical species. The Royal Botanic Gardens, Kew contributed towards the cost of collecting seeds for the Millennium Seed Bank. I would like to thank the Keepers of BM, CGE, K, LIV and OXF for access to specimens, Alan Orange for identification of bryophytes, Peter Sell for discussion, David McCosh for records from the B.S.B.I. Hieracium database and John Edmondson, George Hutchinson, Andy Jones, Serena Marner, Wendy McCarthy, Alison Paul and Sarah Smith for their help. REFERENCES BAKER, J. G. (1879). On a variety of Hieracium caesium from the Great Ormes Head. Journal of botany 17: 360-362. ELLIS, R. G. (1983). Flowering plants of Wales. National Museum of Wales, Cardiff. GRIFFITH, J. E. (1895). The flora of Anglesey and Carnarvonshire. Nixon & Jarvis, Bangor. HANBURY, F. J. (1894). Notes on British Hieracia. Journal of botany 32: 225-229. HUTCHINSON, G. (1986). Graig Fawr and Graig Fach above the Rhondda 13 July [1985]. Cardiff Naturalist’s Society Botanical Section Newsletter 5: 5. Ley, A. (1901). H. cambricum F. J. Hanb. Report of the Botanical Exchange Club of the British Isles 2: 18. LINTON, W. R. (1905). An account of the British Hieracia. West, Newman & Co., London. MARSHALL, E. S. (1901). H. cambricum F. J. Hanb. Report of the Botanical Exchange Club of the British Isles 2:18. PUGSLEY, H. W. (1948). A prodromus of the British Hieracia. Journal of the Linnean Society of London (botany) 54: 1-356. RIDDELSDELL, H. J. (1907a). A flora of Glamorganshire. Journal of botany, 1907 (Supplement): 1-88. RIDDELSDELL, H. J. (1907b). H. argenteum Fr. Report of the Botanical Exchange Club of the British Isles 2: 23% RODWELL, J. S. et al., eds. (1991 et seq.). British plant communities. 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. B.S.B.1., London. WADE, A. E., KAY, Q. O. N. & ELLIS, R. G. (1994). Flora of Glamorgan. H.M.S.O., London. WIGGINTON, M. J, (1999). British Red Data Books. 1. Vascular plants. 3rd ed. Joint Nature Conservation Committee, Peterborough. (Accepted August 1999) Watsonia 23: 311-316 (2000) Bit Conservation of Britain’s biodiversity: Hieracium tavense (Asteraceae), Black Mountain Hawkweed T. C. G. RICH* Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP and L. HOUSTON 2 West Grove, Montpelier, Bristol BS6 5LS ABSTRACT Hieracium tavense is a very rare endemic species, restricted to one waterfall in the Upper Tawe Valley, Brecon (v.c. 42), Wales. It was first found in 1899 but appears to have only been seen a few times subsequently. In 1998, a survey by climbing found 13 inaccessible plants on one cliff. It is a polycarpic perennial which flowers in late July-August and reproduces by seed. Although rare and with a critically small population, it is not threatened except potentially by a severe rock fall. KEYWORDS: Endemic, rare species, Wales INTRODUCTION Hieracium tavense (Ley ex W. R. Linton) A. Ley, Black Mountain Hawkweed, is a very rare endemic species, restricted to a ravine in the Upper Tawe Valley, Brecon (v.c. 42), Wales. Along with a number of other rare hawkweeds, it has been included in the 3rd edition of the Vascular Plant Red Data Book (Wigginton 1999), but very little was known about its detailed distribution or ecology. Hi. tavense was first found by Augustin Ley in the Upper Tawe Valley ‘above Callwen’ in 1899, and was sent unnamed to the Botanical Exchange Club (Ley 1900). Ley took specimens into cultivation, and later submitted them to the B.E.C. as H. protractum Fr. (Ley 1903), but F. J. Hanbury suggested the general appearance of the plants was that of H. rigidum Hartmann rather than H. protractum. It was then described as H. rigidum var. tavense Ley ex W. R. Linton (Linton 1905). Later at Linton’s suggestion, Ley raised it to species noting it was close to H. latobrigorum (Zahn) Roffey (H. auratum Fr.) (Ley 1909). Zahn (1921-1923) later relegated it to a subspecies of A. laevigatum Willd., but this view was not adopted by Pugsley (1948) who retained it as a species. It is currently accepted as a species by P. D. Sell (pers. comm. 1998). This species has only been seen a few times. Pugsley (1948) had not seen it himself and noted that it had apparently not been collected since first found by Ley. B. A. Miles refound it in 1957 and visited it in 1959 and 1960, and M. Porter saw it in 1975 (Ellis 1983). In 1998, a survey was carried out to establish its current status and determine its needs for conservation (Rich 1999). The purpose of this paper is to summarise the work. TAXONOMY Hieracium tavense belongs to sect. Foliosa (Fr.) Arv.-Touv. This is a small section of ten microspecies, mainly distributed in the north and west of Britain (Fig. 1). Pugsley (1948) gives a good description, which appears to be largely based on cultivated material. Syntypes are located in LIV (annotated by Sell & West in 1959) and at BM. *Address for correspondence: e-mail: Timothy.Rich @nmgw.ac.uk 312 T. C. G. RICH AND L. HOUSTON H. tavense 1s almost certainly a neoendemic which may have evolved from another member of the sect., such as H. latobrigorum, since the last Ice Age. There appears to be nothing particularly special about the waterfall on which it occurs (many similar waterfalls occur throughout the Brecon Beacons), and it is not especially isolated from other species in the sect. (Fig. 1). Whether it evolved in situ or 1s relict in this site from a wider distribution is unknown. At least five Hieracium species grow in Nant-y-Llyn. H. tavense can be distinguished from H. diaphanum Fr., H. scoticum F. Hanb. and H. sparsifolium Lindeb. by having numerous stem leaves, the middle ones of which at least partly clasp the stem. It can be distinguished from the other, as yet unnamed, species of sect. Foliosa with which it grows at the waterfall by the linear- lanceolate (not ovate) upper stem leaves (this other species is currently under investigation). Another 23 Hieracium species are listed from the 10-km square SN/8.2 in Sell & West (1968), and more occur to the south in SN/8.1. DISTRIBUTION Locality and habitat information were extracted from the literature and from BM, CGE, LIV. OXF and NMW. Information from the Countryside Council for Wales database of Red Data Book species was provided by R. A. Jones, and from the Hieracium database by D. McCosh. The historical records indicate it has been found in one specific locality only, i.e. ‘waterfall at Nant-y- Llyn’ (Table 1), though Ley cited it in several different forms. He also noted that it was ‘not yet known from any other station other than those in the Upper Tawe Glen’ (Ley 1909) suggesting it could occur elsewhere in this area, but no other stations are known. Nant-y-Llyn is a ravine on the east side of the Black Mountain with a series of rocks and waterfalls. In 1998, 13 plants of H. tavense were found confined to an unstable cliff c. 20 m high on the north-facing side of a small cirque developed around the upper of the two larger waterfalls at c. 510 m altitude (grid reference SN/844.207). The plants could only be reached safely by climbing with ropes. One stem of one plant was removed as a voucher (NMW) and its identity confirmed by P. D. Sell. The rocks of the waterfall are composed of the Brownstones of the Lower Old Red Sandstone of Devonian age (Geological Survey of Great Britain 1979; Barclay et al. 1988). The Brownstones are red-brown and purple sandstones interbedded with red mudstones and siltstones, with occasional intraformational conglomerates and less frequent calcrete nodules. They form beds c. 400 m thick, and outcrop over much of the Brecon Beacons. At the waterfall, there are a series of hard strata c. 20-50 cm thick, interbedded with softer silts, resulting in a series of unstable and overhanging ledges on which H. tavense grows. Lower down Nant-y-Llyn, green sandstone of the Senni beds also outcrops to form a waterfall. TABLE 1. LIST OF HIERACIUM TAVENSE RECORDS Date Collector Site Source 26/7/1899 A. Ley Tawe gorge above Cellwen BM, syntype (Ley 1900) 3/8/1899 A. Ley Waterfall at Nant y Llyn near Cellwen BM (ex. herb. Hanbury) 3/8/1899 A. Ley Upper Tawe Valley BM, LIV; syntype (Ley 1900) 3/8/1899 A. Ley Head of the Tawe BM, CGE, OXF (Ley 1900) 3/8/1899 A. Ley Baggards Glen, Nant-y-Llyn, Upper Tawe BM, CGE 30/7/1902 A. Ley Upper Tawe, cultivated BM, NMW, OXF (Ley 1903) 18/7/1906 A. Ley Upper Tawe Glen, cultivated BM, CGE, LIV, NMW, OXF 1910 A. Ley Upper Tawe Glen, cultivated BM, CGE, OXF (Ley 1911) 26/7/1957 B.A. Miles rocks by the waterfall, 1650’, Nant-y-Llyn, CGE Upper Tawe 1959 B. A. Miles Nant-y-Llyn, Upper Tawe; | plant Note on specimen in CGE 1960 B. A. Miles Nant-y-Llyn, Upper Tawe; 3 plants Note on specimen in CGE 1975 M. Porter Upper Tawe Ellis (1983) 28/8/1998 T.Rich& Nant-y-Llyn; 13 plants NMW L. Houston HIERACIUM TAVENSE B43 te] FIGURE 1. Distribution of Hieracium tavense (@) and Hieracium Section Foliosa (=). ECOLOGY LIFE CYCLE Hieracium tavense is a polycarpic perennial which reproduces by seed. In 1998, the mean number of stems per plant was 2-9 (range 1-7) (Table 2). Nine out of the 13 plants (69%) were flowering, and most (76%) of the stems had flowers. The mean number of capitula per inflorescence was 3-2, which is smaller than the numbers observed on two specimens cultivated by Ley in NMW (eleven and 14 capitula respectively) suggesting sub-optimal growth conditions in the wild compared to cultivation. Some plants rooted in shallow crevices were vegetative, and most plants had some aborted capitula within each inflorescence. Many plants also had dead stems from 1997. Plants were just beginning to flower on 14 August and had virtually all finished flowering by 28 August 1998. Herbarium specimens have flowers from mid-July to the end of August. It is assumed that H. tavense is an obligate apomict like other Hieracium species but this has not been tested. Flowers seen on 14 August had not fruited by 28 August in the wild, but did so rapidly when taken into cultivation. Seed ripening is thus estimated to take 2-3 weeks. The mean number of seeds produced by each flowering plant was 225 (range 35-1015), and the total seed production of the population was estimated as c. 2030 seeds. 314 T. C. G. RICH AND L. HOUSTON TABLE 2. CHARACTERISTICS OF HIERACIUM TAVENSE AT NANT-Y-LLYN ON 28 AUGUST 1998 No. live No. stems Number of capitula No. dead stems Plant stems flowering ineachinflorescence from 1997 Notes 1 3 0 - 4 2 2 0 - 5 + 3 possible seedlings 3 7 a 3,3,4,4,4,5,6 y Huge clump 4 1 0 - 0 Stems growing downwards 5 4 3 22 6 6 2 2 ee) 6 Four capitula picked + soil collected 7 4 1 1 0 Under an overhang 8 1 0 - 0 Very small, + no soil, hanging out of cliff 9 4 ) 1,3,aborted 3 Voucher specimen collected 14/8/1998 10 1 1 8 0 Two capitula collected 14 1 1 1 0 Growing sideways under overhang 12 q 1 2 9 Growing well 13 1 1 Z 0 Plant dangling down, two capitula collected Mean 2-9 1-53 3-2 3:1 The seeds are wind-dispersed. Tests in windless conditions show that seeds fall at a velocity of c. 40-50 cm per second, but are easily kept afloat in the air by a gentle breeze (e.g. gentle blowing). Although the seeds have the potential for dispersal, the paucity of available habitat in the Upper Tawe Valley, restricted calcareous outcrops and low seed production suggest that the chances of H. tavense spreading are limited, though some possible reproduction by seed within the site was noted during the survey (Table 2). All four seedlings which germinated in cultivation had one of the cotyledons split into two, resulting in an apparent appearance of three cotyledons. ASSOCIATED SPECIES AND SOILS The species associated with H. tavense on the waterfall are listed in Table 3. An attempt was made to relate the vegetation type on the rock face to the National Vegetation Classification (Rodwell et al.1991 et seq.), but the site is somewhat heterogeneous. Most of the waterfall face appears to be H12 Calluna vulgaris - Vaccinium myrtillus heath vegetation. H. tavense occurred on the more base-rich ledges, often without any other species, and it has not been possible to ascribe it to a clear N.V.C. type. Plants were rooted in immature rankers on the ledges or sometimes directly into rock crevices with no apparent soil. The soil from the base of one plant was composed of c. 70% large fragments of Brownstone, reddish-brown mineral material and fine organic matter, pH 7-1 (measured with a pHep2 Hanna pocket-sized pH meter in a 50:50 mixture with distilled water). CONSERVATION Although the population size is critically small, the only significant threat to survival of the species is a large rock fall which could wipe. out the entire population. It is under no threat from collecting due to the very difficult climbing conditions and requirement for roped access. Although the area around the waterfall is intensively grazed, the cliff is inaccessible to sheep (H. tavense, like most hawkweeds, is likely to be sensitive to grazing). No site management is currently required, and the only potential management in the future might be control of Sorbus aucuparia trees to minimise shade. Seeds from eight flowering heads were sent to the Millennium Seed Bank at the Royal Botanic Gardens, Wakehurst Place for long-term storage. Attempts are also being made to grow the plant in cultivation. Hieracium tavense is not protected under Schedule 8 of the Wildlife and Countryside Act 1981, but is included in the Vascular Plant Red Data Book (Wigginton 1999), which should draw HIERACIUM TAVENSE 315 TABLE 3. SPECIES ASSOCIATED WITH HIERACIUM TAVENSE, NANT-Y-LLYN, 28 AUGUST 1998 WITH ‘DAFOR’ COVER RATINGS Agrostis capillaris R Oreopteris limbosperma R Anthoxanthum odoratum R Oxalis acetosella O Asplenium viride R Polypodium vulgare s.l. R Athyrium filix-femina O Solidago virgaurea O Calluna vulgaris F Sorbus aucuparia R Carex pilulifera R Succisa pratensis O Deschampsia flexuosa A Thymus polytrichus O Digitalis purpurea O Vaccinium myrtillus O Dryopteris dilatata O Valeriana officinalis R Dryopteris filix-mas R Veronica officinalis R Epilobium brunnescens O Viola riviniana R Epilobium montanum R Bryophytes Euphrasia sp. R Brachythecium plumosum O Festuca ovina R Amphidium mougeotii F Festuca rubra R Blepharostoma tricophyllum R Galium saxatile R Ctenidium molluscum F Hieracium sp. R Diplophyllum albicans O Hieracium tavense O Dicranum scoparium O Hypericum pulchrum R Fissidens taxifolius R attention to its rarity and be sufficient to ensure its survival. The upper part of Nant-y-Llyn is included within the Mynydd Du S.S.S.I, and is also part of the Brecon Beacons National Park. These designations give a significant degree of protection to the site. Thus H. tavense has no specific conservation requirements despite being very rare. The population should be surveyed every five years to monitor the status quo. ACKNOWLEDGMENTS This work was jointly funded by Countryside Council for Wales and National Museum and Gallery, Cardiff as part of a joint collaborative project into conservation and ecology of critical species. The Royal Botanic Gardens, Kew, contributed towards the cost of collecting seeds for the Millennium Seed Bank. We would like to thank the Keepers of BM, CGE, LIV and OXF for access to specimens, Alan Orange for identification of bryophytes, Peter Sell for discussion, David McCosh for Hieracium identifications and records from the B.S.B.I. Hieracium database, John Edmondson, George Hutchinson, Serena Marner, Alison Paul and Sarah Smith for assistance, and Andy Jones, Jamie Bevan and Graham Motley of C.C.W. for their help. The base map was plotted using DMAPW by Alan Morton. REFERENCES BARCLAY, W. J., TAYLOR, K. & THOMAS, C. P. (1988). Geology of the South Wales Coalfield, Part V, the country around Merthyr Tydfil. 3rd ed. H.M.S.O., London. ELLIS, R. G. (1983). Flowering plants of Wales. National Museum of Wales, Cardiff. GEOLOGICAL SURVEY OF GREAT BRITAIN (England and Wales) (1979). Sheet 231, Merthyr Tydfil. Solid and drift edition. 1:50,000. Ordnance Survey, Southampton. KENT, D. H. & ALLEN, D. E. (1984). British and Irish herbaria, 2nd ed. B.S.B.I., London. Ley, A. (1900). H.-. Report of the Botanical Exchange Club of the British Isles 1: 639. LEY, A. (1903). H. protractum Fr. Report of the Botanical Exchange Club of the British Isles 2: 55. Ley, A. (1909). Brecon and West Yorkshire hawkweeds. Journal of botany 47: 8-16 and 47-55. Ley, A. (1911). H. tavense Ley. Report of the Botanical Exchange Club of the British Isles 3: 573. LINTON, W. R. (1905). An account of the British Hieracia. West, Newman & Co., London. 316 T. C. G. RICH AND L. HOUSTON PUGSLEY, H. W. (1948). A prodromus of the British Hieracia. Journal of the Linnean Society of London (botany) 54: 1-356. RicH, T. C. G. (1999). Distribution and conservation of Hieracium tavense, Black Mountain Hawkweed. National Museums & Galleries of Wales, Cardiff. RODWELL, J. S. et al., eds. (1991 et seq.). British plant communities. 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. B.S.B.1I. London. WIGGINTON, M. J. (1999). British Red Data Books. 1. Vascular plants, 3rd ed. Joint Nature Conservation Committee, Peterborough. ZAHN, K. (1921-1923). Hieracium, in ENGLER, H. G. A., ed. Das Pflanzenreich IV, 280: 1-1705. (Accepted July 1999) Watsonia 23: 317—322 (2000) BF Conservation of Britain’s biodiversity: Rubus trelleckensis (Rosaceae), Trelleck Bramble R. D. RANDALL and T. C. G. RICH Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP ABSTRACT Rubus trelleckensis is a rare endemic bramble found only in one 10-km square near Trelleck, Gwent (v.c. 35), Wales. It is a perennial which flowers from mid-July to mid-August, and sets abundant fruit but probably does not spread vegetatively. A field survey was carried out in 1998. Five small populations were found, all in locations which were probably at one time either open heath or open Birch-Oak woodland but are now either conifer plantation, or conifers mixed with broad-leaved trees. Plants were most frequent on acidic podzols in sunny but sheltered spots on level ground. The main threats to its survival are change of land-use from forestry or changes in forestry operations. Seeds have been deposited in the Millennium Seed Bank at the Royal Botanic Gardens, Wakehurst Place. KEYWORDsS: Endemic, rare species, Wales. INTRODUCTION Rubus trelleckensis Edees & Newton, Trelleck Bramble, is a rare Welsh endemic species with a very restricted distribution centred on Beacon Hill near Trelleck in Gwent (v.c. 35, Monmouth). Along with a number of other rare brambles, it has been included in the 3rd edition of the Vascular Plant Red Data Book (Wigginton 1999), but as yet little is known about its detailed distribution or ecology. In 1998 the available information was collated, combined with a field survey, to establish its current status and determine its needs for conservation. A summary of the data are given below; full details are given in Randall & Rich (1999). TAXONOMIC HISTORY R. trelleckensis, as we now know it, has had a long and somewhat confusing history. It is a distinctive species but would probably not have received a name if it had not been distributed in the Set of British Rubi and later identified, erroneously, with a number of European species. It was first collected by A. Ley in 1885 from plantations on Beacon Hill near Trelleck, Monmouthshire. Specimens were sent to W. O. Focke for naming, who commented ‘R. virescens G. Braun, var. glandulosa. The typical plant has no glandular bristles on the flower stalks’. In 1886, Ley again visited Beacon Hill and collected further specimens which he submitted to the Botanical Exchange Club (B.E.C.) as ‘R. Borreri Bell-Salt.’ (Linton 1887). C. C. Babington commented that it was ‘Sprengelii probably’ and Focke agreed. In 1888 Ley again visited Beacon Hill and this time selected two plants for submission to the B. E.C. (Ley 1889). He suggested ‘R. Borreri, Bell-Salt. variety ?’ for one of them but Babington and Focke were unable to place it (this taxon has received no name). The other plant was submitted as ‘R. virescens G. Braun’ on the strength of the name supplied by Dr. Focke for Ley’s 1885 specimen. Ley could not guarantee that he had collected the same plant as previously and, indeed, Dr. Focke’s comments this time were less encouraging: ‘Leaves exactly as in virescens, but the other characters do not agree. I can give no name’. Ley reported that both plants grew with R. sprengelii in the plantation and that the ‘R. sprengelii var.’ partly showed the suberect habit of R. virescens. He was later of the opinion that the two species hybridised there. In 1893, Ley again visited Beacon Hill, accompanied by W. A. Shoolbred on 4 July and by E. F. Linton on 3 August. Material from these visits was distributed as no. 60 in the Set of British Rubi, labelled R. myricae Focke var. virescens G. Braun, forma glanduligera. 318 R. D. RANDALL-AND TC: G. RICH When Focke visited Trelleck and saw living material he withdrew the name he had originally suggested (there is a specimen at BM (Barton & Riddelsdell no. 4107) collected on 31 July 1894 by Ley, Focke and W. H. Purchas; a typed note states that ‘Part of this gathering is in Focke’s scaber packet (159), without name or comment’ ). As the distinctive plant was not clearly identifiable with any known species it was duly described as new as R. orthocladus Ley (Ley 1896). Rogers (1900) pointed out that the name was already in use, viz. R. orthocladus Boulay (1869), and that the Set of British Rubi No. 60 contained representative material of Ley’s plant. At that time it was thought that M. Gravet had discovered Ley’s plant at Louette St. Pierre in Belgium, and it was listed as R. orthocladus Ley in Sudre (1908). Later he revised his views (Sudre 1910) and reduced R. orthocladus Ley to a synonym of R. bracteosus Weihe ap. Le}. & Court. (Lejeune & Courtois 1831), an earlier name that had been applied to the Belgian plant. Watson (1932), at the time referee for the genus, associated a number of forms from around the country under R. egregius var. effeminatus Focke, a name that had originally been given to a plant from Oxfordshire. Included by Watson under this name was material collected by Ley from Beacon Hill Woods on 4 July 1893, labelled R. orthoclados and specimens from an open heath on Beacon Hill on 23 July 1897, labelled ‘R. orthoclados var.’. At the time of the publication of Rubi of Great Britain and Ireland (Watson 1958) and Flore Générale de Belgique (Legrain 1958) it was listed as R. bracteosus Weihe ap. Lej. & Court., which was assumed to be present in both Britain and Belgium. Some specimens in herbaria were later labelled R. euchloos Focke under which name it occurs in Flora Europaea (Heslop-Harrison 1968) on Watson’s authority. E. S. Edees and A. Newton assessed the validity of all the European species which Watson had claimed to be present in Britain, and decided that the name R. bracteosus could not be applied to the Trelleck plant and gave it a new name, R. trelleckensis (Edees & Newton 1978). It fits rather uncomfortably into its current position in Subsection Rubus but would be equally ill-fitting in Series Silvatici, or Series Mucronati. Summary of synonymy R. myricae var. virescens sensu Focke pro parte, non R. virescens G. Braun in Focke R. orthocladus Ley, non R. orthocladus Boulay R. euchloos auct., non Focke in Ascherson & Graebner R. bracteosus auct., non Weihe ex Lej. & Court. R. egregius var. effeminatus sensu W. C. R. Watson pro parte, non Focke IDENTIFICATION There are a number of species which resemble R. trelleckensis, a guide to which 1s given in Rich & Randall (1999) (see also Edees & Newton 1988). Its distinctive characters are the broad-based declining prickles, narrow green long-pointed leaflets, sub-racemose panicle, frequent unequal stalked glands and acicles on peduncles, white flowers, long-pointed green clasping sepals and carpels at first hairy. Hybrids with R. pyramidalis, R. scissus and R. sprengelii have been suggested but require more detailed study. DISTRIBUTION Locality and habitat information was abstracted from the literature, and from herbarium sheets at the Natural History Museum (BM) and the National Museum of Wales (NMW), which between them include much Botanical Exchange Club material and the herbaria of W. M. Rogers, H. J. Riddelsdell and E. S. Edees. The records indicated that R. trelleckensis is very rare, being confined to woods and heathy ground between Trelleck and Tintern in Gwent, and these areas were concentrated on for the field survey. The survey was carried out between mid-July and early September 1998, the best time being mid-July to early August when plants were flowering. About 80+ plants (i.e. distinct clumps) were found in five sites in one 10-km square, SO/5.0, all in v.c. 35 (Figure 1). RUBUS TRELLECKENSIS 319 Trelleck @7 05 00 50 55 Figure 1. Location of sites in western part of SO/5.0 10-km square searched for Rubus trelleckensis in 1998. @ plants found. O no plants found. The inset shows the location of the map in Wales. Site 1. Bargain Wood (SO/520.033) This site was first found by A. Orange in 1997. Seven plants were found scattered along about 50 metres of a footpath in a recently replanted conifer plantation. Site 2. Ninewells Wood (SO/509.037) Two plants were discovered in open heathy birch scrub. This site is immediately south of Trelleck Bog, an old site (see also below). Site 3. Beacon Hill (SO/512.054) This is the type locality for R. trelleckensis, where Ley (1896) described it (R. orthoclados) as ‘occupying a large area of woodland (some three square miles) on Beacon Hill, Monmouthshire’. Beacon Hill is now largely conifer plantation, some parts of which were being felled in rotation resulting in vigorous regeneration of the native vegetation, including brambles. About 50 plants were seen scattered in various forestry compartments across the hill. 320 R. D-RANDALE AND* EC; GF RICH Site 4. Beacon Hill - North Slope (SO/513.057) In light of Ley’s comments about the relative abundance of R. trelleckensis on Beacon Hill, a brief visit was made to the northern slopes. Three plants were found by a forestry track in a small patch where Calluna was frequent. Site 5. Vicar’s Allotment (SO/512.060) A thriving population of over 20 plants was found along a bridleway through a recently felled conifer plantation. Sites searched without success A number of other sites in the area were also visited (Fig. 1). There were historical records for Trelleck Bog (now known as Cleddon Bog) but a check of the developing birch scrub on the south and west edges of the bog failed to locate any plants. Church Hill Common, Penallt (SO/52.10) was also searched as the soil maps (Soil Survey of England and Wales 1983) indicated that a podzol similar to that exhibited on Beacon Hill and the Vicar’s Allotment was present; R. trelleckensis has not been reported from this site and none was found. Erroneous or unconfirmed records The record for ST/2.8 in Edees & Newton (1988) is based on a specimen from Pennsylvania near Castleton (v.c. 35) collected by A. E. Wade in 1941 (NMW). The specimen is not R. trelleckensis but belongs to Series Anisacanthi and is probably R. hibernicus (Rogers) Rogers. A plant from Mitcheldean Meend (v.c. 34), distributed by Ley through the B.E.C. as ‘R. orthocladus Ley var.’ was considered by Sudre to be R. orthocladus x R. gratus and subsequently described as a new species or hybrid R. dobuniensis Sudre & Ley (Sudre 1908). It is probably a hybrid of R. gratus with a species other than R. trelleckensis, perhaps R. melanocladus (Sudre) Riddelsd. which is abundant at Mitcheldean. No specimens have been traced to support the unlikely records from Bewdley (v.c. 37) and Staffordshire (v.c. 39) (Watson 1958), or Dol-y-bont, Llanfihangel and Llyfnant Valley (v.c. 46) (Rogers 1908). LIFE CYCLE R. trelleckensis is, like all brambles, a polycarpic perennial. It is more-or-less evergreen, with leaves often persisting until the new year. Panicles arise from buds in the axils of the previous year’s leaves (as in other species) but vegetative branches were not seen and it is likely that individual stems and branches die back after fruiting and are therefore biennial. The flowering time is mid-July to mid-August. Fruits were abundant at the end of August 1998, and most appeared to ripen about the same time. Dispersal of seed by birds and mammals probably occurs, as in other brambles. The fruits are quite sweet. TABLE 1. PROBABLE NATIONAL VEGETATION CLASSIFICATION TYPES AT SITES OF RUBUS TRELLECKENSIS Site National Vegetation Classification types Bargain Wood H8e Calluna - Ulex gallii heath, Vaccinium sub-community developing in 5-year old cleared and replanted conifer plantation Beacon Hill U2 Deschampsia flexuosa grassland in recently cleared conifer plantation, H8e Calluna - Ulex gallii heath, Vaccinium sub-community in replanted older conifer plantation Beacon Hill - north slope W416 Quercus - Betula - Deschampsia woodland on disturbed edge of possibly original W10 Quercus - Pteridium - Rubus community. Ninewells Wood W16 Quercus - Betula - Deschampsia woodland, replanted with conifers Vicar’s Allotment H8e Calluna - Ulex gallii heath, Vaccinium sub-community developing in 5-year old cleared and replanted conifer plantation RUBUS TRELLECKENSIS 32M Although many Rubus species spread vegetatively by adventitious roots from the ends of the arching stems this does not appear to occur very often with R. trelleckensis. Most bushes observed appear to be groups of stems from a single rootstock. Population sizes appear to depend on the amount of competition from bracken and other species, on the amount of shade, and on soil type and drainage. Plants were most frequent on the podzols of Beacon Hill and the Vicar’s Allotment, with the best-growing plants in sunny but sheltered spots on level ground where there was good moisture retention. At Beacon Hill, the state of the population was closely linked to the forestry management. In recently cleared areas there were scattered patches, mostly stunted due to the dry exposed conditions. In an area replanted with young trees, there were well-grown plants intermixed with other brambles, bracken and gorse. Where young trees were well established and beginning to crowd out the associated underscrub, R. trelleckensis and other brambles were confined to the margins of the plantation. Very few plants were found in the dense, well-established 3-4 m tall plantation except in gaps. HABITATS, VEGETATION AND SOILS All sites seen in 1998 are either conifer plantation, or conifers mixed with broad-leaved trees. An attempt was made to relate the vegetation types to the National Vegetation Classification (Rodwell 1991 et seq.), but this is difficult for recently disturbed conifer plantations (Table 1). The original vegetation types of these sites may have been the W16 Quercus - Betula - Deschampsia woodland type (possibly extending into other related woodlands such as the W10 Quercus - Pteridium - Rubus community), and the H8 Calluna - Ulex gallii heath. Ley’s specimens and his comments on its distribution suggest that R. trelleckensis thrived in woodland. R. trelleckensis is typically distributed on the tops and upper slopes of hills where a peaty soil has developed in situ over quartz conglomerate base rock. It was not found on deeper, more mineral-rich soils or the carboniferous limestone areas of the Wye Valley. Soils collected from the rooting zones in August 1998 are described briefly in Table 2. They are all heathy acidic podzols pH 3-4-4-0, probably derived from the local quartzite rock. At Beacon Hill and Vicar’s Allotment they are classified as the 631le Goldstone Association, a humo-ferric podzol, developed over quartz conglomerate of the Old Red Sandstone, and in the Ninewells Wood - Bargain Wood area as the 541c Eardiston 1 Association, a typical brown earth, developed over the Tintern Sandstone Group of the Upper Old Red Sandstone (Soil Survey of England and Wales 1983; Geological Survey of England and Wales 1974). TABLE 2. SOIL TYPES AT SITES OF RUBUS TRELLECKENSIS Site Soils Bargain Wood A fine grey-brown heathy soil with lots of organic matter, pH 4-0 Beacon Hill A fine brown heathy soil with organic matter, pH 4-0 Ninewells Wood A gritty brown organic soil, pH 3-6 Vicar’s Allotment A blackish organic soil with large quartz sand grains, pH 3-4 CONSERVATION REQUIREMENTS Rubus trelleckensis is a rare plant. It is not protected under Schedule 8 of the Wildlife and Countryside Act 1981, but is listed in the Vascular Plant Red Data Book (Wigginton 1999). None of the sites found in 1998 are scheduled as S.S.S.I.s. Given that it is only known in five sites, most with tiny populations, it would be worth trying to protect the colonies at some sites. Consideration should be given to designating Beacon Hill, the type locality with the largest population, as a Site of Special Scientific Interest for this rare endemic species. The main threats to its survival appear to be changes in the existing forestry regime (e.g. weed killing, longer periods of deep shade, fertiliser application to soils, etc.). The most favourable forestry management is currently thought to be a regime similar to that exhibited at Beacon Hill, where trees have been cropped in rotation with adjacent areas being felled several years apart. This 322 R. D. RANDALL AND T. C. G. RICH allows the possibility of seed from a regenerating section being bird-sown into the most recently felled area. The amount of regeneration on Beacon Hill shows that the seed-bank there is probably sufficient to ensure R. trelleckensis will survive as long as current forestry management continues. Blackberries from plants at Bargain Wood and Beacon Hill have been deposited at the Millennium Seed Bank at the Royal Botanic Gardens, Wakehurst Place. One plant is held in cultivation by A. Orange. ACKNOWLEDGMENTS This work was jointly funded by Countryside Council for Wales and National Museum and Gallery, Cardiff as part of a joint project into conservation and ecology of critical species. The Royal Botanic Gardens, Kew, contributed towards collecting blackberries for the Millennium Seed Bank. We would like to thank Andy Jones and Lindi Rich of C.C.W. for their help, and Alan Orange and Jessica Carey for help with field work. We would like to thank the Keeper of BM for access to the herbarium. REFERENCES EXSICCATAE BOULAY, J. N. (1864-1869). Ronces vosgiennes. Fasc. 1-5, nos. 1-152. Rambervillers. (Fasc. 5, no. 142, 1869, R. orthocladus Boulay). LINTON, E. F. & W. R., MURRAY, R. P., ROGERS, W. M. & LEY, A. (1892-1897). Set of British Rubi. Fasc. 1+4, nos. 1-106 and supplement, nos. 106A—134. (no. 60, 1894, R. myricae Focke var. virescens G. Braun, forma glanduligera). SUDRE, H. (1903-1917). Batotheca Europaea. Fasc. 1-15, nos. 1-750. (no. 218, R. orthocladoides Sudre). SUDRE, H. (1908-1912). Rubi rari vel minus cogniti exsiccati. Fasc. 1-5, nos. 1-140. Albi. (Fasc. 1, no. 7, 1908). LITERATURE EDEES, E. S. & NEWTON, A. (1978). Amendments to the British Rubus List. Watsonia 12: 133-136. EDEES, E. S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. FOCKE, W. O. (1902). Rubus L., in ASCHERSON, P. & GRAEBNER, P. Synopsis der mitteleuropdischen Flora 6 (1). Leipzig. GEOLOGICAL SURVEY OF ENGLAND AND WALES (1974). Sheet 233, Monmouth. Solid and drift edition. 1:50,000. Ordnance Survey, Southampton. HESLOP-HARRISON, Y. (1968). Rubus L., in TUTIN, T. G. ef al., eds. Flora Europaea 2: 12. Cambridge University Press, Cambridge. LEGRAIN, J. (1958). Rubus L., in ROBYNS, W., ed. Flore Générale de Belgique 3: 49. Bruxelles. LEJEUNE, A. L. S. & COURTOIS, R. J. (1831). Compendium Florae Belgicae, etc., 2: 162. Bruxelles. Ley, A. (1889). R. Borreri Bell-Salt variety? and R. virescens, G. Braun. Report of the Botanical Exchange Club of the British Isles 1: 207. Ley, A. (1896). Herefordshire Rubi. Journal of botany 34: 159-160. LINTON, E. F. (1887). R. Borreri Bell-Salt. Report of the Botanical Exchange Club of the British Isles 1: 148. RANDALL, R. D. & RICH, T. C. G. (1999). Distribution and conservation of Rubus trelleckensis, Trelleck Bramble. National Museums & Galleries of Wales, Cardiff. RODWELL, J. S., ed. (1991 et seg.). British plant communities. Cambridge University Press, Cambridge. ROGERS, W. M. (1900). Handbook of British Rubi. London. ROGERS, W. M. (1908). R. orthoclados, Ley. Report of the Botanical Society and Exchange Club of the British Isles 2: 367, . Soil Survey of England and Wales (1983). 1:250,000 Soil Map. Sheet 2. Wales. Ordnance Survey, Southampton. SUDRE, H. (1908). Rubi Europae, etc.. Paris. SUDRE, H. (1910). Les Rubus de Belgique. Bulletin de la Société royale de Botanique de Belgique 47: 185-250. WATSON, W. C. R. (1932). R. egregius Focke *var. effeminatus Focke. Report of the Botanical Society and Exchange Club of the British Isles 9: 645-646. WATSON, W. C. R. (1958). Handbook of the Rubi of Great Britain and Ireland. Cambridge University Press, Cambridge. WIGGINTON, M. J. (1999). British red data books: vascular plants. 3rd ed. J.N.C.C., Peterborough. (Accepted March 2000) Watsonia 23: 323-326 (2000) 323 The nativeness and non-nativeness of species M. B. USHER* Scottish Natural Heritage, 2 Anderson Place, Edinburgh EH6 5NP ABSTRACT It has been usual to classify species as native or non-native. However, in relation to biodiversity conservation and species recovery programmes, it has been recognised that there are “shades of nativeness” between these two extremes. Scottish Natural Heritage has recently adopted a classification of six categories - native, formerly native, locally non-native, long-established, recently arrived and non-native. These six categories are described and examples given. KEYWORDS: alien species, introduced species, native species. INTRODUCTION Webb (1985) made a systematic attempt to establish a series of criteria which could be used to reach a decision on whether a plant species is native or not. His eight criteria were fossil evidence, historical evidence, habitat, geographical distribution, frequency of known naturalisation, genetic diversity, reproductive pattern and possible means of introduction. A ninth criterion, entomological evidence, was added by Preston (1986). There are two particular features of the use of this set of criteria. First, implicitly there is a time scale; this is taken as the ending of the last glaciation 10,000 or so years ago and hence the criteria are not used for species that might have been in the geographical area being studied in inter-glacial periods. Second, the criteria are indicative rather than definitive; thus assessment of “native” or “non-native” status depends upon the balance of probabilities, weighing evidence from all of the nine criteria. Much of the work related to implementation of the United Kingdom’s Biodiversity Action Plan (Anon. 1994) is implicitly, or at times explicitly, concerned with discriminating between native and non-native species of plants and animals. Similarly, species recovery programmes are based on enhancing populations of native species, or aim at re-introducing native species that have become extinct locally or nationally. Examples are the successful re-introduction of Milvus milvus (red kite) into England and Scotland and of Haliaeetus albicilla (white-tailed eagle) into western Scotland, or the on-going studies into re-introducing Castor fiber (European beaver) into Scotland (Anon. 1998b). In contrast, site management often targets non-native species for control; such species include Rhododendron ponticum (Usher 1986) and Fallopia japonica (e.g. Pope 1997). Whereas Webb’s (1985) criteria were designed to discriminate between native and non-native species, Scottish Natural Heritage recognised that these were two of a number of possible categories. The British Ornithologists’ Union, for example, recognised rather more categories in compiling The British List of bird species (Anonymous 1998a). The aim of this study was therefore to review species that occur in Scotland, and to propose a classification of nativeness that would be useful as a factor in conservation practice. CATEGORIES OF SPECIES A geographical scale needed to be defined. For the purposes of this study the whole of Great Britain was used because of its island character. The classification, in Table 1, is therefore related to the whole island, but it could equally be applied to any defined geographical area and portions of that area. * Address for correspondence: e-mail MBU @snh.gov.uk 324 M. B. USHER Native species are those that are presumed to occur in Great Britain by natural means. Examples of such native species would include such common plants as Bellis perennis and endemic species such as Primula scotica. Both of these species are native in Great Britain as a whole: however, if a smaller geographical area is used, both species are native in Scotland, whereas only one, B. perennis, is native in Wales. Formerly native species are those that are known to have occurred in Great Britain since the last ice age, but which have become extinct. Many of the species in this category apparently became extinct because the environment changed (category 2a in Table 1). Birks (1997), for example, listed a number of such plants, including Papaver sect Scapiflora and Cassiope spp. Other species in this category might have become extinct through human activity rather than environment change (category 2b in Table 1). Examples include Trichophorum alpinum, whose only British site was destroyed by marl digging in the 13 years following the species’ discovery in 1791, and Pinguicula alpina, whose only habitat was encroached by agriculture, subsequently invaded by coniferous trees, and not assisted in the mid-19th century by the avariciousness of plant collectors (Lusby 1998). Locally non-native species are those that are native in part of Great Britain, but which have been introduced to the local area being considered. An example is Luronium natans, which is native in England and Wales but not native in Scotland where it has been introduced to at least three sites (Stewart et al. 1994): whereas it is native in Great Britain it is non-native in Scotland. There are few botanical instances where this category might be important from a biodiversity point of view, but in some cases locally non-native animals present a serious threat to native species. An example is the occurrence of Erinaceus europaeus (hedgehog) in the Outer Hebrides. It never naturally reached those islands, but since its introduction it has been creating serious problems for ground- nesting birds. Long-established species have often been described as “naturalised”; they have become part of food webs where the majority of species are native. An example of a long-established species in Britain is Acer pseudoplatanus, which Salisbury (1961) presumed to have been introduced by the Romans of the order of two thousand years ago. This species is long-established in Great Britain, whereas a species such as Fagus sylvatica, which is native in the southern part of Britain, is locally long-established in many northern areas. Recently arrived species are those for which there is as yet insufficient evidence to decide whether they arrived “naturally” (category 5b in Table 1) or were assisted by some human agency (category 5a in Table 1). In a sense, it is a temporary category, a useful classification pending further scientific evidence that would allow the species to be assigned to category | (native) or to category 6 (non-native). An example would be the discovery of Serapias parviflora in Cornwall in 1989. Both Sell & Murrell (1996) and Stace (1997) inclined to the view that this orchid species was deliberately introduced (category 5a), whereas French er al. (1999) considered it likely to be a natural colonisation (category 5b) because it occurs in dry maritime grassland on islands off the Brittany coast. Until there is more evidence, the species cannot be assigned with certainty to categories | or 6. Finally, non-native species are those where all of the evidence points to the fact that they have been brought to Great Britain by people, either intentionally or unintentionally. Logically, if a species has never been native in any part of Britain, then the application of the term can only be to Britain as a whole rather than to any part of Britain. These six categories are not mutually exclusive. Categories 1, 2 and 3 in Table | could be referred to as British native species because, at some time during the last 10,000 years or so, they have occurred naturally in at least some part of Great Britain. Similarly, categories 4 and 6 in Table 1 could be referred to as non-native since neither category has ever occurred naturally in the country. Category 5 is more problematic, but in general 5a would fall into the non-native group whereas 5b are more likely to be treated as native. Although there are therefore certain ambiguities in the definition of the six categories and their subdivisions, they nevertheless prove useful in thinking about the conservation and management of species. NATIVE CATEGORIES 325 TABLE 1. SIX CATEGORIES, TWO SUBDIVIDED, FOR CLASSIFYING SPECIES ACCORDING TO THEIR NATIVENESS. THE CATEGORIES CAN BE APPLIED TO THE WHOLE OF GREAT BRITAIN (B) OR JUST TO A LOCAL PART OF BRITAIN (L). Category Application Sub-divisions 1. Native Bly - 2. Formerly native BE a. Extinct because of environment change b. Extinct due to human activity 3. Locally non-native le - 4. Long-established BL - 5. Recently arrived Bye a. Arrival suspected of being human assisted b. “Natural” means of arrival suspected 6. Non-native B - DISCUSSION The categories of nativeness (Table 1) have been developed for nature conservation purposes. Both habitat loss (and fragmentation) and the effects of non-native species are seen to be the two primary causes of the loss of biodiversity on the Earth (e.g. Heywood 1995). It is therefore important to consider attitudes to the “nativeness” of species and to consider the conservation actions that might follow from such categorisation in the geographical area of concern. The six categories listed in Table 1 were conceived in relation to Great Britain, a single island that is in many respects a single political entity. Inevitably with national biodiversity action plans and with national conservation agencies, political boundaries tend to predominate in considerations of nativeness. However, it might be preferable to conceive of more naturally- defined boundaries, such as the archipelago of islands off the coast of north west Europe (i.e. the whole of the British Isles) or, preferably, biogeographically distinct areas (cf. Usher & Balharry (1996) for a division of Scotland into twelve biogeographical zones). The efforts of the European Commission to establish trans-national conservation, on the basis of a small number of biogeographical regions encompassing all 15 Member States, is certainly an attempt to move from a national to a more natural perspective for the conservation of species and habitats (Anon. 1992). Logically, natural boundaries should be favoured over political ones, but, pragmatically, this is unlikely to be realised! Table 1 can also be compared with the British Ornithologists’ Union’s categorisation of birds for inclusion on the British List (Anon. 1998a). That categorisation includes native species (A), formerly native species (B, with a date of 31st December 1949 as the decision point), introduced species that have established breeding populations (C), species that would be in another category but about which there is some doubt (D), and species that have been introduced and which have not established self-sustaining populations (E). In a comparison with Table 1, categories A and 1 are similar, as are B and 2. Category C relates to both 4 and 6, and possibly also to 5a, whereas category D relates primarily to 5b. Category E relates to 6. Nothing in the BOU list equates to category 3 in Table 1, but this could be predicted with a highly mobile (and migratory) taxonomic group such as the birds. Williamson’s (1996) consideration of biological invasions led him to prepare a conceptual framework, of which two points are particularly pertinent. First, most invasions fail, with only about 10% of introduced species establishing a population. Second, most invasions, probably around 90%, have only minor consequences. This implies that only about 1% of introduced species become “pests”. Does this matter? If one takes Clement & Foster’s (1994) book, which lists 3,586 non-native species recorded in the British Isles, of which 885 are established, then between 36 species (1% of 3,586) and 88 species (10% of 885 species) might be expected to pose problems for native species and natural plant communities. There is therefore potentially a large task to control and manage such a number of problematic non-native species. Nature conservation, or biodiversity action plans, will focus on native species (category 1 in Table 1). Because a proportion of the non-native species (category 6) can become pests, it is useful 326 M. B. USHER to think of controlling them, or even eradicating them, as soon as they are noticed. Re-introduction programmes can be focused on the formerly native species, but only if environmental conditions are still suitable for them to establish self-sustaining populations (i.e. category 2b). Generally there are risks in introducing species beyond their native range (category 3), but care needs to be exercised in deciding to control long-established species (category 4) because they may, over the centuries, have become part of cultural landscapes. Caution is needed with species that have recently arrived; there may be benefits in treating category 5a as if they were category 6, but it has to be remembered that species distributions are dynamic, and that distributions will change naturally under the various scenarios of climate change. There is an increasing scientific basis for adopting a precautionary approach to dealing with the arrival and establishment of non-native species (and, by analogy, also with non-native genotypes). Williamson’s (1996) study documented the problems that biological invasions can pose to a territory’s biodiversity. There is also an increasing practical basis for the early control of non- native species. Once a non-native species has become established, and starts to have an impact on local biodiversity, it can be extremely expensive to mount a control campaign, even if control is ultimately possible. REFERENCES ANONYMOUS (1992). Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal of the European Communities L206: 7-50. ANONYMOUS (1994). Biodiversity: the UK Action Plan. H.M.S.O., London. ANONYMOUS (1998a). The British List: the official list of birds of Great Britain with lists for Northern Ireland and the Isle of Man. British Ornithologists’ Union, Tring. ANONYMOUS (1998b). Re-introduction of European beaver to Scotland: a public consultation. Scottish Natural Heritage, Perth. BirKS, H. J. B. (1997). Scottish biodiversity in a Scottish context, in FLEMING, L. V., NEWTON, A. C., VICKERY, J. A., & USHER, M. B., eds. Biodiversity in Scotland: status, trends and initiatives, pp. 21-35. The Stationery Office, Edinburgh. CLEMENT, E. J. & FOSTER, M. C. (1994). Alien plants of the British Isles. Botanical Society of the British Isles, London. FRENCH, C. N., MURPHY, R. J. & ATKINSON, M. G. C. (1999). Flora of Cornwall. Atlas of the flowering plants and ferns of Cornwall with notes on some species recorded in the Isles of Scilly. Wheal Seton Press, Camborne. HEYWOOD, V. H., ed. (1995). Global biodiversity assessment. Cambridge University Press, Cambridge. Lussy, P. (1998). On the extinct plants of Scotland, in LAMBERT, R. A., ed. Species history in Scotland, pp. 45-62. Scottish Cultural Press, Edinburgh POPE, P. J. (1997). The spread of Japanese knotweed and Himalayan balsam. B.S.B.I. News 76: 62. PRESTON, C. D. (1986). An additional criterion for assessing native status. Watsonia 16: 83. SALISBURY, E. (1961). Weeds and aliens. Collins, London. SELL, P. & MURRELL, G. (1996). Flora of Great Britain and Ireland. Volume 5: Butomaceae - Orchidaceae. Cambridge University Press, Cambridge. STACE, C. (1997). New flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. STEWART, A., PEARMAN, D. A. & PRESTON, C. D., eds. (1994). Scarce plants in Britain. Joint Nature Conservation Committee, Peterborough. USHER, M. B. (1986). Invasibility and wildlife conservation: invasive species on nature reserves. Philosophical transactions of the Royal Society, series B 314: 695-710. USHER, M. B. & BALHARRY, D. (1996). Biogeographical zonation of Scotland. Scottish Natural Heritage, Perth. . WEBB, D. A. (1985). What are the criteria for presuming native status? Watsonia 15: 231-236. WILLIAMSON, M. (1996). Biological invasions. Chapman & Hall, London. (Accepted March 2000) Watsonia 23: 327-330 (2000) i, A new record for Salix x angusensis (Salicaceae) Rech. f. from Ainsdale Sand Dunes National Nature Reserve, S. Lancs. v.c. 59 R. D. MEIKLE Ranscombe Lodge, Wootton Courtenay, Minehead, Somerset TA24 SRA and N. A. ROBINSON 3 Abbey Drive, Natland, Kendal, Cumbria LAY 7QN ABSTRACT Salix x angusensis Rech. f. (S. cinerea subsp. oleifolia x S. repens var. argentea X S. viminalis), a very rare hybrid, previously known only from the type locality, has been found on Ainsdale Sand Dunes National Nature Reserve. Type material of the hybrid has been located, and an amended description drawn up from additional foliage and catkin material. The paper includes notes on the population and ecology of the hybrid. KEYWORDS: willow, hybrid. INTRODUCTION In July 1993 N.A.R. sent R.D.M. some freshly collected specimens of willows from Ainsdale Sand Dunes National Nature Reserve, S. Lancs. v.c. 59. Amongst these were Salix repens L. var. argentea (Sm.) Wimm. et Grab. and the rare $. x doniana G. Anderson ex Sm. (S. purpurea L. x S. repens L.), both of which were known to occur in the area. But a third specimen (Robinson 65) defied immediate identification, though it was evidently a hybrid, with S. repens as one of its parents. This puzzling specimen was successfully rooted and grown in R.D.M.’s garden, and developed into a slender, erect shrub about 1 m high, with small, greyish, acuminate leaves, and numerous narrow-cylindrical female catkins, with hairy ovaries, and remarkably elongate styles and stigmas, superficially not unlike the styles and stigmas of the arctic-alpine S. Japponum L., and very distinct in this respect from other lowland willows. At first, to account for this unusual feature, R.D.M. thought it just possible that the puzzling willow might be S. myrsinifolia Salisb. x S. repens L. (S. x felina Buser ex Camus & A. Camus) but S. myrsinifolia is not known to occur anywhere in the vicinity of Ainsdale, nor apart from the elongate styles and stigmas, was there any evidence of S. myrsinifolia in the make-up of the hybrid. Further searches by Mr Tony Duckels in 1996 and 1997 failed to furnish additional material of the puzzling willow, though S. cinerea L. subsp. oleifolia Macreight (S. atrocinerea Brot.), S. repens var. argentea and S. cinerea subsp. oleifolia x S. repens var. argentea was received from the same area of the dunes. Happily, the puzzling plant was relocated by N.A.R. on 10 August, 1998, by which time R.D.M. concluded that it must be a triple hybrid, S. cinerea subsp. oleifolia x S. repens var. argentea x S. viminalis L., the presence of the last-named being necessary to account for the acuminate leaves of the puzzling plant. This triple hybrid has only once been recorded, as S. X angusensis Rech. f. (Rechinger 1950) from Angus v.c. 90, “Barry Links near Carnoustie, September 1947, K. H. Rechinger, no. 45”. Rechinger’s short description of habit, twigs and foliage did not exactly tally with that of the Ainsdale plant, so it became necessary to examine the type, stated by the author to have been deposited in BM. R.D.M. had earlier tried, without success, to locate this type when preparing an account of Salix for Stace (1975), and a further effort by Mr Roy Vickery (BM) was likewise unsuccessful. An enquiry to E was no more fruitful, though Mr Douglas McKean drew attention to the fact that Rechinger (op. cit.) had also 328 R. D. MEIKLE AND N. A. ROBINSON stated, “Numbers quoted refer to dried specimens, duplicate sets of which have been deposited at the British Museum (Natural History) and at Angus Herbarium”. At Douglas McKean’s suggestion R.D.M. wrote to Miss Rachel Benvie, Curator, Montrose Museum and Art Gallery, Montrose, Angus, but was informed that no material of S$. X angusensis was to be found in the collections there. Almost on the point of giving up hope of finding the missing types, a good photocopy of the type was received from W. Examination of the photocopy satisfied us that the Ainsdale Salix and S. X angusensis are indistinguishable. Since Prof. Rechinger’s original description is brief, and based exclusively on foliage, I subjoin a more detailed description of this rare and interesting hybrid. DESCRIPTION Salix X angusensis Rech. f. in Watsonia, 1(5): 275 (1950); Meikle in Stace, Hybridization and the Flora of the British Isles, p319 (1975). Salix cinerea L. subsp. oleifolia Macreight (S. atrocinera Brot.) x S. repens L. var. argentea (Sm.) Wimm. ex Grab. (S. arenaria L.) x S. viminalis L. Holotype: Scotland; Angus (v.c. 90), “Barry Links near Carnoustie, September 1947, K. H. Rechinger no. 45” (W). Slender, erect or sprawling shrub, c. | m high, but probably taller when more mature; twigs at first densely appressed-pubescent, but soon becoming glabrous or subglabrous and rather lustrous dark reddish-brown. Leaves numerous and rather close together, lanceolate-acuminate, (2-5—)3-5—6(—7) cm long, (O-5—)1—2(—2:5) cm wide, dull green and thinly pubescent above, densely appressed sericeous-pubescent below, margins flat or undulate, entire or subentire, midrib prominent below, lateral nerves numerous, close together ascending; petiole short, usually less than 5 mm long; stipules lanceolate, 3-5 mm long, 1-5—2-5 mm wide, fairly conspicuous on young growths but soon shed. Female catkins numerous and rather close together, suberect or spreading, narrowly cylindrical, shortly stalked, 2—3-5 cm long, 0-4—0-6 cm wide; bracts 3—5, spreading acuminate, 4-7 mm long, 2-3 mm wide, green above, sericeous below; catkin-scales oblong-ovate, acute, about 1-8 mm long, 0-8 mm wide, fuscous except towards base, clothed with long, silvery-silky hairs; ovaries crowded, narrow-ovoid, tapering to apex, 2-2-5 mm long about 1-5 mm wide, shortly stalked, densely subappressed silvery-silky; nectary oblong, truncate, about 0-8 mm long, 0-4 mm wide, style distinct, slender, about 1-5 mm long; stigmas 4, filiform, to about 0-8 mm long, recurving at maturity. Male catkins and seeds not seen. ECOLOGICAL AND DISTRIBUTIONAL NOTES The first specimen was collected in July 1993 in Slack 65 in the north-western corner of Ainsdale Sand Dunes National Nature Reserve, at SD/295.118 by Michael Gee, Site Manager, and N.A.R. The slack is situated at the seaward edge of the fixed dunes, just behind the more mobile frontal dune ridges. The plant community in the slack is known in the National Vegetation Classification (Rodwell 1991-1996) as “Salix repens-Calliergon cuspidatum damp slack”, in which the most conspicuous plant is the leggy, creeping Salix repens and the most consistent plant on the ground surface is the moss Calliergon cuspidatum (Hedw.) Kindb. Other plants present, typical of damp slacks, are Mentha aquatica L., Equisetum fluviatile L., Eleocharis palustris (L.) Roem. & Schult., Epilobium palustre L. and Lythrum salicaria L. The water table varies greatly and the slack is usually flooded during the winter, sometimes into the spring, but can be quite dry in summer. Collections of willows in the slack taken in 1996 and 1997, det. R.D.M., were found not to contain S. X angusensis, but established that although the willow in the slack was mainly S. repens var. argentea, S. cinerea subsp. oleifolia was also present as scattered, stunted bushes, and as a 4 m tall clump at the edge of the slack. The second gathering, which was determined as S. x angusensis, was made on 10 August 1998 from a group of six bushes within the stand of S. repens var. argentea, centred about 7 m south of the tall S. cinerea subsp. oleifolia. The bushes had the same sprawling growth with much horizontal branching, as the surrounding S. repens var. argentea, and, at 0-7 m, were about the same height. In consequence they did not stand out from the rest and were not visible from the edge of the slack. They could be distinguished form the SALIX Xx ANGUSENSIS 329 FIGURE 1. Salix x angusensis Rech. f. del. P. Halliday © 2000 Drawn from N. A. Robinson 65 (cult. Ranscombe Lodge, Wootton Courtenay, Som.) Leaves 4 August 1995; catkins 15 April 1997. A: Foliage, B: Mature leaf, C: Stipules, D: 2 Catkins, E: 2 Flower. 330 R. D. MEIKLE AND N. A. ROBINSON surrounding willows by the longer leaves on the new growth — leaves on the old growth appeared to be much the same. The bushes looked old and broken down, but new growth was coming from the base and also new shoots could be seen round about, presumably from suckers. The fact that they were growing in a compact group suggested that they had arisen from a single initiate, within the stand of S. repens var. argentea. The third parent involved in the S. x angusensis triple hybrid, namely S. viminalis, is known to be present in the reserve, though not in the immediate vicinity, and the rare hybrid S. x friesiana Anderss. (S. repens X S. viminalis) is widespread and common in the v.c. 59 dunes. Some distance away, in the same slack, there is a clump of S. x doniana Sm. (S. purpurea X S. repens). This hybrid was not known outside Scotland until it was found on this part of the coast in 1947 (Savidge et al., 1963). It is present at several locations in the Reserve, and has been found at some other places in the coastal dunes. These willows are in parts of the Reserve which are not generally open to the public. Anyone wishing to see them, or to take samples, should contact the Site Manager, Ainsdale Sand Dunes National Nature Reserve, 2 West End Lodge, Pinfold Lane, Southport. ACKNOWLEDGMENTS We are grateful to Mr Michael Gee and to Mr Tony Duckels for their help in searching for S. x angusensis; also to Messrs Roy Vickery (Natural History Museum) and Douglas McKean (Royal Botanic Garden Edinburgh), and Miss Rachel Benvie (Montrose Museum and Art Gallery) for assistance in our pursuit of the missing type material, and our thanks to Dr Emst Vitek (Naturhistorisches Museum, Botanisches Abteilung, Vienna) for tracing the missing specimens and supplying photocopies. Our thanks also to Miss P. Halliday for the illustration which accompanies this paper. REFERENCES RECHINGER, K H. (1950). Observations on some Scottish Willows. Watsonia 1(5): 275. RODWELL, J. S., ed. (1991-1996). British Plant Communities, 14, Cambridge University Press, Cambridge. SAVIDGE, J. P.. HEYwoop, V. H. & GORDON, V. (1963). Travis’s Flora of South Lancashire. Liverpool Botanical Society, Liverpool. STACE, C. A. (1975). Hybridization and the Flora of the British Isles, pp. 312-319. Academic Press, London. (Accepted July 1999) Watsonia 23: 331—334 (2000) 331 Discovery of male plants of Najas marina L. (Hydrocharitaceae) in Britain R. J. HANDLEY* and A. J. DAVY School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ ABSTRACT The dioecious, annual macrophyte Najas marina is represented in Britain only in the Norfolk Broads. Although it was discovered in 1883 (at Hickling Broad), only female plants had previously been recorded and all seed production was assumed to be apomictic. Following detailed study of the population biology of this species, the occurrence of male plants at Hickling and Upton Broads in the summer of 1998 is reported. Photographs of the inconspicuous male flowers and pollen are provided. Possible reasons for the plants remaining unreported for so long are discussed. The potential for sexual reproduction has implications for the conservation of this Red Data Book species in Britain. Keywords: Annual, aquatic macrophyte, dioecy, Holly-leaved Naiad, rarity. INTRODUCTION Najas marina (Holly-leaved Naiad) is a dioecious annual which grows completely submerged in shallow lakes (1-3 m deep), rooting extensively into the bottom sediments. In Britain, N. marina is only found in the Norfolk Broads and there only in a handful of lakes, where it forms dense beds and can predominate over large areas. N. marina has a long history in the British flora, with fossil evidence for populations being far more widespread during the last climatic optimum than at present (Godwin 1975). In 1883, Arthur Bennett found N. marina for the first time as a living plant (Bennett 1884, 1909) and since then there has been a well reported increase in the number of sites at which it has been found (Barry & Jermy 1953) although there have rarely been more than seven sites in any one year. Najas marina is listed as Vulnerable in the British Red Data Book and in the IUCN European Red Data Categories and is therefore specially protected under the Wildlife and Countryside Act 1981 (Wigginton 1999). Despite considerable interest in N. marina, arising from its rarity value in Britain and Europe, there has been little detailed knowledge of its autecology and life history. In particular, there have been no reports of male individuals of N. marina in Britain and, in the apparent absence of male flowers, there has been speculation that its prolific seed production is entirely apomictic (Stace 1997; Preston & Croft 1997). This paper reports that male plants do in fact occur in Britain and illustrates the features of male flowers. THE IDENTIFICATION OF MALE PLANTS The discovery of male plants of Najas marina was made while snorkel diving to collect plant material in Hickling Broad in East Norfolk (Nat. Grid ref: TG/418.209). Subsequently, plants bearing male flowers were found at Upton Broad, another well established site for this species. Hickling Broad is the largest of the Norfolk Broads and the site at which Bennett first recorded N. marina (Bennett 1884). During 1998, Hickling and Upton Broads were visited at fortnightly intervals throughout the year in order to survey the populations of N. marina. In June and July, only plants bearing female flowers were observed; despite meticulous examination of a considerable quantity of plant material, male flowers could not be found. Plants apparently had only female flowers or no floral *Author for correspondence, e-mail: r.handley @uea.ac.uk 332 R. J. HANDLEY AND A. j. DAVY structures at all. However, in August 1998 male flowers were identified for the first time while diving. These flowers were evident for only a few weeks. The dense beds, with near-continuous cover, and the branching growth form of Najas marina at Hickling and Upton Broads combine to create considerable difficulty in distinguishing individual plants. The bases of the stems often become covered in sediment and root at the covered nodes making the identification or extraction of individual plants impossible. Nevertheless it was clear that the ratio of female to male flower bearing stems in these beds was high, perhaps as high as 10:1, and that male stems occurred in distinct patches, suggesting that isolated, individual males were surrounded by females. With some care it was possible to collect entire stems of N. marina. Examination of these revealed that each stem bears either exclusively male flowers or exclusively female flowers. As an individual plant may comprise several stems arising from the initial node it is not possible to say categorically that all the stems of an individual plant were of the same sex. However, there is no evidence from work on these populations to query the traditional view that N. marina 1s dioecious. FLORAL MORPHOLOGY The flowers of Najas marina arise singly, or sometimes several together, on a fertile shoot with suppressed internodes within a sheath axil (Rendle 1899). The male flower is ellipsoid, c. 2—4 mm long and 1—2 mm wide at the widest point. It possesses a thin membranous spathe which forms a tight envelope about the flower and protrudes above it to form a narrow crown with characteristic spines (Fig. la). The four locules of the single anther are discernible within the perianth (Fig. 1b, lc). The locules are densely packed with pollen grains, which are for the most part ellipsoid, but some are spherical (Fig. 1d). The morphology of these structures conforms well with the classical literature on the Najadaceae (Bailey 1885; Campbell 1897; Rendle 1899). Initial field observations were of male flowers that appeared conspicuously pale in colour, apparently because the white pollen grains were visible through the anther wall and _ thin, membranous perianth. Subsequently, darker male flowers were found. The pale flowers were close to dehiscence, with mature pollen. In these flowers the pollen-filled anther was particularly visible, because elongation of the filament prior to dehiscence caused the anther to split the spathe from the apex downwards (Rendle 1899). Flowers were also found in which the anthers had clearly dehisced, leaving the anther filament and the vestiges of the four locules. Male flowers are similar in size and shape to the maturing fruits of the female plant, and therefore they can be difficult to distinguish in the field. The phase during which the male flowers have a pale colour provides a reliable field diagnostic feature but this was evident only during August. DISCUSSION Given the small size and superficially similar appearance of male flowers to the developing fruits in Najas marina, it is perhaps not surprising that male plants of this dioecious species have gone unrecorded for so long in the British flora. The annual “window” during which the sexes are reasonably distinct appears to be narrow: outside this period, only very close scrutiny would reveal male flowers still within the spathe or the remains of dehisced anthers. In addition, the use of diving equipment and the unusually good water clarity in 1998 would have improved the chances of discrimination. Recently the macrophytes of the Norfolk Broads have been the subject of intensive study (Schutten et al. 1997; Kennison et al. 1998) without male plants of N. marina being reported; it is possible that the high ratio of female to male plants requires a sampling method more sensitive that the double-headed rake commonly used. We cannot be sure, however, that male plants have always been present in the British population and therefore have simply been overlooked previously. An examination of N. marina herbarium material at K did not reveal male flowers that might have been overlooked by previous collectors. However the preservation in herbarium specimens of the minute flowers of N. marina is poor. There are alternative explanations. Sex expression in plants can be fickle and subject to strong environmental influences (Richards 1997). There may be an influence on N. marina that could make the expression of maleness a rare or uncommon event, or may have resulted in males in the British populations for the first time recently. The origins of the current British populations are not MALE PLANTS OF NAJAS MARINA 333 200 um a os ™ . : rf : - : i. ~ ie f - A t = FIGURE 1: (a) Photograph showing the male flower of Najas marina with spathe protruding above the anther to form characteristic spines, a second immature male flower can be seen to the left obscured by the leaf. (b) Photograph showing anther of N. marina with perianth wall removed but still visible, two locules of the anther can be seen through the anther wall. (c) Photograph showing the anther of N. marina with perianth and anther wall removed and one locule of the anther punctured spilling pollen. (d) Photograph of pollen grains of N. marina showing mostly elongate and some spherical grains. 334 R. J. HANDLEY AND A. J. DAVY known: it may have gone unrecorded for many centuries before Bennett’s discovery in 1883 or it may have been a relatively recent colonist from continental European populations, transported in the guts of migrating wildfowl (Agami & Waisel 1986). It is possible that there have been several or repeated colonization events, giving rise to plants of different gender by chance or under environmental control. Najas marina in Britain is an obligate annual; therefore, even a temporary deficiency of males requires an alternative, apomictic mechanism for seed production. In 1998 fruit formation began well before the first male flowers were observed in the field, which suggests the possibility of facultative apomixis, with seed production early in the summer being apomictic and later seed production at least partially sexual (Asker & Jerling 1992). At present it is impossible to say why male plants of Najas marina have not been recorded before. Perhaps the most parsimonious explanation is that male flowers simply went unnoticed until now. More significant is the fresh light shed on the nature of its breeding system, which is of fundamental importance in determining the genetic variability and survival potential of the British populations. The demonstrable occurrence of males gives a new perspective to our strategy for conserving this rare species. ACKNOWLEDGMENTS The collection of Najas marina was carried out by license granted under the Wildlife & Countryside Act 1981. We thank the Biotechnology and Biological Sciences Research Council (BBSRC), The Broads Authority, The Environment Agency and English Nature (Species Recovery Program) for financial support, the Norfolk Wildlife Trust (Hickling) and Norwich Union Fishing Club (Upton) for use of boats while surveying, and the Curator of K for access to herbarium material. REFERENCES AGAMI, M. & WAISEL, Y. (1986) The role of mallard ducks (Anas platyrhynchos) in distribution and germination of seeds of the submerged hydrophyte Najas marina. Oecologia 68: 473-475. ASKER, S. E. & JERLING, L. (1992) Apomixis in plants. CRC Press, Florida, USA. BAILEY, C. (1885) On the structure, the occurrence in Lancashire and the probable source, of Naias graminea, Delile, var. Delilei,. Magnus. Memoirs of the Manchester Literary and Philosophical Society: Third Series 10: 29-75. BARRY, D. H. & JERMY, A. C. (1953) Observations on Najas marina. Transactions of the Norfolk and Norwich Naturalists’ Society 17: 294-297. BENNETT, A. (1884) Plants new to Norfolk with notes on other species. Transactions of the Norfolk and Norwich Naturalists’ Society 3: 633-636. BENNETT, A. (1909) Najas marina, L., and Chara stelligera, Bauer, as Norfolk plants. Transactions of the Norfolk and Norwich Naturalists’ Society 9: 47-50. CAMPBELL, D. H. (1897) A morphological study of Naias and Zannichellia. Proceedings of the California Academy of Sciences: Third Series Botany 1: 1-61. GODWIN, H. (1975) The history of the British flora, 2nd ed. Cambridge University Press, Cambridge. KENNISON, G. C. B., DUNSFORD, D. S. & SCHUTTEN, J. (1998) Stable or changing lakes? A classification of aquatic macrophyte assemblages from a eutrophic shallow lake system in the United Kingdom. Aquatic conservation: marine and freshwater ecosystems 8: 669-684. PRESTON, C. D. & CROFT, J. M. (1997) Aquatic plants in Britain and Ireland. Harley Books, Colchester. RENDLE, A. B. (1899) A systematic revision of the Genus Najas. Transactions of the Linnean Society of London: Second Series Botany 5: 379-444. RICHARDS, A. J. (1997) Plant breeding systems, 2nd ed. Chapman & Hall, London, England. SCHUTTEN, J. et al., (1997). Factors affecting water plant recovery, Technical Report for EC LIFE Project No. 92—93/UK/031, BARS 14c/P-89. Broads Authority/Environment Agency, Norwich. STACE, C. A. (1997) New Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. WIGGINTON, M. J., ed. (1999) British Red Data Books. 1. Vascular plants, 3rd ed. Joint Nature Conservation Committee, Peterborough. (Accepted August 1999) Watsonia 23: 335-348 (2000) 385 Notes CORYNEPHORUS CANESCENS (L.) P. BEAUV. (POACEAE) AT KINVER EDGE, STAFFORDSHIRE: A RE-ASSESSMENT In his review of the distribution and status of Corynephorus canescens (L.) P. Beauv. in Britain, Trist (1998) describes the site at Kinver Edge, Staffordshire, which holds one of the few British inland colonies of this grass, and states that “the population of c. 550 plants has remained fairly constant since 1977.” Though his site description is reasonably accurate, Trist’s estimate of the population size of C. canescens fails to take account of its recent dramatic increase following sympathetic site management. This note, therefore, aims to update Trist’s account and trace the effects on the C. canescens population of conservation measures taken over the last twelve years. BACKGROUND HISTORY C. canescens was discovered on the National Trust’s property at Kinver Edge by I. C. Trueman in the 1970s. Following the appointment of one of the authors (M. E. Blunt) as Warden in 1986, the Trust has pursued active measures to conserve and increase this population. In 1986 C. canescens was present on two areas some 300 m apart near the south-eastern boundary of the property, where the dip-slope of the escarpment inclines gently to the south-east. Site 1 consisted of a bowl of bare sand of approximately 200 m’, heavily trampled by the public; C. canescens survived here as a narrow rim around the bowl. Site 2 was a flatter area of about 140 m*, where the grass was reasonably well established. Trist’s (Joc. cit.) estimate of c. 550 plants is not unreasonable for the size of the population at this date. The two Corynephorus sites were bounded on the south-east by a sandy track and fence, beyond which lay a grassy field in private ownership; this field received periodic treatments of fertilizer and was cut for sileage up to 1986 but not thereafter. Professor Trueman (pers. comm.) has stated that a small quantity of C. canescens was present in this field when the main colonies were discovered. On its other sides the C. canescens locality was surrounded by birch/oak/pine scrub encroaching extensively on lowland Calluna vulgaris (L.) Hull heath. MANAGEMENT SINCE 1986 With C. canescens fairly well established on Site 2, priority in 1986 was given to halting the erosion of Site | and encouraging the expansion of the species across the eroded area. Initially barriers of pine brashings were constructed around the rim of this site to protect existing plants from trampling; but after three seasons there was no noticeable increase in the size of the C. canescens colony. Therefore, in winter 1989/90 pine brashings were spread thinly across the open sand, their branches barely touching to allow good penetration of light. This technique largely eliminated trampling and within a season C. canescens seedlings had colonised the entire area of open sand. At this point the pine brashings were removed, the grass continued to flourish, and public pressure on this site has remained light ever since. The main problem with maintaining C. canescens populations at Kinver Edge is the presence of mat-forming Polytrichum mosses which rapidly establish themselves on sand stabilised by the grass. Polytrichum is followed by a succession of other species, notably Rumex acetosella L., Deschampsia flexuosa (L.) Trin., Festuca rubra L. and Hypochaeris radicata L. After some three years seedlings of Calluna vulgaris and Ulex europaeus appear, and within six the C. canescens population shows some deterioration. Therefore, in the mid-1990s an experiment to reverse this succession by ploughing the whole of Site 1 with a Dutch harrow was tried, with successful results. The moss thatch was broken up, fresh sand was turned over and C. canescens re-colonised from plants which survived in the gaps between the harrow’s blades. Site 2 has had no direct management since 1986 except for regular light trampling from visitors using the adjacent path, and since 1997 from cattle introduced to graze the surrounding heath. This 336 NOTES Watsonia 23 (2000) trampling sufficiently disturbs the sand to allow a few new plants of C. canescens to establish themselves; but on the whole the population of Site 2 is deteriorating slowly and gorse scrub has occupied much of the site. In 1994 a strip of sandy track c. 30 x 4 m parallel to Site 2 and 3 m from it was fenced off and covered with pine brash as described above. Here, too, Corynephorus seedlings covered the entire area within a season and now form a closed sward. This is Site 3 in Table 1 below. In the early 1990s ownership of the field next to the Corynephorus sites passed to the National Trust, since when light grazing by sheep and cattle has been introduced to encourage the establishment of heathland communities. By 1997 it was evident that C. canescens was spreading in two areas of this field adjacent to Sites 1 and 2: these are Sites 4 and 5 listed in Table 1. Other C. canescens populations at Kinver Edge consist of a small area just to the west of Site 1 (numbered Site 6 in Table 1) and individual plants scattered widely along track sides and among the Calluna heath near the main populations. ESTIMATES OF THE CURRENT POPULATION In 1997 and 1998 counts were made of the whole and parts of the Corynephorus sites on Kinver Edge in order to estimate the total population. In summer 1997 M. E. Blunt, assisted by S. Anderson, ran four transect lines across Site 4, finding C. canescens up to 13 m into the field at its furthest point of colonisation. Quadrat counts were made at 2 m intervals along the transects. In July 1998 M. E. Blunt, assisted by S. Fereday, measured the dimensions of all the Corynephorus sites and made population estimates from quadrat counts on all sites. In September 1998 the two authors made further estimates of the numbers of plants in the three main populations (Sites 1, 2 and 3) from random quadrats. The results of these counts are shown in Table 1. The September 1998 estimates were made the day after a severe thunderstorm washed away parts of Site 1, doubtless accounting for the discrepancy between the two estimates for this Site; being on more level ground Sites 2 and 3 were little affected by the storm. The July 1998 estimate for Site 4 was taken over a larger area than that of 1997, since C. canescens was found up to 21 m into the field in 1998: the area given in the Table for this site is that calculated in July 1998. It may be deduced from the Table that the total population of C. canescens at Kinver Edge currently lies between 40,000 and 50,000 plants. CONCLUSIONS In its Kinver Edge stations, C. canescens is the pioneer colonist of open sand; it appears to require a fairly stable surface to establish itself successfully, and seems unable to do so where sand is regularly disturbed, for example by trampling or the activities of rabbits (Oryctolagus cuniculus):; though a few plants do establish themselves on the less disturbed edges of rabbit warrens. Once C. canescens is established, succession of the habitat to gorse (Ulex europaeus) and heather (Calluna vulgaris) scrub begins within three years and C. canescens populations begin to decline within six years unless there are further exposures of bare sand. In short, at Kinver Edge TABLE 1: ESTIMATED NUMBERS OF CORYNEPHORUS CANESCENS PLANTS ON KINVER EDGE Area (m°) 1997 July 1998 Sept. 1998 Site 1 202 ; 19,400 12,700 Site 2 145 . 8,100 8,700 Site 3 129 : 10,900 11,000 Site 4 90 1,400 3,900 - Site 5 40 : 500 - Site 6 =15 : 3,900 - Scattered - - 1,600 - NOTES Watsonia 23 (2000) 337] 2 LA; 5 ip Co ¢. oF, 7. Lasy UN blepy =e — ~ Le —) 7 an 3 Corynephorus = \ ifs canescens = :~\v? sites 36 heathland oo NT Warden's &? Lodge = A Figure |. Plan of Kinver Edge, Staffordshire. Scale approx. 1:2500 C. canescens exhibits a narrow niche requirement, viz. periodic disturbance of sand followed by a cessation of disturbance over a cycle of 3-6 years. Current studies suggest that trampling, wind- blow, rainstorms and the activities of rabbits may, depending on their frequency, both create conditions favourable to C. canescens and also destroy them. Where conditions remain unfavourable over a period, C. canescens seems able to survive as individual plants scattered along the edges of paths, particularly in the lee of overhanging vegetation where they receive little trampling. Observations further suggest that seed production is overwhelmingly the means whereby C. canescens colonises new ground at Kinver Edge; once established, it may continue to propagate itself either from seed or vegetatively. This contrasts with Trist’s (loc. cit.) studies which suggested vegetative means as the grass’s chief method of propagation in Britain. The role of pine brashings in encouraging germination is of interest: besides reducing trampling they may also protect tiny plants from frost and drought, considered by Marshall (1967) as critical factors in seedling survival. Observations also indicate that C. canescens is not selectively grazed at Kinver by sheep, cattle or rabbits, and seeds freely at the grazing intensities encountered there. The authors’ studies of the ecology of C. canescens at Kinver Edge and its response to various management techniques form part of an ongoing programme. It was, however, felt worthwhile bringing current findings to the attention of those interested in this species and its conservation, as well as updating the published record of what now appears to be a major population of this scarce grass in Britain. ACKNOWLEDGMENTS The authors wish to acknowledge the help and advice of Professor I. C. Trueman of the University of Wolverhampton throughout these studies, and for his comments on the draft of this paper. 338 NOTES Watsonia 23 (2000) REFERENCES MARSHALL, J. K. (1967). Biological Flora of the British Isles: Corynephorus canescens (L.) Beauv. Journal of ecology 55: 207-220. TRIST, P. J. O. (1998). The distribution and status of Corynephorus canescens (L.) P. Beauv. in Britain and the Channel Islands with particular reference to its conservation. Watsonia 22: 41-47. A. G. BLUNT University of Wolverhampton, Molineux Street, Wolverhampton WV1 ISB M. E. BLUNT Warden’s Lodge, Comber Road, Kinver, Stourbridge, West Midlands DY7 6HU A REANALYSIS OF THE MISTLETOE (VISCUM ALBUM L.; LORANTHACEAE) SURVEY DATA FROM THE 1970S AND 1990S In 1993, the late John Knight and I, exasperated by the Plantlife 1993 Christmas card depicting unseasonal Bluebells, decided to find a more appropriate species for the following year, and had the idea of repeating the 1970s B.S.B.I. Mistletoe (Viscum album L.) survey reported by Perring (1973). Independently, Jonathan Briggs had also already been investigating the possibility of repeating the survey. Discussions between the two societies resulted in a joint Plantlife/B.S.B.I. survey in the mid 1990s whose main aim was to determine if Mistletoe was declining. The 1994 Plantlife Christmas card included the Mistletoe survey form. For the 1969-1972 survey, data on the occurrence of Mistletoe and its hosts were requested from the B.S.B.I., the Northern Horticultural Society and the general public throughout Britain (Perring 1973). For the 1990s survey, records were again collected from throughout Britain by B.S.B.I. and Plantlife members and the general public, and the survey received extensive press coverage (Briggs 1999). When the two surveys were compared, there was large increase in records since the 1970s. Briggs (1999) was unable to draw any conclusions about possible declines or increases because of the uncertainty of whether there had simply been more records sent in for the 1990s survey (i.e. the increase in records was a result of more recording). I had previously had exactly the same problem with the B.S.B.I. Monitoring Scheme data and have spelt it out in some detail (Rich & Woodruff 1990, 1992, 1996). Fortunately, some further analysis of the data presented by Briggs (1999) is possible. Crucially, the tetrad map for England and Wales (Perring’s Fig. 7) had areas marked with hatched lines where Mistletoe had been searched for “intensively but had not been found”. By assuming that the areas had indeed been surveyed comprehensively during the 1970s it is possible to assess statistically the probability of any significant increases or decreases in these areas. METHODS The number of tetrad records were counted for areas marked with hatched lines in Perring (1973) and for the same areas from the maps in Briggs (1999). The poor reproduction of the map in Perring (1973) meant that determining the limits or exact locations of some hatched areas was very difficult and areas where there was any doubt were rejected. Data for 39 discrete hatched areas which range in size from groups of four tetrads to whole counties representing a total of 2817 tetrads (c. 7% of the tetrads in England and Wales) were available, excluding the large area of Gloucestershire and Herefordshire. The differences in the number of tetrads were tested for significance using a two-tailed test of equality of the two percentages (Sokal & Rohlf 1969) with 95% confidence limits. RESULTS There were 468 tetrad records in the 1970s hatched areas (16-6% of tetrads surveyed intensively), and 562 tetrad records in the same areas in the 1990s (19-95% of tetrads surveyed intensively in NOTES Watsonia 23 (2000) 339 the 1970s). This is a significant increase (p<0-001). Of the discrete hatched areas marked as intensively recorded in the 1970s, Mistletoe has increased in frequency in 22 areas, showed no change in six areas, and declined in eleven areas. DISCUSSION The re-analysis depends on the assumption that the hatched areas marked in Perring (1973) have been surveyed comprehensively. Although Perring indicated areas which had been surveyed sufficiently “intensively” to be reasonably certain that Mistletoe was indeed absent in the 1970s, I have doubts whether hatched areas as large as Surrey were really surveyed thoroughly. Indeed the map of the 1970s survey in Briggs (1999) includes more records showing this to be the case (some from the extension of the survey into 1973). None-the-less, it is reasonable to accept Perring’s hatched areas at face value as a reasonably thorough baseline survey. Briggs (1999) rightly makes no claim that any areas had been surveyed intensively. As it is likely that on average the hatched areas were less intensively recorded in the 1990s than the 1970s, despite the greater number of people participating in the survey, if there was no change or a decline in Mistletoe there should be fewer records. The fact that there are significantly more records despite a less intense survey allows a firmer conclusion to be drawn that there is no evidence for decline for tetrad frequency data in these areas. It is not possible to determine whether Mistletoe has increased or stayed the same because the intensity of the survey in the 1990s compared to that in the 1970s is not known. The inclusion of the Gloucestershire/Herefordshire area, where there is a large apparent decline, may also affect this conclusion. Assuming that the hatched areas are a representative random sample (they are scattered across the country though with local concentrations), it is possible to extend the conclusion to no evidence of decline at a national level for tetrad frequency data. Rich & Woodruff (1990, 1996) found no statistically significant change in frequency of Mistletoe at a 10-km square level between 1930-1960 and 1987-1988. A more appropriate form of survey to determine change than an ad hoc national survey would have been to re-record the hatched areas intensively. Alternatively, a properly designed sample baseline survey including estimates of recorder effort could have been established against which future changes could have been assessed more accurately. The application of standardised survey techniques, as we have for Ashdown Forest (Rich er al. 1996; Rich 1998), would minimise these problems. Once again, the problems of recording bias may have masked real changes in wild plants in the countryside. REFERENCES BRIGGS, J. (1999). Kissing goodbye to Mistletoe? The results of a national survey aimed at discovering whether mistletoe in Britain is in decline. Plantlife, London. PERRING, F. H. (1973). Mistletoe in GREEN, P. S., ed. Plants: wild and cultivated. B.S.B.I. Conference Proceedings 1972. pp. 139-145. E. W. Classey, London. RICH, T. C. G. (1998). Squaring the circles - bias in distribution maps. British wildlife 9: 213-219. RICH, T. C. G. et al. (1996). Flora of Ashdown Forest. Sussex Botanical Recording Society, East Grinstead. RICH, T. C. G. & WOODRUFF, E. R. (1990). B.S.B.I. Monitoring Scheme 1987-1988. Chief Scientist’s Directorate Report No. 1265. Nature Conservancy Council, Peterborough. RICH, T. C. G. & WOODRUFF, E. R. (1992). The influence of recording bias in botanical surveys: Examples from the B.S.B.I. Monitoring Scheme 1987-1988. Watsonia 19: 73-95. RICH, T. C. G. and WOODRUFF, E. R. (1996). Changes in the floras of England and Scotland between 1930- 1960 and 1987-1988: The B.S.B.I. Monitoring Scheme. Biological conservation 75: 217-229. SOKAL, R. R. & ROHLF, F. J. (1969). Biometry. W. H. Freeman, San Francisco. T. C. G. RICH Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP 340 NOTES Watsonia 23 (2000) THREE NEW BRITISH SITES FOR CAREX DEPAUPERATA WITH. (CYPERACEAE) REPRESENTED IN THE IRISH NATIONAL HERBARIUM, GLASNEVIN In the Irish National Herbarium at Glasnevin Botanic Garden, Dublin (DBN), there are four herbarium specimens of Carex depauperata from previously unreported localities in Britain, one of which is of doubtful origin. V.C. 17 SURREY The first specimen is labelled ‘near Farnham, Surrey, August 1846, ex herb. W. Mclvor’. William Graham Mclvor (?-1876) collected a number of plants in Surrey, but does not seem to have communicated the records to the Surrey botanists of the time (e.g. J. D. Salmon or J. A. Brewer). There is plenty of suitable woodland on calcareous soils in this area (e.g. Hog’s Back), not far from the well-known Godalming sites, and this record is accepted. The second specimen is more problematic. It is labelled ‘Leith Hill, Surrey, 30.6.1890, per J. Leitch’. The fact that the label is not in Leitch’s hand-writing, the potential for confusion of the popularly-known ‘Leith Hill’ with the well-known C. depauperata locality at Frith Hill, Godalming, and the predominantly acidic nature of soils on Leith Hill, suggests the specimen may be mislabelled. John Leitch (1849-1896) appears to have collected widely in Kent and Cumbria, but there are few Surrey specimens. His herbarium at Carlisle (CLE) contains two sheets cultivated at Bournemouth on 21 June 1890 (presumably by E. F. Linton), and another from the Royal Botanic Gardens, Edinburgh, June 1871. This record is not accepted. V.C. 83 EDINBURGH A third specimen is labelled ‘Wood above Bonelly (ex. herb. W. R. McNab)’. We interpret the locality as Bonally (sometimes spelt Bonaly), near Edinburgh, grid reference c. NT/21.67, now partly a country park. Dr William Ramsay McNab was born in Edinburgh in 1844. He worked at the Royal Botanic Garden, Edinburgh, lectured in botany in Dumfries in 1869 and in Cirencester from 1870, and became the Scientific Supervisor at the Glasnevin Botanic Gardens in 1880, where there are many of his specimens (Nelson 1990; in press). He died in Dublin in 1889. The handwriting on the herbarium sheet is different to other McNab specimens and it may not have been collected by him, but his local Edinburgh connection reinforces our interpretation of the site. This record is accepted, though it 1s conceivable there may be links between Bonally, McNab and material grown in Edinburgh (cf. above). V.C. 90 FORFAR The final specimen is labelled ‘Forfarshire, T. McFarlane (ex herb. Rev. H. G. Carroll)’ but is undated. We have been unable to trace anything about T. McFarlane to give us any more direct clues about its origin. However, the Rev. Henry George Carroll was an Irish botanist born in c. 1810 and died in 1902, so the specimen was probably collected sometime before 1900. Carroll had obtained other notable specimens from a range of collectors and his herbarium sheets, with their characteristic taxonomic headings and handwriting, represent a significant component of DBN. We would like to know more about his possible connections with Scottish botanists. The specimen also casts new light on the record in Hooker (1821) ‘Woods near Forfar, G. Done accepted by Gardiner (1848) but rejected as doubtful by Ingram & Noltie (1981) as it could have originated from Don’s garden (the source of several other confusing records). We believe this apparently independently-collected specimen establishes C. depauperata as a Forfarshire species. There are no other specimens collected in Scotland in ABD, DEE, E, GLAM or PTH. This now brings the total number of sites for C. depauperata in the British Isles to 14. Although the British collections in DBN are relatively poorly known, it is none-the-less surprising that at least three new records should turn up in the same herbarium folder. We hope that these specimens lead to further information and, above all, to the rediscovery of C. depauperata in these sites. ACKNOWLEDGMENTS We are grateful to David Allen, Richard Brinklow, Stephen Hewitt, Douglas McKean, Henry Noltie, Mark Simmons, Joyce Smith and Keith Watson for their help. NOTES Watsonia 23 (2000) 341 REFERENCES GARDINER, W. (1848). The flora of Forfarshire. Longman, Brown and Green, London. HOOKER, W. J. (1821). Flora scotica. Archibald Constable & Co, Edinburgh. INGRAM, R. & NOLTIE, H. J. (1981). The flora of Angus (Forfar, v.c. 90). Dundee Museums and Art Galleries, Dundee. NELSON, E. C. (1990). William Ramsay McNab’s herbarium in the National Botanic Gardens, Glasnevin (DBN). I. Its early history and acquisition. Glasra 1: 1-7. NELSON, E. C. (in press). William Ramsay McNab’s herbarium in the National Botanic Gardens, Glasnevin (DBN). II. A catalogue of specimens from the Royal Gardens Kew (1805-1810) relating to Hortus Kewensis (edition 2), the Botanic Garden, Edinburgh (c. 1815) and Scottish native species. Glasra. Ts G.G. Rico Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP R. A. JONES Countryside Council for Wales, Plas Gogerddan, Aberystwyth, Dyfed SY23 3EE M. JEBB Herbarium, National Botanic Gardens, Glasnevin, Dublin 9, Ireland ELEOCHARIS PARVULA DISCOVERED IN SCOTLAND On 29 July 1999 we visited the Cromarty Firth S.E. of Pitglassie, Dingwall, E. Ross (v.c. 106), where the River Conon flows in a number of channels separated by low-lying islands with marshes and pools. The marshes and pools contain some species characteristic of fresh water and others which typically occur in more brackish conditions. When we arrived at high tide the channels were full, but as we left we could see that the tide had gone out to reveal extensive mud banks beside them. It was while looking at one of these banks, at grid reference NH/553572, that we noticed a small Eleocharis which turned out on further examination to be E. parvula (Roem. & Schult.) Link ex Bluff, Nees & Schauer. This species has hitherto been recorded in Britain on the coasts of southern England (Devon, Dorset, S. Hampshire) and North Wales (Merioneth, Caernarvonshire). The occurrence on the Cromarty Firth therefore represents the first record for Scotland and an extension of its British range some 500 km northwards. The identification has been confirmed by S.M. Walters and voucher specimens have been deposited in CGE and E. The Eleocharis parvula population at Pitglassie grew on both sides of the channel in a substrate of fine mud, or in predominantly stony areas with mud in the interstices of the stones. On the date of our visit it would have been submerged under 50-75 cm of water at high tide. In places it grew under a very sparse stand of Bolboschoenus maritimus, but the most frequent associate was Callitriche stagnalis, which grew as small, prostrate, fruiting plants which were abundant on the mud banks but did not form a complete cover. Other associates were very occasional plants of Armeria maritima, Cochlearia officinalis and Triglochin maritimum. In late July we could find only vegetative stems, and there was still no sign of flowering when Stewart Angus, Ro Scott and Peter Wortham revisited the site on 31 August 1999, when Peter Wortham returned on 30 September or when Ro Scott went again on 22 October 1999. Even in July the characteristic white turions of E. parvula were present, albeit sparingly, on the slender rhizomes. By the end of September the plants were described as yellowing. The habitat of Eleocharis parvula at its English and Welsh sites has recently been described by Byfield (1999). As in the Cromarty Firth, it grows “in large tidal rivers ... on firm bare muddy substrates ... close to the upper limits of tidal influence, avoiding the strongly saline conditions associated with many saltmarshes.” However, the occurrence of FE. parvula with frequent Callitriche stagnalis does not appear to have described elsewhere in Britain: no Callitriche species are recorded with Eleocharis parvula by Rhind & Jones (1994) in their account of the E. parvula saltmarsh community at Glaslyn Marsh, Merioneth, for example. Indeed, we have not traced any description in the standard British sources of tidal mud bank communities dominated by Callitriche species. Our very limited experience suggests that these are a feature of the large estuaries in N.E. Scotland: on 2 August 1999 we visited the River Beauly E.S.E. of Tomich House, 342 NOTES Watsonia 23 (2000) NH/54.47., where C. stagnalis formed an abundant green sward on the sloppy mud banks and flats alongside the large single channel of the river. Rodwell (1995) does not describe these communities, though he mentions the occurrence of Callitriche stands in temporarily or seasonally flooded sites such as tracks or woodland rides, the edges of fluctuating ponds or the summer-dry upper reaches of chalk streams. C.D.P. was similarly unaware of the tidal mud communities when preparing the habitat descriptions for Callitriche species in Preston & Croft (1997). However, R. M. Burton (in litt., 1999) has since told us that he has seen C. stagnalis on estuarine mud sheltered by eyots by the River Thames in west London. The absence of flowers in the Scottish population of E. parvula in 1999 is consistent with its behaviour elsewhere in Britain and Ireland: Byfield (1999) notes that “in many localities the plant flowers and fruits very poorly’, and this certainly appears to be true of the more northerly populations: “the plant rarely flowers in Wales” (Rhind & Jones 1994) and the population by the River Bann in Co. Londonderry “apparently does not flower” (Praeger & Megaw 1938). Although the discovery of E. parvula in northern Scotland would not have been predicted from its known British distribution, it is not surprising when viewed in a wider, European perspective. It is perhaps most frequent in Europe around the coast of southern Sweden and the Baltic Sea, including the Gulf of Finland and the southern part of the Gulf of Bothnia, where it grows at higher latitudes than the recently discovered Scottish site (Hultén & Fries 1986; Mossberg 1997). Indeed, its northernmost sites are further north than Shetland. It is, as Byfield (1999) writes, “an inconspicuous plant that usually grows in an uninviting habitat” and it may well have been overlooked elsewhere in Scotland. ACKNOWLEDGMENTS We are grateful to B. Goater, G. M. Kay & D. A. Pearman for their help on fieldwork in Scotland, to R. Scott and P. Wortham for providing details of their later visits to the site, to R. M. Burton for his observations on estuarine Callitriche stagnalis, to S. M. Walters for checking our material and to D. A. Pearman for comments on a draft of this note. C.D.P. thanks Arto Kurtto, Pertti Uotila and Leena Helynranta, who demonstrated swards of Eleocharis parvula at Uusimaa, Gulf of Finland, on an Atlas Florae Europaeae excursion in 1997, thus increasing his awareness of this inconspicuous species. REFERENCES BYFIELD, A. J. (1999). Eleocharis parvula (Roemer & Schultes) Link ex Bluff, Nees & Shauer (Cyperaceae), in WIGGINTON, M. J., comp. & ed., British red data books I vascular plants, 3rd ed., p. 134. Joint Nature Conservation Committee, Peterborough. HULTEN, E. & FRIES, M. (1986). Atlas of north European vascular plants north of the Tropic of Cancer. 3 vols. Koeltz Scientific Books, K6nigstein. MOSSBERG, B., STENBERG, L. & ERICSSON, S. (1992). Den Nordiska Floran. Wahlstr6m & Widstrand, Solna. PRAEGER, R. L. & MEGAW, W. R. (1938). A Flora of the north-east of Ireland, 2nd ed. Quota Press, Belfast. PRESTON, C. D. & CROFT, J. M. (1997). Aquatic plants in Britain and Ireland. Harley Books, Colchester. . RHIND, P. M. & JONES, A. (1994). Brackish saltmarsh communities in the Glaslyn Marsh Trust Reserve. Field Studies 8: 373-384. RODWELL, J. S., ed. (1995). British plant communities, 4. Aquatic communities, swamps and tall-herb fens. Cambridge University Press, Cambridge. T. D. DINES Rhyd y Fuwch, near Bethel, Caernarfon, Gwynedd, LL55 3PS C. D. PRESTON Institute of Terrestrial Ecology, Monks Wood, Abbots Ripton, Huntingdon, Cambs. PE28 2LS NOTES Watsonia 23 (2000) 343 SUAEDA VERA FORSSK. EX J. F. GMEL. (CHENOPODIACEAE), SHRUBBY SEA-BLITE, PRESENT IN ANGLESEY (V.C. 52), WALES On the B.S.B.I. field meeting of 13 June 1999, T. C. G. Rich, M. and. J. Davidson and J. A. Green discovered a new locality for Suaeda vera Forssk. ex J. F. Gmel., Shrubby Sea-blite, on Anglesey (v.c. 52). Four large plants were found at the high water mark at the top of a stony beach on the north side of the eastern end of Black Creek on the Bodior Estate, Rhoscolyn, Holy Island (grid reference SH/293.757). The plants were to c. 1-3 m tall, were much branched at the base and evidently old, and were just coming into flower and hermaphrodite. They were tucked into a very sheltered rocky corner opposite a small stone fisherman’s hut and under low cliffs. It is no surprise that they have not been seen before as they are hidden from view except in the immediate vicinity and the site is on a private estate away from public footpaths. No further plants were found during additional searches of the coastline in this area in June and August. A side branch was collected as a voucher and has been deposited in the National Museum of Wales (NMW). Suaeda vera is generally regarded as native around the coasts of South-east England from Lincolnshire to Dorset, with occasional introductions with ships ballast in North-east England and South Wales (for a recent map, see Stewart et al. 1994). The Anglesey locality is highly disjunct from these other sites, but the information available to date does not suggest it was deliberately introduced by man. Other typical saltmarsh species occurred nearby including Juncus maritimus Lam. var. atlanticus J. W. White and Zostera angustifolia (Hornem.) Rchb., and no introduced species other than Spartina anglica C. E. Hubb. were present in this pristine area. There are no records of planting by the estate. No ripe seed was found even in August, and it is possible that the four plants have originated by layering (cf. Chapman 1947). We thus regard this locality as the first native site in Wales. We presume that the plants have either arisen from long distance dispersal by sea water or wildfowl, or are relict from a previously wider, more continuous distribution (it is certainly widely distributed in Europe, Asia, Africa, and in Atlantic islands such as Madeira, Canaries and St Helena, on the coast as well as inland). Ridley (1930) states that Suaeda seeds appear to sink very speedily, but quotes no data for S. vera. To provide more information on seed floatation, 30 ripe seeds were collected from bushes on Chesil Beach, Portland in September 1999 and placed in a beaker of sea water, with or without calyces depending on how they had fallen off the plants. After stirring to simulate wave action, the number of seeds floating were recorded at intervals for one month (Table 1). Most seeds sank within 5 days, but one seed floated for at least one month. No germination occurred in sea water. To test that the seeds were still viable after immersion in sea water for a month, they were rinsed in fresh water and placed on damp tissue paper in a petri dish and left to germinate. Five seeds (17%) germinated within two weeks but none thereafter. The poor flotation time suggests that plants are unlikely to have arrived at this site directly by sea water transport, and are more likely to have been brought by wildfowl which regularly winter on Anglesey. A number of other essentially maritime species with disjunct distributions also reach their northern limit on Anglesey including Juncus capitatus Weigel, Asparagus officinalis L. subsp. prostratus (Dumort.) Corb. and Rumex rupestris Le Gall. Frankenia laevis L., another south- TABLE 1. NUMBER OF SEEDS OF SUAEDA VERA FLOATING IN SEA WATER WITH TIME (N = 30). Days Number floating (%) 0 30 (100%) 3 20 (67%) 4 8 (27%) =) 5 (17%) 6 4 (13%) 10 1 (3%) 30 1 3%) 344 NOTES Watsonia 23 (2000) eastern species of saltmarshes and rocks with a virtually identical distribution to S. vera, was found nearby on Anglesey in 1965 (Roberts 1975); the discovery of the Suaeda suggests it could also be native. ACKNOWLEDGMENTS We would like to thank Frank Roberts and the Bodior Estate for access to the private land. REFERENCES CHAPMAN, V. J. (1947). Biological flora of the British Isles. Suaeda fruticosa Forsk. Journal of ecology 35: 303-310. RIDLEY, H. N. (1930). The dispersal of plants throughout the world. L. Reeve & Co., Ashford. ROBERTS, R. H. (1975). Frankenia laevis L. in Anglesey. Watsonia 10: 291-292. STEWART, A., PEARMAN, D. A. & PRESTON, C. D. eds. (1994). Scarce plants in Britain. Joint Nature Conservation Committee, Peterborough. i CsG: Rich Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF10 3NP N. BROWN Treborth Botanic Garden, University College of North Wales, Treborth, Bangor, Gwynedd LLS572RQ A PROBABLY NATIVE AND REGENERATING POPULATION OF TILIA PLATYPHYLLOS SCOP. IN BEDFORDSHIRE Evidence that Tilia platyphyllos is native in scattered localities in England and Wales was reviewed by Pigott (1981), where most of the sites discussed are on cliffs or steep slopes of hard limestone or volcanic rocks in the Welsh borders, Derbyshire and Yorkshire and where annual rainfall exceeds 800 mm. Since 1981 as many as 23 localities have been discovered on the chalk of the South Downs in west Sussex and Hampshire (Abraham & Rose; unpublished report 1998). These populations are usually of large old trees, or a mixture with younger trees, but there is almost no regeneration. Quite characteristically in this drier part of England the sites are at the foot of the north-facing escarpment of the South Downs where the trees grow close to the spring-line, often at the head of coombes and along their flanks. In 1993 Graham Bellamy, manager of the Barton Hills National Nature Reserve in Bedfordshire, told me that there were three large lime-trees in Leat Wood on the steep east-facing slope of the chalk coombe south of Barton-le-Clay (51°57'N 0°25'W; GR 52/08.29). Shoots were obtained from the emergent crowns by climbing the trees, and proved to be of typical Tilia platyphyllos Scop. subsp. cordifolia (Besser) Schneider. The upper part of Leat Wood is dominated by beech (Fagus silvatica), which was probably planted about 1830, but the steep slope is occupied by a mixture of ash (Fraxinus excelsior), sycamore (Acer pseudoplatanus) and horse chestnut (Aesculus hippocastanum). There is patchy regeneration of ash, beech and sycamore with scattered saplings of lime. The largest lime had a basal diameter of almost 3-0 m. As an indication of its age it may be noted that this is as large as the tree of T. platyphyllos which was planted in 1565 in the open and in a more fertile site at Burghley House, Northamptonshire, and was therefore 450 years old in 1995. Leat Wood was shown on the OS. map (1:10560 sheet 30NW) of 1891 (surveyed in 1880-81) and also on the pre- enclosure map of 1778, where it is called “The Lead’. Below Leat Wood, a stream flows northwards down the valley. Its source is two vigorous springs at the head of the lower coombe at 52/089.295, and beside them grows a tree of T platyphyllos in a typical natural habitat for the species. The tree originally had two trunks but in 1995 one was a stump and the other 18 m tall with a diameter at 1-3 m above the ground (dbh) of 1-09 m. A section cut from the stump gave a ring count of 173 but the outer 70 mm was decayed. NOTES Watsonia 23 (2000) 3.45 From the average ring width in the adjacent wood this represents another 30 years, to give an approximate age of 203 years. Having two trunks, the tree had probably been cut and regrown and was therefore older. Above the spring is the most remarkable feature of the site which is perhaps unique in Britain. On the steep north-east facing slope at 52/088.294 is an area of woodland containing 32 trees of T platyphyllos, mixed with ash and beech of similar size. Most of the trees are maidens (with a single stem) and variation in diameter (dbh) is approximately normal with a mean of 0-31 m, SD + 0-09 m (March 1995). The immediate impression was that the trees were of one age and might therefore have been planted. On the pre-enclosure map of 1778 the area was south of the boundary of Leat Wood and shown as “common pasture’. On the O.S. map it was in 1880-81 still treeless and shown as rough pasture. In Dony (1953) a photograph (plate 20), which was probably taken about 1945, shows large trees around the spring and all the slopes above are downland, while the particular slope has sparse scrub above and below the prominent footpath. The hills were part of the estate of Hexton Manor and the present owner Mr J. A. Cooper told me that before World War 2 the hillsides were grazed by sheep, which were later confined to the east of the sharp boundary of the scrub in Dony’s photograph. Parts of the scrub were cut but at no time this century were trees planted on the slope. Cores were taken in March 1995 with a Pressler corer from close to the base of the trunk of a sample of lime trees. Ring counts showed that the trees were, in fact, of several ages. The oldest trees (dbh 0-40—0-49 m) had 51-55 rings and therefore date from 1939-43, several trees (dbh 0-22—0-35 m) had 41-45 rings and date from 1949-1953, which is before the demise of rabbits following myxomatosis, and the youngest tree sampled (dbh 0-14 m) dated from 1966. One of several large hawthorns (Crataegus monogyna) dying beneath the limes dated from 1942. The field-layer under the limes was sparse and included species from the original downland (Primula veris) but consisted predominantly of woodland species dispersed by birds and mammals (Arum maculatum, Geum urbanum, Hedera helix, Sanicula europaea). Mercurialis perennis, which was abundant in Leat Wood and near the spring was conspicuously absent from most of the new woodland. The evidence that 7. platyphyllos is native at Barton is circumstantial and is essentially the age of the oldest trees and their presence at a spring-head site. It must, however, be recognised that T. platyphyllos sets plentiful fertile seeds and in most years seedlings are common in England, so that it has the potential to regenerate and to be dispersed. Indeed, at Barton the colonisation of ungrazed downland and the ability to spread and become established in suitable habitats is dramatically demonstrated. ACKOWLEDGMENTS I thank Mr Graham Bellamy for his enthusiastic support and for cutting the slice of the stump, Dr Jonathan Spencer and Mr Ian Evans for sampling the crowns of the old trees, Mr J. Ashley Cooper for giving me information about the history of the site, Dr Eric Till for confirming the date of planting of the great lime at Burghley. and Dr Francis Rose and Miss Frances Abraham for allowing me to quote from their unpublished data. REFERENCES Dony, J. G. (1953). Flora of Bedfordshire. Corporation of Luton, Museum and Art Gallery. PIGOTT, C. D. (1981). The status, ecology and conservation of Tilia platyphyllos in Britain, in SYNGE, H, ed. The biological aspects of rare plant conservation, pp. 305-317. Wiley, Chichester. C,DsPiGOrE Greenbank, Cartmel, Grange-over-Sands, Cumbria LAI1 7SQ 346 NOTES Watsonia 23 (2000) FURTHER BRITISH SPECIES OF RUBUS L. (ROSACEAE) IN NORTH-WEST FRANCE Following the discovery in the Cotentin Peninsula of west Normandy of eight Rubus species previously unknown outside the British Isles (Allen 1966), wider exploration of north-west France as Well as further study of material already collected have resulted in the detection of several more. A number of British species already known there have also been found to occur more widely. In this second list first published records for mainland Europe are highlighted by an asterisk, as before. Determinations assented to by A. Newton are indicated by an obelisk (7+) against the locality in question. For départements renamed in recent years the previous name is adhered to, for the sake of consistency with earlier records — by analogy with the British Isles vice-county system. My specimens have all been lodged in BM. R. nobilissimus (W. C. R. Watson) Pearsall Calvados: Bois de St.-André, Falaiset. New to France, but collected in Belgium in 1970 by H. Vannerom, det. Newton (Vannerom 1998). R. oxyanchus Sudre_ Calvados: a colony in wood at west end of Rochers des Parcs, Clécy. This trans-Channel species seems much rarer in France than in England. *R. plymensis (Focke) Edees & Newton Manche: margin of heath, La Pernelle, near Quettehour. This Cornubian species has also recently been found in Guernsey. R. pyramidalis Kalt. Extends westwards at least to central Brittany (Ille-et-Vilaine: Forét de Paimpont). *R. riparius W. C. Barton ex Newton Calvados: abundant both in heathy scrub at west end of Rochers des Parcst and in Bois de St.-Clair, Clécyy. Only recently recognised in south Wessex, before which it was believed endemic to north-west Wales. *R. viridescens (Rogers) T. A. W. Davis Loire-Inférieure: patch in a boggy ditch, wood on west side of Herbignac, det. Newton. R. rubritinctus W. C. R. Watson Locally common in Cotés-du-Nord, where Rogers turns out to have collected it in 1897 (BM). Very local in central Normandy. R. hylophilus Rip. ex Genev. Rare. Calvados: foot of Rochers de Brisevieille, Cossesseville; Bois de St.-André, Falaise. Cotés-du-Nord: Lande de Fréhel. Single bushes or patches in each case. R. neomalacus Sudre One of the commonest species of the region at the mouth of the Loire, its French headquarters, yet, puzzlingly, in Britain virtually confined to Surrey. R. stenopetalus Lef. & P. J. Mueller Calvados: Bois de St.-André, Falaise. R. adscitus Genev. Mostly thinly distributed. Seen in plenty only in certain woods and plantations in Calvados, Sarthe, Maine-et-Loire and Cétes-du-Nord. *R. thurstonii Rilstone Manche: Mont du Roule, Cherbourg, locally frequent on lower slopes‘. Collected by Newton in 1984 near Quettehou, in the north-east of the Cotentin Peninsula. R. vestitus Weihe Orne: Chateau d’Or plantation, Montrée, on clay. The nearest this basicole seems to come to the granite massif covering most of north-west France. *R. coombensis Rilstone Manche: Mont du Roule, Cherbourg, locally common in heathy scrubr*; wood margin, La Glacerie Eglise, near Cherbourg’. R. echinatus Lindl. Rare or absent north of the Loire valley, where it becomes plentiful from Angers eastwards. R. leightonii Lees ex Leighton Though seemingly absent from the Cotentin Peninsula proper. locally common in Calvados, Cétes-du-Nord, Maine-et-Loire and Indre-et-Loire. R. bloxamii (Bab.).Lees Increases southwards to become locally comon in Orne and Mayenne. but apparently absent from Brittany. NOTES Watsonia 23 (2000) 347 R. botryeros (Focke ex Rogers) Rogers The supposed first record from the European mainland (Allen 1996), near Cherbourg in Manche, turns out to have been anticipated by Rogers in 1897 at Dinan in Cétes-du-Nord+(BM). *R. cantianus (W. C. R. Watson) Edees & Newton Manche: Bois du Mont du Roc, near Cherbourgy. A surprising occurrence in this strongly Atlantic florula of this otherwise south-east England species, unknown west of Wiltshire. R. flexuosus P. J. Mueller & Lef. Not seen outside the Cotentin Peninsula. R. insectifolius Lef. & P. J. Mueller Calvados: locally common on gorse-covered hillside, Rochers des Parcs, Clécy. Apparently rare in north-west France, R. asperidens Sudre ex Bouvet Calvados: one clump at east end of Rochers des Parcs, Clécy. Apparently very rare in Normandy and absent from Brittany, coming in quantity only in Anjou. *R. rilstonei W. C. Barton & Riddelsd. Morbihan: Lanmeur, 1969, H. Vannerom, det. Newton. Cétes-du-Nord: common on heath margins, Lande de Fréhel*. *R. tamarensis Newton Calvados: in three places near Clécy — locally common on north bank of R. Orne at foot of Rochers de Brisevielle, Cossessevillet, Bois de St.-Clair, Meslay+; hedge bottom by La Vigne, Culey-le-Patry. R. tumulorum Rilstone Manche: Bois de Pépinvast, Val de Saire+. Second record for the Cotentin and mainland Europe. R. scaber Weihe Ille-et-Vilaine: Forét de Paimpont, one patch (new to Brittany’). R. nemorosus Hayne ex Willd. Penetrates west to the Loire mouth, where it is common in the fens of the Parc des Briéres. R. transmarinus D. E. Allen The wider distribution claimed by Rogers, earlier doubted (Allen 1994), has proved to be the case after all: thinly distributed in Calvados, this is also the prevailing member of Sect. Corylifolii in the coastal parts of C6étes-du-Nord (where it may be the plant collected by Coilliot at Le Val-André which Sudre described as R. x namneticus). Of the 208 species included in the monograph by Edees & Newton (1988) that were believed at that time to be endemic to the British Isles, no fewer than 36 have now been detected on the European mainland. Of these, 17 have been found as yet only in France and 14 in the Low Countries (cf. Vannerom 1988). A further seven are also now known to extend to the Channel Islands, which belong to the mainland geographically rather than to Britain. REFERENCES ALLEN, D. E. (1994). Four new Anglo-Sarnian species of Rubus L. (Rosaceae). Watsonia 20: 143-151. ALLEN, D. E. (1996). British species of Rubus L. (Rosaceae) in the Cotentin Peninsula of Normandy. Watsonia 21: 129-133. EDEES, E. S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. VANNEROM, H. (1998). Wijzigingen aan de Rubus-inventoris (Sect. Corylifolii Lindl. en Sect. Rubus) in de derde editie van de ‘Flora van Belgie’. Dumortiera 70-71: 45—49. D. E. ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ RUBUS ANGUSTICUSPIS SUDRE (ROSACEAE) IN SCILLY In a recent revision (Allen 1997) of Rubus fruticosus agg. in the Isles of Scilly, v.c. la, the granite archipelago 45 km south-west of Land’s End, 18 morphotypes were listed as occurring, of which five appeared to be undescribed. Two of the latter, both members of ser. Hystrix Focke, are locally 348 NOTES Watsonia 23 (2000) abundant. One of these, already known to be widespread in Brittany, has subsequently been described as R. venetorum (Allen 1998); the other has now belatedly been recognised as an unusually robust version of R. angusticuspis Sudre, a British endemic hitherto recorded only from the southern end of the Welsh Marches, North Somerset, v.c. 6, and one wood in the Isle of Wight, vce 0: The discovery of this species in Scilly is of interest for three reasons apart from the major extension to its known range that this represents. For a start, its local profusion on the two largest islands, St. Mary’s and Tresco, and the robustness of its growth, in full sun no less than light shade and despite the thinness of the soil and exposure to strong westerly winds, would seem to indicate that it finds here conditions conducive to its optimal development. The other English and Welsh populations thus have to be regarded as for the most part depauperate by comparison. The Scilly plants, however, differ from the generality of those not only in dimensions - the terminal leaflet of the stem leaves ranges up to 11 x 6 cm as against 8 x 5 cm given as the limit in Edees & Newton (1988) — but also in two qualitative characters: in the specimens examined the styles were noted as yellowish, not “red-based”’, and the petals instead of being “bright pink” are uniformly more or less white. In this last respect R. angusticuspis in Scilly parallels R. mollissimus Rogers, another of the islands’ Rubus species with a much more easterly range in Britain. It may be that the high actinic content of the sunlight for which the islands are noted places pink flowers at a selective disadvantage there. A third point worthy of note is the coexistence in quantity in this comparatively remote island group of members of three quite different Rubus florulas: R. iricus Rogers, otherwise all but confined to the western half of Ireland, R. venetorum D. E. Allen, a species predominantly of Brittany and R. angusticuspis, a member of the Severn Bay Florula (Newton 1980). This is one further illustration of the value of Rubus for the finer discrimination of phytogeographical affinities than most other biota are able to provide. REFERENCES ALLEN, D. E. (1977). Rubus L. (Rosaceae) in the Isles of Scilly: a revised list. Watsonia 21: 355-358. ALLEN, D. E. (1998). Five new species of Rubus L. (Rosaceae) with transmarine ranges. Watsonia 22: 83-96. EDEES, E. S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. NEWTON, A. (1980). Progress in British Rubus studies. Watsonia 13: 35-40 D. E. ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ Watsonia 23: 349-358 (2000) 349 Book Reviews A Rum Affair. Karl Sabbagh. Pp. ix + 223. Allen Lane, The Penguin Press, 1999. £16.99, ISBN 0-713-99277-8. Karl Sabbagh is a Cambridge graduate and a television producer but not a botanist. This is a book about a botanist, about the unmasking by fellow botanists of fraudulent botanical records made by Professor J. W. Heslop Harrison in the Hebrides during his pioneering researches there with students from Newcastle University in the period 1935-51. In particular, it focuses on the part played by J. E. Raven and his visit to Rum in 1948, after he had been let down by the distinguished botanists that it was originally planned would accompany him. Raven found persuasive circumstantial evidence of plants being fraudulently introduced which he wrote up in a riveting manuscript that he brought back to his fellow dons at Kings College, Cambridge, where it was hushed up for 50 years. John Raven, the kindest and most sensitive of people and an expert and meticulous botanist found himself in the most invidious position. He alone of all those who had doubted Heslop Harrison’s records had found evidence of fraud in the field, others just failed to refind the plants and insects claimed by Heslop Harrison. Raven had been a conscientious objector in the war, he would never have launched a personal vendetta. Sabbagh has taken the trouble to research the nature of botanists, and field botanists in particular, he perceives that “botany is very different to a love of cultivating plants. In fact, there seems sometimes to be very little overlap” and, quoting from John Raven’s friend, Tim Clutton- Brock, “my impression is that botanists are really fascinated with the details and the facts, whereas zoologists like working in bigger, broader-brush, theoretical frameworks”. Heslop Harrison came to zoology before botany. Sabbagh’s perceptions of us botanists, and his misconceptions, are refreshing, though some of the misconceptions might have been avoided if his research had included a day on a B.S.B.I. field meeting where he would have learnt an appreciation of the endless scope for errors in fieldwork. The book is a vignette on the social structure of British botany in the nineteen thirties and forties. There are class prejudices, there is the Oxbridge-Redbrick divide, there are amateurs and professionals, there is Rum itself as a personal fiefdom of the Bullough family. There are cliques and rivalries but, transcending all, there is a culture in which there was a great reluctance to expose a fellow scientist especially when he had a family of successful scientists following him. There is much excellent biographical detail not only on the Heslop Harrisons and Ravens but also of other prominent naturalists of the day, including Frank Balfour-Browne, “Willie” Clark, J. E. Dandy, Dick David, E. B. Ford, George Taylor, Max Walters and A. J. Wilmot. It is not clear if Sabbagh grasps why publication of an exposé was in a sense unnecessary. The botanical grapevine was sufficient. The relatively few who had a need to know because it might affect their research were tipped off. Inevitably, the insinuations sometimes suggested too black a picture. Be that as it may, many of Heslop Harrison’s “dubious” records have since been refound. Many a botanist has left a trail of dubious records behind him, and much trouble they have caused, but fraud, as such, is usually thought to be rare. Sabbagh studies the motives for scientific fraud, which he illustrates with other examples and suggests it to be much more common than generally thought. Quoting Felix Franks “If ever the delicate balance between making a living, searching for the truth, and obtaining the approbation of one’s peers is upset, then there is the danger of deviant science”. Nowhere, however, does he suggest that advancing years could have had something to do with it, or that some of the arrogance and dogma with which Heslop Harrison ruled his university department could have affected him under the maxim “power corrupts, absolute power corrupts absolutely”. This book is a first rate read, you will not put it down until the end. And, if you are a botanist, you will be left, fraud aside, with personal heart searching as you seek a humble approach to the intractable intricacies of nature. M. E. BRAITHWAITE 350 BOOK REVIEWS A Naturalist’s Shetland. J. Laughton Johnston. Pp xii + 506. T. & A. D. Poyser. 1999. £27.95, ISBN 0-85661-—105-0. It is some 20 years since the first publication of Sam Berry and Laughton Johnston’s monograph on “The Natural History of Shetland” in the Collins New Naturalist series. Since then, much has changed in Shetland and much of this change has had a major effect on both the islands’ natural heritage and our understanding of it. Of course this change had been largely anticipated at the time of writing, as the oil industry was already having a tremendous impact on the social, environmental and economic fabric of the islands. It is therefore timely that the state of the environment and our knowledge of it should be revisited, and who better than one of the original authors of the New Naturalist book. The book is not simply a re-working of the previous book: indeed it has been thoroughly revised and an enormous amount of new material added, especially information and understanding gained from the many ecological and environmental studies that have taken place in the intervening years. Although the extent and scope of the book remains as broad as it was 20 years ago, the structure of the book remains largely unchanged (despite different chapter headings), moving through the physical influences on the landscape; the history of colonisation; and then on to the nature of the biological communities that make up Shetland’s natural environment. The new information is crucial though; for the botanist, the publication of Walter Scott & Richard Palmer’s The Flowering Plants and Ferns of the Shetland Islands, published in 1987 after the original New Naturalist volume is a case in point, and Laughton Johnston has incorporated a substantial amount of recent botanical work including the improved understanding of the Alpestria hawkweeds (now ‘protected’ under the U.K. Biodiversity Action Plan) and the many studies on the enigmatic, but endlessly fascinating, serpentinite soils of the Keen of Hamar N.N.R. The final chapter of the book, reflecting perhaps the time we live in, deals with issues of sustainability: issues previously dealt with under the chapter devoted to conservation! The influence of the oil industry on the natural history has been all-pervasive. In some cases the effect has been direct, in terms of built development e.g. the giant oil terminal at Sullom Voe) and the effect of spilt oil on the coastal and marine environments, especially from high profile incidents such as the Esso Bernicia and the Braer. But the oil industry has brought wealth to the islands, which has been used to fund much other change. The influence of grants to the agricultural industry is one of the most noticeable in terms of its impact on Shetland’s semi-natural vegetation. Laughton Johnston deals with these subjects effectively, and brings us right up to the present and the challenges that the islands (and their environment) face as we start another century. The book is a tour de force. Its format and presentation have been well thought out, and the insertion of many new figures, sketches, and colour photographs give it a sense of quality that justifies its price. Some of the photographs have been strangely cropped and appear too small, but this rarely detracts from the overall feel of a real gem of a book, that will last well for at least another 20 years. Buy it if you are at all interested in Shetland, or plan to go at some time in the future. There’s everything in it that you’d ever want to know about Shetland and a lot more besides! A. DOUSE British Red Data Books I Vascular Plants 3rd edition. Compiled and edited by M. J. Wigginton, Pp. 465. J.N.C.C. Peterborough 1999. £30, ISBN 1-86107—-45 1-4. It is 16 years since the publication of the 2nd edition of the Red Data Book for Vascular Plants, and in this time there have been many field surveys, county Floras, and local red data books. The previous slim volume was very much a basic factual account which stimulated more detailed work and helped set the scene for better things to come. This book, with 465 textual pages, expands significantly on the autecology, distribution and changes in populations of the species that have taken place especially in the last 20 years. Comments are also provided on the European or world- wide status. BOOK REVIEWS efoul| The introductory pages set out the background and context of the book and summarise the all- important data sources. Without accurate record-keeping and meticulous checking, it would not have been possible to produce a meaningful and rigorous account for each species, or the summary tables. Accurate records also provide the essential platform for the analysis of numbers of populations, spatial changes, and colonisation of new areas in a truly scientific, rather than conjectural, manner. There are 289 native or probably native species and subspecies that are considered to be threatened or near-threatened in Britain addressed in this volume, compared with 317 in RDB 2nd edition. This is due to the fact that 44 taxa have been excluded because they are non-native, introduced, casual, escaped, varieties, hybrids or of dubious status. There are now 20 extinct species as opposed to 19 in the 2nd edition; the most recent extinctions being Agrostemma githago, Crepis foetida and Neotinea maculata. On the other hand species identified by Scarce Plants in Britain as being rare have been added. A significant change is in the revised I.U.C.N. threat categories, which have now been agreed after much debate, and which aim to provide clearer guidelines for evaluating rarity. Before, there were four categories and now there are eight. They are intended to help assess priorities for conservation on an international scale. For each species the text follows a similar format, beginning with habitat, associates and reproductive biology. Historical and present geographical ranges are then described with population details. Information on threats and conservation management is provided and, finally, commentary on the distribution in a wider context i.e. European and/or world-wide. These accounts were contributed by many botanists, both amateur and professional, but the majority have been distilled by the editor, Martin Wigginton. His dedication and meticulousness have ensured that much of the existing and new information has been incorporated into the book. Margaret Palmer’s patience and guidance ensured that these efforts came to fruition. The valuable contributions from other authors, following the excellent example set by the Nationally Scarce Plants volume, ensure that the individual accounts are unique, and provide variety for the reader. Some accounts are more erudite and detailed than others, reflecting the author’s depth of knowledge and experience of their species. But this does not unduly detract from the work, as we know there is still much to be discovered about our rarer plants, and it leaves room for additional notes in the 4th edition! Some of the most useful information is contained in tabular form e.g. the list of species protected by the various international and national regulations, and a list of endemics, 71% of which are threatened or near-threatened. Occurrence of species in different habitats shows the importance of our coastal areas, calcareous grasslands and mountains. Planners will be delighted to see the rare species listed for counties and unitary authorities. In the conservation section, the many practical and positive steps that have been taken during the past 20 years through the activities of English Nature’s Species Recovery Programme and Plantlife’s Back from the Brink campaign, plus the Wildlife Trusts, the National Trust and the Wildflower Society, are discussed. Threats from habitat destruction, pollution, collection and global warming continue to exist but, on the whole, the decline of the rarest species has been halted and the status of many has become more secure. Recently, thoughts have been focused on genetic aspects, and this is a subject which has been neglected botanically. Now there is an increased awareness of the conservation of genetic diversity and a section summarises current aspects and highlights this as an area of work that needs increased effort. There are a few drawbacks to note. The volume is definitely too heavy to carry far and the maps for some species could be considered a waste of space (those with one or two dots or numbers). There is also the exclusion of Ireland and the Channel Isles (but there is the Irish Red Data Book 1988 to compensate). I am well aware of all the hard work and collaboration that is needed to produce such a volume, and I have no hesitation in saying well done to all concerned, on producing a work which is informative, interesting to read and, on the whole, well laid out, and a substantial improvement on the previous editions. L. FARRELL 352 BOOK REVIEWS Sir Joseph Dalton Hooker, traveller and plant collector. R. Desmond. Pp. 286. Antique Collector’s Club, Woodbridge, with the Royal Botanic Gardens, Kew. 1999. Hardback £29.50. ISBN 1—85149-305-0. No British botanist can have started life with so favourably auspicious a pedigree as the younger Hooker: a professor of botany who would shortly become Director of Kew, for a father, Dawson Turner for a grandfather and James Dalton for a godfather. He would subsequently reinforce that by marrying the daughter of Henslow and, when she died, the widow of Scotland’s leading naturalist, Sir William Jardine. Various other scientific luminaries can be found slotted into that dynasty on all its sides. Though very different in temperament from his father, it must have seemed foreordained that he would have a no less illustrious career in the same line of work. That Ray Desmond should write this major biography seems almost to have been foreordained too. After many years as librarian and archivist of Kew he has produced in a steadily more impressive stream a history of the India Museum, a history of the botanical exploration of India and the long-needed, definitive history of the Royal Botanic Gardens themselves. Now this massive accumulation of expertise has eventually all come together in this scholarly yet highly readable, lusciously illustrated account of the man who will always be associated first and foremost with the exploration of the flora of the Himalayas and the enrichment of our gardens with so many of their finest rhododendron species. As early as the age of five or six Joseph Hooker gave signs of having inherited the family’s taxonomic blick by identifying the moss Bryum argenteum on a Glasgow wall. In addition to that, though, he turned out to have enormous physical stamina, acute powers of observation, that painstaking industriousness typical of so many Victorian ‘workaholics’ and, not least, a shyness which bred a complete self-sufficiency. This combination of qualities ideally equipped him for the role of explorer, the role which he himself must surely have recognised as the one most central to his being. This took him first to Antarctica, as a naval surgeon in his early twenties; then came the years in India — the most momentous experience of his life, to which Desmond fittingly devotes no fewer than six chapters, the main bulk of the book — and finally, as late palliatives for his enduring restlessness and love of travel, shortish trips to the Middle East, North America and Morocco. As a result he acquired a wider knowledge of the World’s flora at first hand than any of his contemporaries and was able to make important pioneer contributions to plant geography. In particular, he supplied valuable ammunition to Darwin, his long-time friend and confidante, who chose him as the first to hear of his conversion to the view that species are not immutable. Unlike the elder Hooker’s achievements, which were pre-eminently in building institutions, as a teacher and administrator, the son’s were essentially in his individual capacity as a scientist. Kew was not necessary to his career except in providing him with a permanent base as a taxonomist, and it was not indeed till he was nearly forty that that base became a secure one, when at last the post of Assistant Director was created for him by the government. That was only a half-solution, however, for he was not at all cut out by temperament for life as a public administrator: short- fused and often brusque, he was tactlessly impetuous and inflexible in dealings with civil servants and politicians where his father had been the soul of diplomacy. An academic position would have suited him far better, if only he had not been precluded from that by a lifelong aversion to speaking before audiences. If the author has one weakness as a historian and biographer, it is in closing his eyes overmuch to the frailties of his subjects. It is only in the very last paragraphs that he allows us to learn of Hooker’s petty and shameful treatment of W. H. Fitch, the artist to whom he owed so much (and some of whose superlative hand-coloured lithographs are reproduced in this book). And we are not told at all of his unpardonable subverting of Welwitsch in that heroic man’s legal battle with the Portuguese government over his Angolan collections. Of more moment to readers of this journal, though, is the absence of any reference to Hooker’s publications on British botany. That there were indeed any of these at all is, on the face of things, surprising, for he would appear to have scarcely looked at a wild plant in these islands after his student days were over. Rather, as H. C. Watson was moved to complain to J. G. Baker in 1866, he seemed to hold “British botany in a sort of contempt”. It is all the odder, therefore, that he undertook The student’s flora of the British Islands for Macmillan’s around that very time, a period, moreover, when he was seriously behind with the Flora of British India and having to carry the entire load of Kew’s administration (for Thiselton-Dyer only arrived in 1875 to relieve him of much of that). It is known that his then solitary scientific assistant, Baker, whose BOOK REVIEWS 353 knowledge of British plants was profound, contributed at least the excellent section on Rosa and there must be a suspicion that he ghosted a good deal more. For the work certainly met with a favourable reception, went to three editions and several reprintings and was still the preferred text of some field botanists as late as the 1950s, when Clapham, Tutin and Warburg finally edged it into extinction. Hooker’s name was to become very much more familiar, though, as one of the duo to whom that all-too-hardy evergreen, the Handbook of the British flora, was accredited from the 5th edition onwards. The creation originally of George Bentham, who had knocked it off back in 1858, as only that incredible man could have done, as a before breakfast relaxation (“for the ladies”, as he once admitted), this brought to bear on the taxonomy of this country’s vascular plants the broad species concept advocated by those, at Kew and elsewhere, who were struggling to impose order on the almost overwhelming mass of novelties pouring in from newly-explored regions of the globe. The ruthless ‘lumping’ that this approach involved conveniently resulted in a much simplified text which held a particular appeal to beginners. But by lending his authority to what amounted to a dismissal of years of critical work by those who knew our flora far better, Bentham did serious and lasting harm to British field botany. On the credit side, however, was the equally lasting influence of the companion volume of illustrations by Fitch and Worthington Smith, first introduced in 1880. Those black-and-white figures, which cried out to be painted in, were to send hundreds over the years scouring the British Isles on a lifetime quest to register by that means the finding of as many as possible of the species depicted — as a substitute for the herbarium specimens which would previously have served as trophies. Again, this was an incongruous publication for Hooker to have taken on, though it is not at all clear whether its editing required much from him in the way of either work or knowledge. Probably he did so in part out of loyalty to Bentham’s memory and in part because in his retirement he felt in need of the money that this even steadier seller could be relied on to continue to bring in. It is ironical, and sad, that so misguided a work, in which his involvement may well have been tangential at most, outlasted by far everything else that he published and served to keep his name before the public. For a man who ranks as one of Britain’s all-time botanical greats it was very far from being an appropriate memorial. D. E. ALLEN Atlas Florae Europaeae. Distribution of Vascular Plants in Europe. J. Jalas, J. Suominen, R. Lampinen & A. Kurtto, eds. Committee for Mapping the Flora of Europe & Societas Biologica Fennica Vanamo, Helsinki, 1999, vol. 12 Resedaceae to Platanaceae, pp. 1-250. ISBN 951—9108—12-2. Volume 12 is the latest in the series from the Atlas Florae Europaeae Committee. As I shall explain later, it is much more than a set of maps, although there is no shortage of these in the present work (343 in all). Although eleven families are dealt with altogether (Resedaceae, Sarraceniaceae, Droseraceae, Crassulaceae, Saxifragaceae, Parnassiaceae, Hydrangeaceae, Escalloniaceae, Grossulariaceae, Pittosporaceae and Platanaceae) most of the volume is taken up with treatments of Sedum and Saxifraga. Taxa are mapped down to the level of subspecies, with spots and other symbols indicating presence/absence, extinction/probable extinction, native/alien status, etc. A cut off date of 1930 is used to distinguish recent from older records. This does not give an especially accurate picture of present-day distribution, such has been the pace of extinctions in some taxa, although even with this criterion the vulnerability of some taxa is clearly evident, e.g. the chronic decline of Saxifraga hirculus across much of its continental distribution. The authors/editors of the respective accounts have taken the opportunity to revise the treatments provided in Flora Europaea (editions 1 and 2) and, at least as far as Saxifraga is concerned, they have included not only necessary nomenclatural changes, but also new taxa that have been recently recognised or discovered in places as far apart as Spain and northern Scandinavia. In many such cases a certain narrowing of the species concept can be detected from the earlier Flora Europaea account. In my view, this can often be a good thing, because scrutiny of the resulting distribution maps reveals interesting patterns of geographically correlated variation that would otherwise have been obscured. Analysis of these patterns from a phylogeographical perspective is now possible using the various molecular techniques that are available, and by this means we learn more about how our flora has evolved and become distributed. 354 BOOK REVIEWS Apart from the detailed maps, the notes on distribution, the revised nomenclature and updated taxonomy, what else have the authors/editors provided for us? The answer is an extraordinary amount of information by way of literature references not only to the nomenclature and taxonomy but also to studies of biosystematics (which here includes almost any aspect that has a bearing on variation or relationships), with chromosome numbers being singled out for detailed attention. The treatment of chromosome numbers deserves special praise because it is both comprehensive (citing a list of the countries from which the counts were made) and critical (indicating aberrant, problematic or otherwise erroneous counts), and includes references to the original literature. Speaking as someone who has attempted such a review of the genus Potamogeton, I know just how much work is involved in this task alone. In short, the authors and editors are to be congratulated on a most impressively detailed and accurate piece of work which will join its predecessors as an essential reference for anyone with an interest in plant geography, evolution, floristics and conservation. R. J. GORNALL Flora of Cornwall. Atlas of the Flowering Plants and Ferns of Cornwall, with notes on some species recorded on the Isles of Scilly. C. N. French, R. J. Murphy & M. G. C. Atkinson. Pp. 400. Wheal Seton Press, Camborne. 1999. £40, ISBN 0—953461 3-0-0. The need for an entirely new Cornish Flora has been keenly felt for some while. The present work, unlike its predecessors, is an Atlas. It is based on systematic tetrad recording by a small team of volunteers begun in 1987, along with a wealth of old and not-so-old records gleaned from other sources. The scope of the Flora is formidable, with entries for some 2,600 vascular plant taxa. Its treatment of aliens and critical groups like brambles and dandelions is particularly impressive. Many sub-taxa not mentioned in Stace (1997) are included, though the taxonomic value of some of these must be questionable. (What, for example, is Festuca ovina var. hispidula?) The volume is in A4 format, with the usual introductory sections covering geology, soils, climate, vegetation history and so on, plus descriptions of the main plant habitats and essay-length pieces on four regions (Bodmin Moor, the Lizard, West Penwith and the Isles of Scilly). The habitat/vegetation accounts are detailed and well-rounded but, rather oddly, only “wetlands” and heaths are described in terms of their NVC communities. The book attempts to deal with the whole of Cornwall although, as the authors admit, coverage of the Isles of Scilly is considerably less thorough than the rest. The islands are (inexplicably) omitted from the distribution maps, which means that for species with no mention of Scilly in the text one is left not knowing whether they occur there or not. For example, on p. 58 Ranunculus bulbosus, R. parviflorus, R. sardous and R. sceleratus are all found on Scilly, but (unlike R. repens) are not listed as such (and incidentally R. marginatus, given as “thought to be extinct”, is still extant on St Martins). The introductory essay on the Isles of Scilly (p. 31) lists certain “missing” plants that are common on the mainland, including at least two (Stachys officinalis and Dipsacus fullonum) that do occur there. This reviewer, for one, ended up wondering whether the Isles of Scilly would have been better left out of the book altogether. The species accounts provide useful summaries of habitat preferences, conservation status, distribution and abundance. Tetrad dot maps - showing old (pre-1980) and recent (post-1979) records - are placed within the text, and for all the less common species there are lists of selected localities. Production constraints have meant that maps are limited to four per page. This is not a problem when the unmapped species are rarities with all their localities listed in the text, but it is slightly annoying that some “less -rare” plants are without maps too (e.g. Cerastium semidecandrum, Ranunculus sardous, Picris hieracioides and Persicaria amphibia). Assessments of “conservation status” are based on Scarce Plants in Britain (1994) and, unfortunately, the badly out of date 2nd edition of the Red Data Book (1983) rather than the 3rd edition (1999). Readers should therefore take the categorisation of rare species (and the maps on p. 6) with a hefty pinch of salt! The “locally scarce” list on p. 381 is an updated version of one produced for the Red Data Book for Cornwall and the Isles of Scilly (1997), but the rationale behind the selection of species is not stated and the omission of Brachypodium pinnatum, Carex disticha, Catabrosa aquatica, Neottia nidus-avis and Trifolium fragiferum, for example, is hard to understand. BOOK REVIEWS 395 Floristic similarities and differences with neighbouring Devon, and beyond, would perhaps have merited further commentary and analysis. To anyone living “up-country” it must seem strange that in Cornwall Berula erecta, Briza media, Dactylorhiza fuchsti, Hordeum secalinum and Leontodon hispidus are all extremely rare, while Valeriana dioica and Limonium vulgare (to name but two) are missing altogether. Each county has its own special quirks, of course, but Cornwall seems to have more than its fair share! Even the most thoroughly researched Flora will fall short of being the last word on its subject. As the authors rightly say, “this Flora ... should be considered as part of an ongoing study” (p. 1). It is an important milestone that will, in its turn, stimulate a new wave of botanical recording in the county. And, yes, I am pleased to report that there are plants out there still waiting to be discovered: a brisk walk along the coast between Millandraeth, Looe and Portnadler Bay in October produced a list of 200-odd species, including new tetrad records for Atriplex portulacoides and Verbena officinalis, and post-1979 “firsts” for Asplenium marinum and Geranium rotundifolium. At £40 this Flora is not cheap, but as a turn-of-the-century guide to the botanical delights of this strange and wonderful county it is probably indispensable. Despite its few shortcomings, it stands as a fitting tribute to the enormous effort of all those involved in its production. Anyone with more than a passing interest in Cornwall’s plants should make sure they get hold of a copy. S. J. LEACH Flora of Cornwall on CD-ROM. Cornwall Business Systems. 1999. £40.00 from CBS House, Albany Road, Redruth, Cornwall, TRI15 2HY By producing the Flora of Cornwall on CD-ROM format, it is possible to include much more than appears in the book. All of the introductory chapters and appendices can be read on screen if desired, but the bonus is that the database on which the Atlas section of the Flora is based is also included, from which maps can be generated, including species not mapped in the Flora. The CD- ROM database also contains some records extra to the Flora. Many more photos than could be produced in the book are also included. Species are searched for using Latin or common name lists arranged alphabetically. On selection, the species caption from the flora is reproduced, one or more photos of the species appear, if available, the map is generated and the records listed. The map can be generated on one of a choice of six bases including a geological base, at one of four sampling scales, 10-km, 5-km, tetrad and 1-km scales. The records for the species are listed giving grid reference at tetrad level, locality, date and recorder(s) and can be ordered by grid reference or year. Some limited querying of the records can be undertaken on a date basis, such as records for a particular year, before or after a chosen year, or for a date range. A gazetteer of Cornwall is included, searchable by place name or by grid reference, and pinpointing places on the map. This links to the Flora database, making it possible to list the records for a particular 1-km square, or to produce a printable species list for the square. Like the Flora records and map, these lists can be limited to particular date choices. Other add-ons of interest are a tourist map of the county and photographs and information on local landmarks. The species photographs can be searched and viewed on their own. Using a recent PC with a fast CD-ROM, the map and records were generated fairly quickly for the rarer species, less than 15 seconds for species such as Polygonum maritimum or Rumex rupestris, which have about 100 or less records. However for commoner species, there is some waiting time involved; to generate Agrostis capillaris, listed as the 100th most commonly recorded species, took 90 seconds if the records were listed by year or 160 seconds if listed by grid reference. Other than a few error messages on installation, which were notified with this review copy, the only problem some users may have is resetting their usual display resolution on use to be able to view the CD-ROM full screen. As the first county Flora to be produced on CD-ROM as well as traditional book format, this is surely an indicator of the future. This CD-ROM will be invaluable for any botanist regularly requiring more detail than can be gleaned from the printed media. A. STEWART 356 BOOK REVIEWS Britain’s Rare Flowers. Peter Marren. Pp. xv + 334. Poyser Natural History, London. 1999. £24.95, ISBN 0-85661-114—X. Over a decade ago a book published with the title Wild Flowers in Danger caused widespread concern because it was mainly a guide to where to find our rarest plants and the book itself added to their danger. So, you might expect Britain’s Rare Flowers would take you down the same path to the exact spot — but you would be wrong. Peter Marren has written a celebration of our 300 rarest species bringing them to life along with all those botanists, past and present, who have found them and enthused about them over the last four centuries — a book for the armchair rather than the field. What the author set out to do, and achieved so triumphantly, was, as the American writer John Burroughs said of Gilbert White, ‘to seize the significant and interesting features and put the reader into sympathetic communication with them’. He takes you on a journey, first to meet the fathers and elders of British botany, from William Turner to George Claridge Druce, who discovered our flora, and then to areas like the Lizard Peninsula, Avon Gorge, Breckland, Upper Teesdale and the Scottish Highlands which hold the highest concentrations of rarities. During the journey he manages to convey the excitement of discovering or rediscovering a rare plant in a wild place even after a tiring day in ghastly weather. Later he looks at individual species and considers the value of research in assisting their conservation. He can be wise and challenging and points to a growing gap between theoretical scientists and practical conservationists in the last 30-40 years (partly a direct result of the disastrous break-up of the Nature Conservancy Council). However, his admission that he is not a scientist perhaps explains his ambiguity about the effects of collecting and a failure to appreciate that the picking of daffodils can be beneficial and that banning it has caused decline. On the other hand his non-academic view that long-established introductions like the spring crocus and spring snowflake deserve protection and were properly included in Eds 1 & 2 of British Red Data Book 1: Vascular Plants and should not have been excluded, as they were in Ed. 3, is surely to be welcomed. The non-scientific background also shows in his failure to appreciate variation within species e.g. that there are native and introduced forms of sainfoin, the former in the ancient turf of the Wiltshire Downs and only the latter brought in as a fodder crop in the 17th century, or that there are native forms of weed species on beaches and river banks and others which have come in as followers of man. Elsewhere he gets into a muddle with Batrachian Ranunculus hybrids placing R. aquatilis x tripartitus in the New Forest where the taxon is R. omiophyllus x tripartitus (R. x novae-forestae S. D. Webster). It is the mixture of fact and anecdote which make the book so palatable and informative but, as these examples show, it cannot be used as an entirely reliable source. There is a wicked story of one of the 14 remaining wild cotoneasters on the Great Orme being cut down by ill-trained conservationists but the truth is that there are only seven native bushes and none of these or any translocated plants have been destroyed. Inaccuracies have crept into the captions (a non-botanical picture-editor?) so that Welsh botanists will be surprised to learn that Chamaemelium (sic) nobile is ‘now restricted to scattered localities in southern England’ when there are 8 10-km square records on their coast north to Lleyn. The photo of Lundy cabbage, Coincya wrightii, given as on p. 121 in the index turns up on p. 124 with a caption reading ‘two kinds of beetles seem to eat nothing else’ whilst the adjacent text names three. The index is the worst feature of the book — totally user-unfriendly. It is arranged in alphabetical order of English names starting with -the first word so that to find references to our mountain saxifrages you need to look under Alpine, Drooping, Marsh and Tufted, but there is no consistency — both native Limes are together! Moreover, recommended English names are not always used: to find Goldilocks Aster, Aster linosyris, you would need to know that the Marren name is Hair-leaved goldilocks. There is no alphabetical list of Latin names. So if you want reliable facts about Britain’s rare plants (and no index at all!) turn to Ed. 3 of the British Red Data Book (Wigginton 1999 see Review p. 350), but it will not compare as a good read with Peter Marren’s compelling story — but regard some of that as fiction — and the index is friction. ; F. H. PERRING BOOK REVIEWS 357 Vegetation of the British countryside — the Countryside Vegetation System, Ecofact Volume 1. R. G. H. Bunce, S. M. Smart, H. M. van de Poll, J. W. Watkins & W. A. Scott. Pp. 224. Dept. of Environment, Transport and the Regions, London. 1999. Price £48.00. ISBN 1—85112—155-2 Measuring Change in British vegetation, Ecofact Volume 2. R. G. H. Bunce, S. M. Smart, H. M. van de Poll, J. W. Watkins & W. A. Scott. Pp. 144. Dept. of the Environment, Transport and the Regions, London. 1999. Price £10.00. ISBN 1—870393-47-3 Ellenberg’s indicator values for British plants, Ecofact Volume 2 Technical Annexe. M. O. Hill, J. O. Mountford, D. B. Roy & R. G. H. Bunce. Pp. 45. Dept. of Environment, Transport and the Regions, London. 1999. Price £5.00. ISBN1—870393-48-1 One of the biggest problems faced by distribution ecologists and conservation campaigners alike is that the British vascular plant flora (impoverished though it is by continental standards) appears too vast in terms of species to survey quantitatively. Where is the plant equivalent of the common bird census, plant conservationists cry — well, this may be it. These three volumes are the first major publication resulting from the Countryside Survey: the results from the Countryside Surveys of 1978 and 1990 undertaken by the Institute for Terrestrial Ecology have been analysed and compared, giving for the first time a quantitative, objective and repeatable survey of the vegetation of the British countryside. The data are collected from 508 1-km squares containing over 13000 plots, selected by stratified random sampling so that the full range of landscape types are included. The data collected include species lists with abundance from quadrats and the results have been analysed in a multitude of ways. So, what do we get in these three publications? To start with, 37 groupings of plants by their ecological preferences, and 100 broad vegetation classes called the Countryside Vegetation System or CVS. Does this replace the National Vegetation Classification (NVC)? No — the two systems are very different — the NVC covers semi-natural plant communities in detail and the sample plots were selected deliberately rather than randomly; the CVS aims to cover the vegetation we encounter in the wider countryside, with the random samples allowing statistical analysis for monitoring change in the British vegetation. In Volume 1, there is an abundance of coloured distribution maps of vegetation classes, pie charts and histograms of landscape associations, soil and land cover, together with comparisons with and relationships to the NVC and also Grime’s plant strategy characterisations (based on three basic strategies — competitor, ruderal or stress-tolerant, or combinations of these three). Volume 2 provides lists of the vegetation classes and indicator species groupings. There are even Statistics on the decline of butterfly (larval) food plants and effects on lowland farm birds. The second volume is also supported by a technical annexe containing re-calibrated Ellenberg values for the British flora, which I have already found invaluable. If you haven’t encountered Ellenberg values before, these constitute a set of ecological indicator values based on a plant’s tolerances of temperature, light, pH, etc. and are particularly useful in attempting to quantify some of the vaguer aspects of descriptive plant ecology. Overall, the amounts of data contained within are staggering, perhaps even overwhelming, but the project is well on its way to confirming what many of us have suspected for some time — that nitrogen-loving competitors are on the increase at the expense of less competitive species. There is no doubt that the data can be used as serious campaigning tools for changes in modern agricultural methods, but my impression is that these messages are almost lost in the mass of information; hopefully we will see a series of papers published as extracts from these reports. The programme is being repeated as part of Countryside Survey 2000, so we can expect further updates on the state of British vegetation, in addition to Volume 3, Causes of change in British vegetation (in prep.). Following Countryside Survey 2000, it will be interesting to see if the periodicity of the surveys is right. The point of monitoring is to detect change and ten years may tell us of change and of loss — but will it give us enough time to do anything about it? In the introduction to Volume 1, the NVC, the Phase I Handbook and the Flora Britannica are listed as essential items for the ecologist’s tool box. While I might take issue with the inclusion of Flora Britannica, I would suggest that these volumes are essential reading for any professional plant ecologist, especially those involved in plant conservation. However, at over £60 for the three softback volumes, together with the dense technical content, it will be the truly dedicated amateur botanist who dips a toe into the Countryside Vegetation System. S. J. WHILD 358 BOOK REVIEWS The liverwort flora of the British Isles. Jean A. Paton. Pp. 626. Harley Books, Colchester. 1999. £52.50, ISBN 0—946589-60-7. Liverworts in the British Isles (BI) grow in an almost continuous range of habitats from the lowlands to the highest mountain summits. Of about 6,000 species known in the world, the number in these islands, 296 species, two subspecies and five varieties, seems quite insignificant, but they comprise nearly 70 per cent of the European liverwort flora, and their richness and diversity in the western parts of Britain and Ireland is of great international significance. This is not the first liverwort book for the BI to be produced this century; but it is a definitive, modern work. The pages are approximately A4, the text is in two columns and each of the 314 text figures contains many individual drawings, all by the author. After the preliminaries an Introduction of 21 pages contains sections on Collection and Preservation, and Examination of Material. One on Habitats and Distribution lists species of diverse habitats and then species characteristic of five more specialised habitats. There are sections on Conservation and on how to use the text and figures. Ending the book is a useful illustrated Glossary, a list of Vice-Counties of the British Isles with a map, a Bibliography and an Index. The bulk of the book is the Flora wherein every taxon at present known in the BI is keyed out, described and illustrated. Full descriptions, even including unpublished chromosome counts, are followed by statements on habitat, ecology, altitude, associates and distribution both within the BI and without. Then there are notes on distinguishing features, comparisons with other species and observations on morphological variations. Each taxon has a fine Figure or Figures which accompany the descriptive section and illustrate everything one would want. For example, that of the common liverwort Cephalozia bicuspidata which is *... in almost any moist or wet, shaded or insolated habitat ... from sand, loam, peat ...; also on rotting wood ...’, has over 60 individual drawings which include several habit figures (female, male and sterile), leaves, cells and stem sections. Primarily an identification manual, its importance to those studying plant communities in the BI and Europe will be profound and cannot be over-emphasised: the rich accounts of the ecological requirements and associated species were gathered from the author’s intimate field studies of the plants over many years. With the excellent, all-encompassing illustrations, keys and full descriptions, this book is a highly significant event in the study of European liverworts. Through the publication in the 1900s of several important books on mosses and liverworts in the BI, and an active British Bryological Society, our bryological flora is probably the best known in the world. It is because of the intense interest thus aroused and of the expertise of dedicated bryologists like Jean Paton that species are still being added to the British and Irish lists, some of them new to science. This hardback with covers bearing splendid colour photographs is not for the pocket: it weighs over 2:5 kg. Harley Books has done well to add this attractively produced work to their already distinguished list. Jean Paton is to be congratulated on a splendid achievement. This is a veritable tour de force that will compel the respect and admiration of bryologists world-wide and become the most used liverwort book in the British Isles and Europe in the new century. A. R. PERRY Watsonia 23: 359-372 (2000) 359 Obituaries JOAN WENDOLINE CLARK (1908-1999) Joan Clark (née Rust) passed away on 6 July 1999 at Fort William, Inverness-shire. She had lived in the area for some 60 years and was known and loved by all sectors of the wider community. During the last 25, she developed a deep interest in the botany of North-West Scotland and the Islands. Joan was born on 6 October 1908 at Benholm, Kincardineshire, the only child of Wendoline and John Rust, City Architect of Aberdeen. Her father died when she was ten, and Joan and her mother went to live in Worthing, Sussex, to be near her mother’s family. She continued her education in Worthing before going to a ‘finishing’ school in Switzerland. As a result she was good at languages, and on returning to England went to a college in Brighton to study French, German, and the traditional ‘Shorthand and Typing’. After working in the Foreign Office in London, and in the British Embassy in Paris for two years she came back to London, where her Scottish roots surfaced and she joined the newly formed Scottish Nationalist Party. She was a staunch supporter of the SNP until her death, attending annual conventions almost without fail until 1998. In 1932, at the London meetings, she met Branch Secretary Alastair Clark, (a Glaswegian of Mull origins) who was to become her future husband; they were married in London in April 1935. They moved to Glasgow where Alastair set up a fur-trading business but the outset of World War II in 1939 severely restricted his work. They decided to move north to Loch Leven where they bought and successfully ran the Loch Leven Hotel, a hostelry well-known by many a traveller botanist who had queued for the Ballachulish ferry! Here they were the ideal Highland hosts, making many life-long friends, both locally and from those who came to stay at the hotel during the next 19 years. After the untimely death of Alastair, in 1959, Joan sold the hotel and moved to a bungalow. Tigh Solais, built on nearby land they had owned. I first met Joan in 1972 when she came to Kindrogan Field Centre to the first Sedges Course I had been asked to tutor. There were only four students: a very fit sixth-former preparing for university (with whom I just about coped), two ‘B.S.B.I. Ladies’ (who knew much more about sedges than I did!), and this quiet Scottish soul who had recently joined the Wild Flower Society and wanted the challenge of a critical group. Her advancement to Valhalla was being monitored by Mary MacCallum Webster (which Joan later admitted was a wonderful, albeit terrifying training) and Joan was determined to be a model student. We were two days into the course before she produced a walking stick and timorously admitted she was waiting for a hip replacement! I learnt afterwards that she thought I might send her straight home if it had been apparent from the outset. When we came to a day walking up Lawers Burn, she struggled to the weir where we had lunch. I took her aside and said I felt she might find the scramble through deep heather and bog to the Yellow Corrie (then in the descending cloud) to see the Carex microglochin, too much. I could see her disappointment but she, seeing my real concern, said “I will sit here, and will not move”. The rest of us went on, and having found the species we had come to see and dropped back over the col to return to the Burn, through the mist Joan Clark appeared lifting one leg at a time over the deep sedge litter. I could not disappoint her again and we went up to see the ‘new’ species and joined the other three at the Burn for tea. This determination to ‘complete the course’ and to do as good a job as she could was, I was to find out, the hallmark of Joan’s contribution over the next 20 years to Scottish botany. She had ‘graduated’ from the W.F.S. with a sound understanding of Scottish wild flowers. As a W.F.S. ‘interne’ with ‘Mary Mac’, Joan had offered to type Mary’s MS for the Flora of Moray, Nairn & Easterness (published in 1978). This took many weeks but Joan persevered, and learnt much from the exercise. Joan joined the B.S.B.I. in 1977. She took full advantage of the Society’s Specialists and soon realised the value of voucher specimens and pressed material for checking. Her eldest daughter Sine lived out in N. Uist, and Joan made regular visits to this land of low-altitude nutrient-rich 360 OBITUARIES lochans and extensive coastline. With her own little Morris Minor and son-in-law Archie’s boat, she covered the island. Water plants became a special interest and she sent much material to J. E. Dandy at the Natural History Museum. He was once chiding me for not taking the trouble to ‘float out’ a grass-leaved Potamogeton, saying there “were few botanists of the calibre of Joan Clark and Dorothy Cadbury, when it came to collecting pondweeds”. As some will realise, this was indeed a great compliment to Joan. With the pondweeds sorted out Joan took to other specialist collecting, sometimes ferns when we were looking for Dryopteris expansa, and then, in a big way, dandelions, where she again took great trouble in arranging the plants in the press and arranged her visits to be there in May. By this time she was in regular correspondence with the R.B.G. in Edinburgh and sent many herbarium specimens to the Scottish National Collections. Rubus and Hieracium were other genera that received her careful attention. Her fieldwork in Uist was further encouraged by Richard Pankhurst and others at the “~BM’ when writing the Flora of the Outer Hebrides and her contribution was considerable. When John Cannon relinquished the v.c. Recordership for Mull in 1978, Joan Clark took over v.c. 103. She took her stewardship seriously and made extensive annual visits. We encouraged her to maintain our mini-vice-counties as a recording base on Mull and she added many new records. She also took to recording plants of Coll and Tiree seriously, especially the former where she made several friends through her regular annual visits. Although she was an Aberdonian Scot from the East, she took to the West and its islands with eagerness. Being naturally friendly she wanted to chat with people she met and in the Islands she realised this meant speaking Gaelic. She studied diligently and passed her O-grade. She quickly saw that vernacular Gaelic names given to plants varied from area to area and made a plan to collect these as widely as possible. Between 1975 and 1985 she circulated lists of what names were in the literature, asking for additional ones in use in her contacts’ area. She was asked by Richard Pankhurst to provide names for the Hebridean Flora and she used her knowledge of Gaelic to coin names for taxa not 1n everyday use. For this she enlisted the help of another Uist man, Ian MacDonald (now Director of the Gaelic Books Council) and between them they came up with a definitive master list for all vascular plants growing in Scotland. This 20-year labour-of- love (Gaelic Names of Plants - Ainmean Gaidhlig Lusan, was finally published (at her own expense but with some help from the Gaelic Book Council and the B.S.B.I.); although Joan saw it all, including the coloured cover in final proof, in spite of all stops being pulled out by the printers it reached Tigh Solais seven days after her death. Joan will be greatly missed by many in Lochaber. She was a devoted mother, grandmother and great-grandmother to Sine, Anna and Eilidh and their respective families. She was also a wonderful friend and correspondent to many a botanist who will mourn her passing and remember her with warm affection. A. C. JERMY DAVID EDWIN COOMBE M.A., Ph.D. (1927-1999) Dr David Coombe died in Cambridge on 28 June 1999. Although his name will be known to many B.S.B.I. members for his studies of the Lizard peninsula, and of the British and European species of Trifolium, few will be familiar with the wide range of his interests and the depth of the scholarship with which he pursued them. David Coombe was born in Bath on 9 March 1927, the only child of Charles William and Phyllis Coombe (née Weaver). His father, a ‘motor cycle specialist’, ran a small workshop. He won a scholarship from South Twerton primary school to King Edward’s Grammar School, Bath, and then went on to Christ’s College, Cambridge, in 1945. After gaining a first-class degree and Ph.D., he worked in the Botany School, University of Cambridge, for the rest of his career, being appointed a University Demonstrator in 1952 and a University Lecturer in 1957. He was elected a Fellow of Christ’s College in 1951 and subsequently served his college in many capacities, including Vice-Master. As a bachelor he was able to live in college rooms until he retired prematurely in 1989, after a prolonged period of ill-health. Coombe’s interest in the British flora developed at an early age. He was given his first Flora, Skene’s Flower Book for the Pocket, on his 10th birthday. He was able to botanise in Dorset as OBITUARIES 361 well as in his native Wiltshire in the 1930s, as at least 20 of his mother’s relatives lived near Portland. “Several of the couples were childless,” he recalled in 1991 in a letter to David Pearman, ‘and I was taken out in turn by these kind people on long summer holidays - with all sorts of rituals - the train from Easton to Sandsfoot, and a picnic at “Vetch Corner’ (where Vicia bithynica, V. lutea and Lathyrus nissolia have still been so abundant in recent years).” By 1939, aged 12, he was already noting pure and albino Stork’s-bill, Sea Holly, Sea Samphire, Sea Purslane and the Shrubby Sea-blite at Small Mouth, as well as the abundance of butterflies, especially Painted Ladies, but this holiday was abruptly terminated on 27 August 1939 when all visitors were advised to leave Portland because of the imminence of war. In 1944 he started a detailed card index of the plants he had seen, but he was unable to find time to keep it up after his appointment to a teaching post in 1952. Coombe’s first botanical studies in Cambridge used analyses of plant growth to explain the distribution of species in the field. His Ph.D. thesis, Plant Growth and Light in Woodlands (1952), was supervised by Dr G. C. Evans. He used Impatiens parviflora as an experimental plant, solving the problem of germinating this awkward species and growing it in different light regimes at Madingley Wood near Cambridge. (The plants he grew spread and the species is now well naturalised in the wood.) Coombe and Evans developed the technique of measuring the complex light climate of woodland by hemispherical photography of the canopy. At first they used the camera with the original ‘fish-eye’ lens designed by Robin Hill, who lent it to them until commercial models became available. Coombe continued his work on plant growth after he became a University Demonstrator. Visits to Nigeria and the Cameroons led to studies of the tropical shrub Trema guineensis and the tree Musanga cecropioides, rapidly growing plants of abandoned farmland and forest clearings. Even when he was working for his Ph.D. thesis he developed the interest in the Lizard peninsula which he retained throughout his life. His first trip to the Lizard was made with his parents in the spring of 1950. They had struggled to scrape together enough petrol coupons for the journey, and when they arrived at the rather squalid village round Lizard Green his mother burst into tears of disappointment. However they soon began to appreciate the scenery and botanical interest of the area. In June 1950 Coombe led the first of numerous Botany School excursions to the Lizard. Lewis Frost, then an undergraduate at Emmanuel College, joined this excursion and subsequently the two teamed up to study the flora and vegetation of the area. Later excursions included several held at Easter to study bryophytes, which were jointly led by Dr Harold Whitehouse. It was on excursions to the Lizard that many students were introduced to Coombe’s remarkable ability to synthesise knowledge from many different fields into a ‘holistic’ appreciation of the ecology of an area. His understanding of the Lizard was based on the critical interpretation of evidence from such diverse sources as Cornish place-names, Anglo-Saxon boundary charters, tithe redemption maps, the accounts of Victorian travellers, published tourist guides and the memories of the people of the Lizard themselves, as well as the more usual ‘scientific’ sources such as meteorological, geological and archaeological studies, aerial photography, botanical records and his own knowledge of the soils of the region and its vegetation at all seasons of the year. He developed an acute ability to ‘read’ a landscape. The detailed, long-term observations (including the repeated recording of permanent quadrats) carried out with Lewis Frost provided great insight into the ecological processes at work in the Lizard vegetation, and in particular the way in which rare species at the edge of their range were able to exploit small areas where the vigour of the dominant plants was reduced by different combinations of summer drought, winter flooding and human disturbance. All this information was committed to Coombe’s prodigious memory, and could be recalled at will. Coombe and Frost’s first papers on the Lizard described the heathland vegetation in relation to the underlying soils. A crucial discovery was the fact that, whereas some of the heaths were rooted in soils derived directly from the underlying serpentine rock, the Agrostis curtisii heaths grew over wind-blown loess which had originated elsewhere. Although these papers were perhaps their most influential, a later paper by Coombe alone provides a greater insight into his approach to less Strictly scientific subjects. In a bibliographic study of C. A. Johns’ classic guide A week at the Lizard, he drew on a wide range of sources to clarify the complex publishing history of this work, which was incorrectly described in all the standard bibliographies and major library catalogues. These conclusions were not written up in a dry bibliographical paper but in a highly readable account full of detailed observations and spiced with humour. 362 OBITUARIES To British botanists, the Lizard is famous for its rare annual Trifolium species, and Coombe and Frost were particularly interested in their population dynamics. It was while on fieldwork at the Lizard on 27 September 1957 that Coombe noticed a perennial clover which he described in his field notebook as “‘a thick leaved, +glaucous, hairy-petioled Trif. repens (collected). This turned out to be a diploid species which he described as Trifolium occidentale in 1961. (His original intention to call it 7. atlanticum was dropped when H. Gilbert Carter pointed out that strictly speaking this epithet applied to the Atlas Mountains.) Few British species can have been described with such a wealth of observational and experimental detail, and the paper is in effect a worked example of the biological species concept. Coombe got considerable satisfaction from the fact that botanists elsewhere in western Europe soon came to recognise T. occidentale, and those who knew it in the field had no difficulty in accepting it as a distinct species. (Many of its characters are lost in the herbarium.) He later prepared the account of the large genus Trifolium for Flora Europaea. It was characteristic of his approach that, rather than base his descriptions on published accounts or herbarium material, he grew as many species as possible at Cambridge from seed distributed by botanic gardens all over Europe (much of which turned out to be misidentified). As a result he was able to write an accurate account and a key that actually works. When I first visited the Lizard with David in September 1976, the clovers were germinating in abundance after the heavy rain which followed the droughts of 1975 and 1976. He was able to find all the rare and many commoner species, and to show me how to identify seedlings with only a single trifoliate leaf. Coombe’s field studies were usually backed up by the cultivation of plants in the Botanic Garden at Cambridge, many of them grown in the experimental glasshouses which he had helped to design. He developed methods of growing species such as Isoetes histrix and Ophioglossum lusitanicum, and cultivated many dwarf or prostrate ecotypes from the exposed Cornish coasts. He took into cultivation all the surviving plants of Juniperus communis at the Lizard, rooting them from cuttings (an operation which required horticultural skill and great patience). This provided the basis of a public exhibit in the Botanic Garden where these plants were grown with progeny raised from seed from one of the Lizard plants and with other junipers from all over Britain. The differences between the columnar plants from southern England and the dwarf plants from Fair Isle (which Coombe likened to a Wilton carpet) were very striking, and the display had a great impact on visitors to the garden. Erica x williamsii, the hybrid between E. tetralix and E. vagans, was another special interest. Coombe is the only person to have found more than one plant of this hybrid (only eleven have ever been discovered, all at the Lizard) and he grew all the surviving clones at Cambridge. These were introduced into commercial cultivation, one as cultivar ‘David Coombe’. He also carried out a long series of experiments at the Garden on soil samples collected in the 1950s, which demonstrated the long dormancy of the seeds or spores of several rare or scarce species, including /soetes histrix, Juncus capitatus, Juncus pygmaeus, Lythrum hyssopifolia, Riccia beyrichiana and R. crozalsii. Unlike many botanists he was a good gardener, and he acted for some years as Hon. Garden Steward at Christ’s, designing a garden which contained all the plants mentioned in Milton’s poem Lycidas. His interest in garden plants led him to realise that Limonium companyonis (now L. hyblaeum), discovered at Rottingdean and originally reported as an “enigma of doubtful origins’, was widely cultivated and almost certainly a garden escape in Britain. Although Coombe did not find the East Anglian countryside as attractive as that of counties further south or west, he nevertheless acquired a detailed knowledge of the ecology of Wicken Fen, Breckland, the chalk grassland and woods of East Anglia and the Chilterns, to all of which he led student excursions in the Long Vacation. His approach to botany proved inspirational to many students. When he was in his prime his lectures to the Part II botany students were superb, and Peter Grubb remembers two absolutely outstanding formal papers to British Ecological Society meetings. After a day in the field, or a discussion in his college rooms, one left his company exhilarated by his insight into plant ecology. He influenced many who passed through Cambridge University in the period when the Botany School was one of the leading departments for whole- plant biology. One might have thought that as a “congenital conservative” (his words), and someone who could talk about his subject at great length, he might have seemed a remote figure to students, or even a bore. In fact he developed a great rapport with many students, and some have even commented that he was the one don they met during their time at Cambridge whom they really admired. They doubtless responded to his great enthusiasm and knowledge, his transparent integrity and his willingness to spend time with them. He also treated students as individuals and OBITUARIES 363 was able to see good points in almost all of them. He responded very generously to requests for information from other botanists, professionals and amateurs alike. In many cases where he failed to respond to letters it was not because of any lack of desire to help, but because he started on a very detailed exposition of the subject in a letter which he was unable to find time to complete. He could, however, be scathing in his criticisms of professional botanists whose work did not measure up to his own standards. Coombe applied the same academic rigour to all he undertook. He appeared to be incapable of considering anything superficially, and he usually seemed to be either an expert in a subject or completely ignorant of it. As a young lecturer he had been required to teach a wide- ranging introductory course to students, during which he had obtained a detailed knowledge of several cryptogamic groups, including algae and bryophytes. (In March 1958 Coombe and Whitehouse discovered the moss Hennediella_ stanfordensis, new to Europe, on one of the Lizard bryophyte excursions.) Coombe had a very good command of German, having spent 3 months in Heidelberg aS a research student, and he collaborated with Peter Bell on a translation of Strasburger’s textbook of botany. To his botanical interests were added many others, including books (inevitably), English furniture and the visual arts. His tastes were traditional, and he responded particularly to authors like Thomas Hardy and painters like John Constable who shared his sense of place. As a relaxation he composed limericks and clerihews. It was a matter of great regret to many of Dr Coombe’s friends and colleagues that much of his work remained unpublished. They knew how great a contribution he was capable of making, and tried in vain to persuade him to write more. In his early career he published a series of excellent papers on diverse topics. Unfortunately his output then slowed, and he published little in the late 1960s and 1970s. The detailed Biological Flora accounts of the Lizard rarities which he planned to write, for example, were never completed, and his studies of the semi-natural vegetation and the crop yields on the patterned soils of Breckland were not written up for publication. No doubt his heavy teaching load, his service on innumerable college, university and other committees and his work as an editor of Journal of Ecology all contributed to this. The editorial work, in particular, made great demands on his time and energies as he was meticulous in his attention to detail and haunted by the prospect of misprints. His very appreciation of the interaction of factors which made him such an astute ecologist also appeared to make it difficult for him to pursue a single line through to publication. Added to this was an innate perfectionism: when I once encouraged him to prepare a Biological Flora account of Juncus pygmaeus he said he would not do so, as although it would be better than many accounts in the series it would not be as good as the best. However, none of these factors really seem to provide a sufficient explanation for his failure to publish more at the peak of his career. ate David Coombe at Traunsee, Austria, | August 1956 (photo. taken by Dipl. Ing. B. Weinmeister on the 11th International Phytogeographical Excursion). 364 OBITUARIES A more tangible impediment to Coombe’s botanical work arose in 1978, when he developed a lung disease which was diagnosed as cancer. It was only after a major operation that it was recognised as a fungus (Aspergillus) infection. The adverse effects of this unnecessary operation remained with him for the rest of his life: in particular his right arm was often painful, especially in cold weather. In these later years he turned from his Cornish work to a series of studies of East Anglian topics. Although limited in breadth, these were all pursued with the same depth of scholarship. Several were written up for Nature in Cambridgeshire, where Gigi Crompton provided initial encouragement and Philip Oswald was a sympathetic ‘midwife’ and editor. Two of these papers resulted from his discovery of new populations of Veronica spicata on Newmarket Heath. This led him to a consideration of the lost chalk heaths of Cambridgeshire, which he reconstructed from the fragmentary surviving evidence. He also published a detailed paper on ‘Milton’s Mulberry’ in his college magazine. This tree was said to have been planted at Christ’s College by the poet when he was an undergraduate. Coombe not only demonstrated that it was almost certainly planted many years after Milton’s death, but went on to develop a theory that many of the connections between great men and ancient trees were invented during the Romantic period. Sadly his health prevented him from writing up several of the topics he had been working on in this period - William Turner’s Cambridgeshire plant records (which had been neglected by authors of successive Cambridgeshire Floras), the changing vegetation of Ringmere and the history of the Lombardy Poplar in Britain. The poplar paper would have been particularly interesting, as much of the evidence was drawn from landscape paintings, prints and literature references. In the last years of his life Coombe’s health deteriorated still further, and he was only rarely able to get into the field. He made his last two visits to the Lizard in 1994, to help survey populations of Juncus capitatus and J. pygmaeus for the Red Data Book and to follow up a developing interest in Pyrus cordata. In 1999 he learnt that he was suffering from Parkinson’s disease and began to prepare for the inevitable disabilities it would bring. However in June 1999 he contracted a bacterial infection which failed to respond to treatment, and he died peacefully in hospital only a few days later. David Coombe will be remembered as an ecologist with an unrivalled insight into the flora and vegetation of southern England. He left a substantial corpus of published work, all completed to his own high standards and some of it of outstanding merit. However, his major contribution perhaps lay in his teaching. By great good fortune one of his pupils was Oliver Rackham, who was thus introduced to historical ecology. Dr Rackham has developed and extended Coombe’s approach and in numerous books and papers brought it to the attention of a wider audience. By this and other routes the holistic studies which Coombe pioneered came to influence an entire generation of ecologists, and to change the way in which we all look at the English countryside. PUBLICATIONS BY DR D. E. COOMBE 1947 The Sixty-one Society. Christ’s College Magazine Easter Term 1947: 16-17. 1951 (with F. White) Notes on calcicolous communities and peat formation in Norwegian Lappland. Journal of Ecology 39: 33-62. dz The Wordsworths and botany. Notes and queries 5 July 1952: 298-299. Clapham, A. R., Tutin, T. G. & Warburg, E. F. Flora of the British Isles. [Review.] The Cambridge Review 74: 88-90. Ellenberg, Heinz. Landwirtschaftliche Pflanzensoziologie. Vol. 1. Unkrautgemeinschaften als Zeiger fiir Klima und Boden. [Review.] Journal of Ecology 40: 409-410. 1954 Carex humilis Leyss, in C. D. Pigott & S. M. Walters, On the interpretation of the discontinuous distributions shown by certain British species of open habitats. Journal of Ecology 42: 111-113. Ancient earthworks. Letter to The Times 26 April 1954: 7. Early Devon botanists, in W. Keble Martin & T. J. Wallace (eds), Forty-sixth report on the botany of Devon. Report and transactions of the Devonshire Association for the Advancement of Science, Literature and Art 86: 284. OBITUARIES 365 1956 (with L. C. Frost) The heaths of the Cornish serpentine. Journal of Ecology 44: 226-256. (with L. C. Frost and mineralogical data by M. le Bas & W. Watters) The nature and origin of the soils over the Cornish serpentine. Journal of Ecology 44: 605-615. Biological Flora of the British Isles. Impatiens parviflora DC. Journal of Ecology 44: 701-713. Notes on some British plants seen in Austria. Veroffentlichungen des Geobotanischen Instituts Riibel in Ziirich 35: 128-137. Botany, in Britannica Book of the Year 1956, pp. 88-89. Encyclopaedia Britannica, London. 1957 The spectral composition of shade light in woodlands. Journal of Ecology 45: 823-830. Botany, in Britannica Book of the Year 1957, pp. 82-83. Encyclopaedia Britannica, London. 1958 Botany, in Britannica Book of the Year 1958, pp. 84-85. Encyclopaedia Britannica, London. 1959 (with G. C. Evans) Hemispherical and woodland canopy photography and the light climate. Journal of Ecology 47: 103-113. (with F. H. Perring & S. M. Walters) Lythrum hyssopifolia L. Proceedings of the Botanical Society of the British Isles 3: 286-288. (with F. H. Perring & S. M. Walters) Lythrum hyssopifolia L. [Abstract.] Proceedings of the Botanical Society of the British Isles 3: 327. Botany, in Britannica Book of the Year 1959, pp. 94-95. Encyclopaedia Britannica, London. 1960 An analysis of the growth of Trema guineensis. Journal of Ecology 48: 219-231. (with G. C. Evans) Hemispherical photography in studies of plants. Medical and Biological Illustration 10: 68-75. 1961 Trifolium occidentale, a new species related to T. repens L. Watsonia 5: 68-87. 1962 (with W. Hadfield) An analysis of the growth of Musanga cecropioides. Journal of Ecology 50: 221-234. 1965 (with P. Bell) Strasburger’s textbook of botany, rewritten by R. Harder, W. Schumacher, F. Firbas & D. von Denffer, translated from the 28th German edition by P. Bell & D. Coombe. Longmans, Green & Co., London. 1966 The seasonal light climate and plant growth in a Cambridgeshire wood, in R. Bainbridge, G. C. Evans & O. Rackham (eds), Light as an ecological factor, pp. 148-166. Blackwell Scientific Publications, Oxford & Edinburgh. 1967 (with P. Morisset) Further observations on Trifolium occidentale. Watsonia 6: 271-275. 1968 Trifolium L., in V. H. Heywood (ed.), Flora Europaea, Notulae systematicae ad Floram Europaeam spectantes. Feddes Repertorium 79: 54. Trifolium L., in T. G. Tutin et al. (eds), Flora Europaea, vol. 2, pp. 157-172. Cambridge University Press, Cambridge. J. A. Paton. Wild flowers of Cornwall and the Isles of Scilly. [Review.] Cornwall Naturalists’ Trust Newsletter October 1968. 1969 In free time. Letter to The Times 2 December 1969: 11. 1970 The editions of A week at the Lizard by the Revd Charles Alexander Johns. Journal of the Society for the Bibliography of Natural History 5: 259-269. Which “Week at the Lizard”? The Lizard 4(2): 11-17. (with D. P. Dymond, M. D. Hooper & O. Rackham) The hedge’s worth and age. Letter to The Times 6 April 1970: 9. Reprinted in K. Gregory, The first cuckoo: a selection of the most witty amusing and memorable letters to The Times 1900-1975, pp. 303-305. George Allen & Unwin & Times Books, London, 1976 (hardback) and Unwin Paperbacks, 1980 (paperback). 366 OBITUARIES 1973 The prostrate junipers at Gew Graze. The Lizard 5(1): 7-12. Thomas Johnson, in Dictionary of Scientific Biography, vol. 7, pp. 146-148. C. Scribner’s Sons, New York. 1974 An anthology of very wild flowers round the coast. The Lizard 5(2): 12. 1975 An anthology of very wild flowers round the coast. Part 2: ‘I sent thee late a rosy wreath’. The Lizard 5(3): 15-17. 1976 (with P. Bell) Strasburger’s textbook of botany, rewritten by D. von Denffer, W. Schumacher, K. Magdefrau & F. Ehrendoefer, translated from the 30th German edition by P. Bell & D. Coombe. Longman, London & New York. 1977 Air photography and plant ecology, in J.K.S. St Joseph (ed.), The uses of air photography, 2nd ed., pp. 86- 102. John Baker, London. The Lizard, Cornwall, in D. A. Ratcliffe (ed.), A Nature Conservation Review, vol. 2, pp. 146-148. Cambridge University Press, Cambridge. Miltonic plants in Library Court. Christ’s College Magazine May 1977: 3-6. 1979 A botanical hazard at the Lizard. The Lizard 6(3): 15. 1981 (with A. F. G. Douse & C. D. Preston) The vegetation of Ringmere in August 1974. Transactions of the Norfolk and Norwich Naturalists’ Society 25: 206-217. 1982 The Rottingdean Limonium. |Abstract.] Watsonia 14: 228. 1983 (with L. C. Frost & M. H. Martin) The effects of the cessation of grazing and of drought on the grasslands in the Caerthillian valley. University of Bristol Lizard Project Report no. 9. The University, Bristol. Sir John Plumb’s mastership. Christ’s College Magazine 72: 8-10. 1984 Stony ground. The Lizard 7(1-2): 28. Grid reference SW 752182. The Lizard 7(1-2): 31. 1987 Of Milton and mulberries. Christ’s College Magazine 76: 15-20. Spiked speedwell, soil stripes and polygons, and the vanishing chalk heaths of Cambridgeshire. Nature in Cambridgeshire 29: 26-37. Longevity of spores of some terrestrial Jsoetes species. [Abstract.] B.S.B.I. News 45: 38. 1988 More about stone stripes, polygons and fairy rings. Nature in Cambridgeshire 30: 13-15. Hayward, John. A new key to wild flowers. [Review.] Endeavour n.s. 12: 48-49. Propogation of Polygonum maritimum L. [Abstract.] B.S.B.I. News 49: 43. 1989 Vegetative propogation of Lythrum hyssepifolia. [Abstract.] B.S.B.I. News 51: 46. [See also correction in B.S.B.I. News 52: 3 (1989).] 1990 (with T. J. N. Mills & P. Upchurch) The history of Madingley brickpits: contrasts in succession. Nature in Cambridgeshire 32: 3-14. Lythrum hyssopifolia L.: two postscripts. [Abstract.] B.S.B.I. News 54: 47. (with G. Crompton) Ferula communis L.: mythology and biology. [Abstract.] B.S.B.I. News 54: 47. 1992 Cymbalaria muralis forma toutonii. [Abstract.] B.S.B.I. News 60: 62. Ophioglossum lusitanicum: 31 years in cultivation. [Abstract.] B.S.B.I. News 60: 62. OBITUARIES 367 1994 ‘Maritime’ plants of roads in Cambridgeshire (v.c. 29). Nature in Cambridgeshire 36: 37-60. Trifolium occidentale Coombe, Trifolium ochroleucon Hudson, Trifolium ornithopodioides L., Trifolium squamosum L. and Trifolium suffocatum L., in A. Stewart, D. A. Pearman & C. D. Preston (comps & eds), Scarce plants in Britain, pp. 414-418. Joint Nature Conservation Committee, Peterborough. (with C. Chatters) Scilla autumnalis L., in A. Stewart, D. A. Pearman & C. D. Preston (comps & eds), Scarce plants in Britain, p. 374. Joint Nature Conservation Committee, Peterborough. (with S. J. Leach) Trifolium glomeratum L., in A. Stewart, D. A. Pearman & C. D. Preston (comps & eds), Scarce plants in Britain, p. 413. Joint Nature Conservation Committee, Peterborough. 1996 (with O. Rackham) Madingley Wood. Nature in Cambridgeshire 38: 27-54. 1999 Erica vagans L., in M. J. Wigginton (comp. & ed.), British Red Data Books 1 Vascular plants, 3rd ed., pp. 139-140. Joint Nature Conservation Committee, Peterborough. (with L. E. Friday & P. J. Grubb) The Godwin Plots at Wicken Fen: a 55-year record of the effects of mowing on fen vegetation. Nature in Cambridgeshire 41: 3346. OBITUARY NOTICES An obituary appeared in The Independent newspaper on 30 July 1999 and further obituaries will be published in Christ’s College Magazine and Nature in Cambridgeshire. ACKNOWLEDGMENTS I am very grateful to Mrs G. Crompton, Dr G. C. Evans, Lady Rosemary FitzGerald, Dr P. J. Grubb, P. H. Oswald, Mrs J. A. Paton, D. A. Pearman, Dr O. Rackham and the late Dr H. L. K. Whitehouse for help in preparing this obituary or for comments on a draft of the text. C. D. PRESTON LEWIS CHARLES FROST (1926-1998) Lewis Frost was very proud to have been born and bred in Norfolk and remained deeply attached to the county throughout his life even though in later life he rarely returned. He came from farming stock with a Belgian mother to whom he was very close. He was to have been called Louis but the Registrar disapproved of foreign names and wrote Lewis on the birth certificate. Hence, he was known as Louis to his family, as Lewis to most in the academic world, to those of us on more familiar terms as Lew and Uncle Lew to our children. At one time it was intended that he should enter the priesthood and whilst he continued his connection with the church during his early undergraduate days at Emmanuel College, Cambridge, it would seem that national service followed by his academic successes in the natural sciences tripos persuaded him to follow a scientific and academic career. After attending school in Bungay he joined the Royal Army Educational Corps in 1944 and when he was demobbed in 1948 went up to Cambridge. His career in Cambridge was distinguished. He obtained First class in all three parts of the tripos and this led to starting a Ph.D in genetics in the school of agriculture in 1952 under the supervision of D. G. Catcheside. After one year he was appointed Demonstrator in Botany (a post he held until 1957); H. L. K. Whitehouse became his supervisor when Catcheside left Cambridge. His Ph.D was awarded in 1956. In October 1957 he was appointed Lecturer in Genetics within the Department of Botany at Bristol University. In Bristol Lewis taught genetics to a wide variety of undergraduates. He was for many years the only genetics lecturer in the university, so he taught medical and veterinary students as well as biologists. He taught genetics until his retirement in 1983 with a further three years of part-time re- engagement. However, he also developed courses on nature conservation as his research interests in genetics waned and were overtaken by his deep commitment to conservation ecology. Indeed, Lewis will probably be remembered by most biologists not for his genetics teaching, but for his ecological and conservation activities primarily within two areas: the Avon Gorge and the Lizard in Cornwall. 368 OBITUARIES In Cambridge Lewis had teamed up with David Coombe and together they developed an interest in the Lizard area of Cornwall; their joint work commenced whilst Lewis was still an undergraduate, during a field course led by David Coombe in June 1950. Lewis’s passion for the Lizard (and the associated Cornish cream teas) lasted throughout the rest of his life and he would often visit the Lizard two or three times a year. In 1956 David Coombe and Lewis published their classic paper on the heaths of the Cornish serpentine - even today this is still regarded as the definitive paper and known to most ecologists as “Coombe & Frost”. This was followed by another paper on the soils of the Lizard in which they made the startling discovery of wind blown deposits overlying the serpentine of the Goonhilly Downs. Today it is accepted that such loess deposits are wide-spread in the south west but at the time many treated the findings with scepticism. Lewis took groups of Bristol students to the Lizard to imbue in them his enthusiasm and idiosyncrasies. Those who attended these trips soon found themselves referred to as “Lizard kiddies”. A favourite diversion was the grade one cream teas: these were taken at the farm on the main road to the Lizard where a side road is signposted to “Grade 1’. A highlight of every visit to the Lizard was the systematic visitation of all the serpentine turners in the village and the bargaining with them for a few pence off the choice items of serpentine ornaments. Throughout his career Lewis was a great battler for conservation causes. He raised considerable funds for the Cornwall Wildlife Trust’s purchase of land and establishment of a nature reserve on the Lizard and the Cornwall Wildlife Trust made him an honorary member. He inveigled, bullied, and persistently pestered the Nature Conservancy Council to do the same, with the result that eventually substantial areas of the Lizard became National Nature Reserves and Sites of Special Scientific Interest. In Bristol he set up the Lizard Appeal for funds to carry out research. This received good support and allowed numerous conservation and research projects to be undertaken, which are documented in a long series of over 20 Lizard Reports, and the engagement of a series of Ph.D students including John Hopkins, Marion Hughes and Andrew Byfield. His own research concentrated on the rarer plant species such as Trifolium incarnatum ssp. molinerii and in particular Herniaria ciliolata ssp. ciliolata. During his early life in Bristol Lewis was a sub-warden in Wills Hall and he recognised the Avon Gorge as a vast untapped study area on the doorstep of Wills Hall. He quickly formed an informal field club and together with a trusty and dedicated group of students set about a wide range of botanical studies, including detailed examination of the ecology of species such as Trinia glauca, Arabis scabra, Hornungia petraea and Carex humilis. These interests culminated in the establishment of the Avon Gorge Appeal Fund, this attracted considerable sums of money and helped finance innumerable studies on the Avon Gorge and its rare plants, with Clive Lovatt and Libby Houston as collaborators. After his retirement Lewis remained as an Honorary Research Fellow in the School of Biological Sciences and continued producing research reports right up to the time of his going into hospital. Locally Lewis played an active role in the early days with the Gloucestershire Wildlife Trust and took part in many of the early surveys of sites later to become nature reserves in that county. Another highlight was his long involvement with the Badgeworth Nature Reserve and the Adder’s- tongue Spearwort. The fact that this reserve was officially declared the smallest nature reserve in the world and appeared as such in the Guinness Book of Records gave him great pleasure. Most of us will remember Lewis for his dry sense of humour; even a few days before he died we were able to have a joke together and that impish glint came to his eyes. We will also remember the ways that Lewis always savoured a bargain. It gave him immense pleasure to be able to bargain a few pence off virtually any purchase from relatively expensive hi-fi equipment, to a new pair of shoes, even to a packet of biscuits. But this aspect of his character changed totally with friends and colleagues who were in trouble when he would be enormously generous, helpful and considerate. Those of us who were fortunate to experience this side of Lewis were indeed privileged. M. H. MARTIN & M. D. AMES Postscript: At the time of his death David Coombe was in the process of writing an obituary for Lewis Frost. Unfortunately due to his illness he did not manage to write this, but the authors are aware of how much more David would have included. We are however, very grateful to M.H.M. and M.D.A. for taking on the writing of the obituary in these circumstances. Ed. OBITUARIES 369 HUGH ALLISON LANG 1920-1999 Dr Hugh Lang died on 19 November 1999 in Newton Stewart where he had been a much respected single-handed family doctor for 27 years. He joined the B.S.B.I. in 1973 although he had botanical interests going back many years. He was essentially a field botanist and very much an individualist who became well known to members of the society in Scotland, where he had served on the B.S.B.I. Committee. The late Olga Stewart described him in her “Flowering Plants of Kirkcudbrightshire” as a “mountaineer and mountain plant botanist” which summed up his love of wild and remote places and the plants which grew there. He was a large man who had great strength and stamina and enjoyed scrambling up to and exploring particularly inaccessible habitats and was an expert photographer who recorded most of his finds on film. He was also a keen observer of many aspects of the natural world and these included astronomy, geology and meteorology. He was born in Sauchiehall Street, Glasgow and entered Glasgow University in 1937. After a science degree he obtained his medical degree in 1943 and immediately joined the airborne division of the Royal Army Medical Corps and was posted to India. He was one of the first doctors parachuted into the Japanese prisoner of war camps in Burma and saw at first hand the appalling condition of the Allied prisoners there. He was demobbed in 1947 and after entering General Practice in Stranraer and Barrhill, he eventually set up his own practice in Newton Stewart in 1958 helped by his wife Anne, whom he married in 1953. He was ideally suited to attending emergencies in the hills and put his organisational experience to good use, being instrumental in founding the Galloway Mountain Rescue Team and in 1975 becoming its first chairman. He was also a past president of the local Rotary Club. As a side line he made an impressive collection of aircraft wreckage collected from military jet crash sites. He retired from his practice in 1985. Hugh knew the rocky coastline and hills of Galloway and the location of its botanical rarities better than anyone and was especially fond of the crags of the Merrick, the highest point in Southern Scotland. He told me how early on he had met the late Dr Humphrey Milne-Redhead, another doctor/botanist and vice-county recorder, and showed him a willow collected high in the hills. Humphrey pronounced it to be Salix lapponum (Downy Willow) an extremely rare plant of the district and was duly impressed. He exhorted Hugh to take his botany seriously and this paid off for in 1973 he added Saxifraga oppositifolia (Purple Saxifrage) to the flora of the Galloway hills; a third site for Southern Scotland. However, Hugh’s great discovery had come previously in 1971 when exploring the remote hills of Easter Ross, he found Artemisia norvegica (Norwegian Mugwort) in a third British station. It is Hugh’s photograph of this species which appears as the frontispiece in Ursula Duncan’s “Flora of East Ross-shire”. It is intriguing to note that it was Humphrey Milne-Redhead who found the second British site for this species in West Ross 17 years before. His enthusiasm was such that only a few months before his death he had managed to photograph Arabis alpina (Alpine Rock-cress) at its recently discovered second site in the Cuillins. Hugh served on the Torrs Warren reserve committee and was undertaking a photographic survey of this nationally important dune system. He also monitored the sole remaining colony of Woodsia ilvensis (Oblong Woodsia) in the Moffat Hills. The arctic and the history of its exploration had always fascinated him and on retirement he was at last able to visit the high arctic regions of North America and particularly Greenland. He didn’t miss a single summer visit in the 14 years between his retirement and death. His tiny blue tent was a constant feature of these expeditions and there was always speculation as to how he managed to fit his large frame into it. His ultra-long sturdy home-made trekking pole gave him extra stability especially when crossing glacial torrents. He had a very good eye for the small plants of the arctic flora and noted anything which looked unusual including the occasional lichen and fungus. His collecting was unorthodox but effective, using a notebook in his hip pocket and his collections were sent to the Botanical Museum of the University of Copenhagen and to Geoffrey Halliday at Lancaster University. It was from Hugh’s high quality photograph of a Pedicularis sp. (a Lousewort) from north-western Greenland that Geoffrey identified P. sudetica, new to the Greenland flora. It had been collected previously, but was misidentified as the flower colour fades in the herbarium and it resembled a closely related species. He was delighted on his last expedition to see and photograph Potentilla stipularis, a very rare and elusive Siberian and Greenland species 370 OBITUARIES of Cinquefoil. His archaeological interests were stimulated by ancient Eskimo remains and he was in correspondence with the Danish authorities about possible new sites. On expeditions Hugh was a steadfast companion who never let bad conditions get on top of him. He had an inexhaustible fund of verse and quotation to suit any situation and this was delivered in his own inimitable way. This was sure to ‘break the ice’ with strangers and boost flagging morale among friends. It was this same side of his personality that made his slide shows at the Scottish Exhibition meetings so memorable with descriptions which could reduce the audience to helpless laughter. He loved Drambuie and his ceremonial decanting from bottle to plastic container in the airport departure lounge to save weight, was a regular pre-expedition feature, causing many an eyebrow to be raised. He was a member of the Arctic Club and served on the committee of the Scottish Arctic Club. He is survived by his son Angus. Sadly, his wife Anne predeceased him four years earlier. R. W. M. CORNER BERTRAM EVELYN SMYTHIES (1912-1999) Bertram Evelyn Smythies, usually known as Bill, died on 27 June 1999, aged 86. He had been a member of the B.S.B.I. since 1974 and was elected a Fellow of the Linnean Society in 1969. Bill was born in India on 11 July 1912, at Naini Tal, Uttar Pradesh. His father worked for the Forest Service as the Assistant Conservator of Forests for the United Provinces and his mother was a noted tiger huntress. Thus, he was introduced to natural history at a very early age. He was sent to a preparatory school back in England at the age of eight, but during the holidays stayed with his grandfather (who had himself worked in India teaching botany at the Dehra Dun forest school) in North Devon. Some years ago, Bill sent me a Christmas card by Elizabeth Twining, daughter of Richard Twining (tea merchant) and Elizabeth Smythies, illustrating the Grossulariaceae. He added the note that “My grandfather had the 2 vols of our ancestor’s work in his library, and I used to lie on the floor and learn the names of the “Natural Orders’ by heart - no TV or even Radio in those days! Sadly the artistic gene was not passed to me, but Elizabeth Twining’s work may have aroused my interest in bot. Ill., along with Fitch’s line drawings in ‘Bentham & Hooker’”. It was no surprise that he went up to Balliol College, Oxford to read Botany and Forestry. It was here that he met G. K. Yates, a leading bird photographer, who taught Bill bird photography and how to prepare a bird skin. Bill went off camping and photographing rare birds at the nest in Caithness and Sutherland, skills he was to put to good use later. After graduating, Bill went in 1934 to work for the Colonial Forest Service in Rangoon, Burma, where he was quickly recruited by H. C. Smith to collect bird skins for the British Museum (Natural History). This led to production of the Birds of Burma, published by the Mission Press, Rangoon, in 1941. Before leaving Burma in February 1942, Bill managed to rescue the copper printing blocks from the abandoned Mission Press and made possible the production of an enhanced second edition in 1953. Some copies of the original edition were sold to expatriate Europeans, but most were taken back to Japan where they were destroyed in an air raid: it is thus a sought-after rare work. After the war, Bill returned to Burma, but left in 1948 when the country left the Commonwealth. He then took up a post in the forest service in Sarawak, publishing a 300-page Annotated Checklist of the Birds of Borneo in 1957. This was the foundation for his glorious Birds of Borneo (1960), a classic work that was later revised with a second edition in 1968 and a third in 1981. Although he also worked for a while in Brunei in the 1950s, he returned to Sarawak to produce Sarawak Forest Trees, published in 1965 by the Borneo Literature Bureau. It aimed to enable foresters to identify tree species and is a fundamental work for the new Tree Flora of Sabah and Sarawak currently being prepared. In 1964 Bill married Jill Rogers and shortly afterwards retired from the Sarawak Forest Service to live in Estepona, southern Spain. It was here that he pursued his love of botany, taking part in some of the John Carr Expeditions to Andalucia. He travelled widely in the area and produced a Flora of Spain: Checklist of Vascular Plants, published in three parts (1984, 1984, 1986) as OBITUARIES 374 volume three of the periodical Englera of the Botanischen Garten und Botanischen Museum Berlin-Dahiem. This phenomenal catalogue enumerates 7138 taxa compared to the 5948 in Flora Europaea (and an estimated 8600 in Flora Iberica) and illustrates Bill’s industry. This increase was not achieved by simply taking a more radical view of species, but by a careful review of the extensive recent literature and consulting specialists. However, it is not for this fundamental work that botanists most remember him, but for Flowers of South-West Europe: a field guide, a very successful popular botanical guide produced in collaboration with Oleg Polunin and published in 1973 by Oxford University Press. It is a credit to them that even today it is still also a best seller in its Spanish edition. Part of the success is due to the clear, double-page spread line drawings of plants by Jill Smythies. Jill became a competent botanical artist as is also shown by her archive of drawings and paintings of Spanish plants presented after her death in 1994 by Bill to the University of Reading Herbarium. Following an accident to Jill’s right hand which ended her drawing, Bill endowed the Linnean Society of London with the Jill Smythies award for botanical drawing which contributes to plant identification. After moving back from Estepona to Surrey, Bill became a frequent visitor to the Linnean Society's rooms. Although always of few words, he was a good correspondent, and it was here that his wit and humour was seen: “In my grandfather’s day, a century ago at Dehra Dun the mail was carried by a dak runner trotting along from village to village; for long distances horse transport was used. I suspect the Italian postal service is at the same state of development - your card postmarked Rome 9 Dec. took three weeks to get here. In the days of Imperial Rome it would surely have taken less!” With Bill’s death, the story does not end, for botanists will continue to benefit through the legacy of his library to the Linnean Society of London and the greater part of his estate to botanical charities. S: L. JURY ERIC VERNON WATSON 1914-1999 Dr Eric Watson, the distinguished bryologist and greatly respected university lecturer died at his home in Goring on 25 October 1999. He had been his usual self and working in the University during the summer. Although he found lately that he had been getting very tired, he had kept up his botany at home, almost to the last. He was diagnosed as having been suffering from leukaemia. He joined the staff of the Department of Botany at the University of Reading in 1946 as Lecturer, later becoming Senior Lecturer before his retirement in 1979. He was very well known in bryophyte circles, especially for his books (The Structure and Life of Bryophytes, Hutchinson Universal Library, London, 3 eds and British Mosses and Liverworts, Cambridge University Press, 3 eds). These were particularly successful in getting students of all ages interested in and able to understand and know bryophytes. He had amassed a considerable bryophyte herbarium, putting specimens in both the University collection and his own private one (this is left to the Royal Botanic Garden, Edinburgh; the University of Edinburgh was his old alma mater). However, it must be said that in fact he was an excellent all-round botanist and knew the British flora well, having been a member of the B.S.B.I. for almost 50 years. I remember vividly ‘demonstrating’ to students in his anatomy classes in the 1970s, when the class was expected to trace the vascular tissues through a plant node by making a series of sections and interpreting the results as a 3-D drawing. All the examples (Vicia, Lamium and strawberry runners) came from his garden. His teaching abilities were such that most students were inspired to work incredibly hard to do the most difficult example (Vicia, the stipules complicate the structure enormously). His tutees were regularly invited out to his house for supper and a slide show. He held regular moss forays starting with morning coffee at his house and ending back for afternoon tea. These invariably started with the species to be found in the garden and over the years a considerable checklist of species was made. Generations of students will remember him with affection and many had kept in touch with him — he was a great letter writer. 372 OBITUARIES Dr Watson was a keen natural historian, accomplished ornithologist (founder member of Reading Bird Club after the war), gardener, artist (watercolourist: he was a leading member of Reading Guild of Artists), golfer and earlier a very good tennis player. Taking part in so many local societies and activities made him well-known locally. He will be greatly missed. We send his widow Joyce and family our condolences. Sek. JURY Watsonia 23: 373-375 (2000) 373 Report ANNUAL GENERAL MEETING, 8 MAY 1999 The Annual General Meeting of the Society was held at the Royal Pharmaceutical Society, | Lambeth High Street, London, at 11.50 a.m. The President, Mrs M. Briggs took the Chair in the presence of 102 members. Apologies for absence were read and Minutes of the 1998 Annual General Meeting, published in Watsonia 22: 297-298 (1999), were accepted as correct, approved and signed by the President. REPORT OF COUNCIL The President took members through the Report of the Council, which had been previously circulated to members, commenting on the main achievements of the Society during the year. She mentioned especially the progress of Atlas 2000 and the setting up, with the Joint Nature Conservation Committee (J.N.C.C.) and the Country Agencies, of the Threatened Plants Database (T.P.D.B.) and the appointment of Mr A. Lockton and Mrs S. J. Whild to run this and act as B.S.B. I. Co-ordinators. She finally thanked all those who worked so hard for the Society on a voluntary basis and congratulated Mr R. G. Ellis on producing another excellent Report. The adoption of the Annual Report was proposed by Mr C. R. Boon, seconded by Mr R. M. Walls and accepted unanimously. HON. TREASURER’S REPORT AND ACCOUNTS The Treasurer reported that as usual, only summary accounts had been published in the Annual Report, full sets of accounts were available at the meeting and on application to him. He commented that only 30% of the resources used in the Society’s activities were derived from subscriptions, much of the remainder coming from other bodies in the form of grants for Atlas 2000, T.P.D.B., etc. He was again pleased to report a small surplus for the year but, with the disappointing returns from our recent Covenant campaign, in order to continue to cover expenditure by income a modest increase in subscriptions was needed. He then proposed the adoption of the report, which was seconded by Mr J. Ounsted and carried unanimously. The President thanked the Treasurer for all the hard work he put into managing our financial affairs so competently. RULE CHANGES In introducing the proposed rule changes, the President commented that the change to Rule 2 was aimed at allowing the Society to involve itself with the study of British taxa in Europe and elsewhere, while the change to Rule 4 would replace the now disbanded Conservation Committee with the newly formed Science and Research Committee as a Permanent Working Committee. Members were assured that the Society would continue to take an interest in plant conservation, mainly through the S & R Committee, and the revised Terms of Reference for that Committee were available at the meeting. Proposed amendment to Rules 2 ( Aims of the Society) That Rule 2 of the Society’s Rules be deleted and the following be substituted as recommended by Council: 374 REPORT AIMS AND OBJECTS 2. The aims of the Botanical Society of the British Isles shall be those of a learned society of professional and amateur botanists dedicated to the study and conservation of the British and Irish vascular plant and charophyte flora. The objects of the Society shall be a) To promote the study of and interest in the British and Irish vascular plant and charophyte flora. b) To support, encourage, carry out and participate in research into the taxonomy, ecology, biogeography and conservation of the British and Irish flora and to co-operate with Continental and other botanists in matters of mutual interest and concern. c) To provide a forum for the dissemination of knowledge about the British and Irish flora, by means of publications, Information Technology, meetings and advice. d) To support and encourage the training of botanical skills. e) To collaborate with statutory and other voluntary bodies engaged in the conservation of the British and Irish vegetation and flora. f) To maintain a botanical recording network in Britain and Ireland. g) To provide, for the benefit of members, access to identification and other expert services for the study of wild plants. h) To support studies of the British and Irish flora or related taxa elsewhere, in furtherance of the Society’s aims. The change to Rule 2 as set out above was proposed by S. L. Jury, seconded by E. F. Greenwood and carried unanimously Proposed amendment to Rule 4 (Permanent Working Committees) That Rule 4 of the Society’s Rules be amended. Council, in consultation with the Committees concerned, decided that the furtherance of the Society’s aims for conservation would be better achieved by the Committee reorganisation as follows: (c) Conservation be deleted and replaced by (c) Science and Research The change to Rule 4 as set out above was proposed by Dr N. K. B. Robson, seconded by Dr F. H. Perring and carried unanimously INCREASE IN SUBSCRIPTIONS The Treasurer proposed an increase from January Ist 2000 as follows (current rates in parenthesis): For members in Britain and Ireland Ordinary £20.00 (£18.00) Family £2.00 (£2.00) Junior £10.00 (£9.00) Senior £13.00 (£12.00) Institutional Member £20.00 (£18.00) For other overseas members £22.00 (£20.00) This was seconded by Mr A.O. Chater and carried unanimously. ELECTION OF PRESIDENT-ELECT The President commented that it was a pleasant duty to propose Council’s nomination of Dr G. Halliday, who, as Chairman of Records Committee, had been involved with many of the Society’s REPORT a5 activities in recent years. The nomination was seconded by Mr M. Walpole and carried unanimously with acclamation. ELECTION OF VICE-PRESIDENT The election of Mr D. J. McCosh, as nominated by Council, was proposed by Dr P. Macpherson, seconded by Mr R. D. Pryce and carried unanimously. RE-ELECTION OF HON GENERAL SECRETARY AND HON. TREASURER Council had nominated Mr R. G. Ellis and Mr M. E. Braithwaite for re-election to their respective posts. This was proposed by Mr P. H. Oswald, seconded by Miss A. P. Conolly and carried unanimously with acclamation. REAPPOINTMENT OF EDITORS, INDEXERS AND REPRESENTATIVES OF COMMITTEES In proposing the reappointment of Editors, Indexers and Representatives of Committees, the President praised them for the hard work they carried out, voluntarily, on behalf of the Society. She gave special thanks to Mr P. J. Fry (Assistant Secretary) who gave so freely of his time in answering postal and telephone queries at the B.S.B.I. desk in the Natural History Museum, and to Mr P. Thomson (Council Minuting Secretary). These were received with acclamation. ELECTION OF COUNCIL MEMBERS In accordance with Rule 11 nominations had been received for Mr P. H. Oswald (proposed Dr C. D. Preston, seconded Mr R. G. Ellis), Mrs J. M. Robertson (proposed Mr A. C. Jermy, seconded Mr R. M. Walls) and Dr B. S. Rushton (proposed Mrs M. Briggs, seconded Mr R. G. Ellis). Profiles had been published and election of these members was carried unanimously. ELECTION OF HONORARY MEMBERS The President mentioned that three members had been nominated for Honorary Membership of the Society this year. All had, in different ways, made outstanding contributions to the Society. Sponsors for all three gave short appreciations of their respective candidates (to be published in BSBI News) and their election was carried unanimously with warm applause. The three new Honorary Members (and sponsors) were: Mr E. J. Clement (Mr T. B. Ryves, read by Mr R. G. Ellis), Dr B. S. Rushton (Mr M. Walpole) and Prof. G. A. Swan (Dr A. J. Richards, read by Mrs M. Briggs). RE-ELECTION OF HONORARY INDEPENDENT EXAMINER The President warmly thanked Mr John Coats, of Greaves West & Ayre, Chartered Accountants, Berwick upon Tweed for his exemplary examination of our accounts. His re-election was proposed by Mr M. Walpole, seconded by Mrs M. Lindop and carried unanimously. ANY OTHER BUSINESS In reply to a written query by Dr T. C. G. Rich concerning the non-availability of a pre-publication offer for the new Red Data Book published by J.N.C.C., the President reported that attempts to persuade J.N.C.C. to give a discount to B.S.B.I. members had been unsuccessful. There being no other business, the meeting closed at 12.47 p.m. GWYNN ELLIS — CHWNIADARWN TS 72 West Cornwall Scilly East Cornwall South Devon North Devon South Somerset North Somerset North Wiltshire South Wiltshire Dorset Isle of Wight South Hampshire North Hampshire West Sussex East Sussex East Kent West Kent Surrey South Essex North Essex Hertfordshire . Monmouthshire . Glamorgan . Breconshire . Radnorshire . Carmarthenshire SCOTLAND V3. 74. 1D: 76. TT. 78. 79. 80. 81. 82. 83. 84. S2; IRELAND a. H2. M3: H4. HS. H6. H7. H8. H9. H10. H11. H12. H13. H14. Dumfriesshire Kircudbrightshire Wigtownshire Ayrshire Renfrewshire Lanarkshire Peeblesshire Selkirkshire Roxburghshire Berwickshire East Lothian Midlothian West Lothian Fife South Kerry North Kerry West Cork Mid Cork East Cork Co. Waterford South Tipperary Co. Limerick Co. Clare North Tipperary Co. Kilkenny Co. Wexford Co. Carlow | Laois NAMES OF VICE-COUNTIES IN WATSONIA pa 22, 29: 24. 2D: 26, 27 28. Zo: 30. eae a2. 33: 34. 36. ah 38. 39. 40. 40. D2: 45. 46. 47. 48. 49. 86. 87. 88. 89. 90. 91. v2, 93; 94. 93: mG. 96b. a 98. H15. H16. BET, H18. lathe aa H20. H2 1. H22. HZ3: H24. H25: H26; R27. H28. ENGLAND Middlesex Berkshire Oxfordshire Buckinghamshire East Suffolk West Suffolk East Norfolk West Norfolk Cambridgeshire Bedfordshire Huntingdonshire Northamptonshire East Gloucestershire West Gloucestershire Herefordshire Worcestershire Warwickshire Staffordshire Shropshire Shropshire South Lincolnshire Pembrokeshire Cardiganshire Montgomeryshire Merionethshire Caernarvonshire Stirlingshire West Perthshire Mid Perthshire East Perthshire Angus Kincardineshire South Aberdeenshire North Aberdeenshire Banffshire Moray Easterness Nairnshire Westerness Main Argyll South-east Galway West Galway North-east Galway Offaly Co. Kildare Co. Wicklow Co. Dublin Meath Westmeath Co. Longford Co. Roscommon East Mayo West Mayo Co. Sligo North Lincolnshire . Leicestershire . Rutland Nottinghamshire Derbyshire . Cheshire South Lancashire . West Lancashire . South-east Yorkshire . North-east Yorkshire . South-west Yorkshire . Mid-west Yorkshire . North-west Yorkshire . Co. Durham . South Northumberland . North Northumberland . Westmorland . Furness . Cumberland . Isle of Man . Denbyshire . Flintshire . Anglesey . Dunbarton . Clyde Isles . Kintyre . South Ebudes . Mid Ebudes . North Ebudes . West Ross . East Ross . East Sutherland . West Sutherland . Caithness . Outer Hebrides . Orkney . Shetland . Co. Leitrim . Co. Cavan . Co. Louth . Co. Monaghan . Fermanagh . East Donegal . West Donegal . Tyrone . Co. Armagh . Co. Down . Co. Antrim . Co. Londonderry BOTANICAL SOCIETY OF THE BRITISH ISLES (B.S.B.1.) BOBI The B.S.B.I. traces its origin to the Botanical Society of London founded in 1836 and has a membership of 2,850. It is the major source of information on the status and distribution of British and Irish flowering plants and ferns. This information, which is gathered through a network of vice-county recorders, is the basis for plant atlases and for publications on rare and scarce species and is vital to botanical conservation. The Society published Atlas of the British flora in 1962 and revised editions and is working towards a new atlas, to be completed in 2000. It organises plant distribution surveys, publishes handbooks on difficult groups of plants and has a panel of referees available to members to name problematic specimens. The B.S.B.I. arranges conferences and field meetings throughout the British Isles and, occasionally, abroad. It welcomes as members all botanists, professional and amateur alike. Details of membership and any other information about the Society may be obtained from: The Hon. General Secretary, Botanical Society of the British Isles, c/o Department of Botany, The Natural History Museum, Cromwell Road, London SW7 5BD. BSBI handbooks Each handbook deals in depth with one or more difficult groups of British and Irish plants. No. | Sedges of the British Isles A. C. Jermy, A. O. Chater & R. W. David. Revised edition, 1982. 272 pp., with descriptions, drawings and distribution maps for all 73 species of Carex. Paperback. ISBN 0 901158 05 4. No. 2 Umbellifers of the British Isles T. G. Tutin. 1980. 200 pp., with descriptions and drawings of 73 species of Apiaceae (Umbel- liferae). Paperback. ISBN 0 901158 02 X. No. 3 Docks and knotweeds of the British Isles J. E. Lousley & D. H. Kent. 1981. 208 pp., with descriptions and drawings of about 80 native and alien taxa of Polygonaceae. Paperback. ISBN 0 901158 04 6. Out of print. New edition with distribution maps in preparation; orders recorded. No. 4 Willows and poplars of Great Britain and Ireland R. D. Meikle. 1984. 200 pp., with descriptions and drawings of 65 species, subspecies, varieties and hybrids of Salix and Populus. Paperback. ISBN 0 901158 07 0. No. 5 Charophytes of Great Britain and Ireland J. A. Moore. 1986. 144 pp., with descriptions and drawings of 39 species and varieties of Characeae and 17 distribution maps. Paperback. ISBN 0 901158 16 X. No. 6 Crucifers of Great Britain and Ireland T. C. G. Rich. 1991. 344 pp., with descriptions of 148 taxa of Brassicaceae (Cruciferae), 129 of them with drawings, and 60 distribution maps. Paperback. ISBN 0 901158 20 8. No. 7 Roses of Great Britain and Ireland G. G. Graham & A. L. Primavesi. 1993. 208 pp., with descriptions and drawings of 13 native and nine introduced taxa of Rosa, descriptions of 76 hybrids, and 33 maps. Paperback. ISBN 0 901158 22 4. No. 8 Pondweeds of Great Britain and Ireland C. D. Preston. 1995. 352 pp., with descriptions and drawings of all 50 species and hybrids of Potamogeton, Groenlandia and Ruppia, most of them with distribution maps; detailed introductory material and bibliography. Paperback. ISBN 0 901158 24 0. No. 9 Dandelions of Great Britain and Ireland A. A. Dudman & A. J. Richards. 1997. 344 pp., with descriptions of 235 species of Taraxacum, most of them illustrated by silhouettes of herbarium specimens; drawings of bud involucres of 139 species and 178 distribution maps. Paperback. ISBN 0 901158 25 9. No. 10 Sea beans and nickar nuts E. Charles Nelson. 2000. 156 pp., with descriptions of nearly 60 exotic seeds and fruits found stranded on beaches in north-western Europe (many illustrated by Wendy Walsh) and of the mature plants (some with drawings by Alma Hathaway), accounts of their history and folklore, growing instructions etc. Paperback. ISBN 0 901158 29 1. Other publications English names of wild flowers J. G. Dony, S. L. Jury & F. H. Perring. 1986 (2nd ed.). 126 pp. Recommended English names for British and Irish vascular plants (Latin-English and English—Latin). Paperback. ISBN 0 901158 15 1. List of vascular plants of the British Isles D. H. Kent. 1992. 400 pp. Nomenclature and sequence as in Clive Stace’s New Flora of the British Isles (1991, 1997), with selected synonyms. Paperback. ISBN 0 901158 21 6. Supplied with errata lists and Supplement 1. Alien plants of the British Isles E. J. Clement & M. C. Foster. 1994. 616 pp. Lists 3,586 recorded non-native species (of which 885 are established), with English names, frequency of occurrence, status, areas of origin, location of voucher specimens, references to published descriptions and illustrations, and selected synonyms. Paperback. ISBN 0 901158 23 2. Alien grasses of the British Isles T. B. Ryves, E. J. Clement & M. C. Foster. 1996. 234 pp. A companion volume to the last, listing over 700 non-native grasses; includes keys to bamboos and eight of the larger and more difficult genera and 29 pp. of illustrations. Paperback. ISBN 0 901158 27 5. Plant crib 1998 T. C. G. Rich & A. C. Jermy. 1998. 400 pp. An identification guide for some 325 ‘difficult’ taxonomic groups, with explanations, keys and illustrations of plant details. A4 paperback. ISBN 0 901158 28 3. British Red Data Books I Vascular plants M. J. Wigginton, ed. 1999 (3rd ed.). 468 pp. Up-to-date information on 408 taxa (including 118 microspecies) regarded as threatened in Great Britain, with details of their habitats and associated species, maps showing pre-1970, 1970-1987 and post-1987 records by 10-km squares, and 1-km square frequency maps for 63 taxa. Hardback, published by JNCC, Peterborough. ISBN 1 86107 451 4. Scarce plants in Britain A. Stewart, D. A. Pearman & C. D. Preston, comp. & ed. 1994. 518 pp. Accounts of 254 nationally scarce taxa (occurring in 16-100 10-km squares in Great Britain) and of 71 taxa formerly thought to be so, with details of their habitats and associated species, reproductive biology, changing British distribution and world range; updated distribution maps. Hardback, published by JNCC, Peterborough. ISBN 1 873701 66 7. Aquatic plants in Britain and Ireland C. D. Preston & J. M. Croft. 1997. 365 pp. Summarises the distribution, habitat and reproductive biology of 200 aquatic plants in 72 genera, with 200 distribution maps and 72 drawings. Hardback, published by Harley Books, Colchester. ISBN 0 946589 55 0. Available from the official agents for BSBI publications, Summerfield Books (Jon & Sue Atkins), Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX (Tel: 017683 41577. Fax: 017683 41687. E-mail: bsbipubs @beeb.net). . = . =, z — 7 = - =< 7 ~ -. & - - a ' = ie 7 = > — a-= a) = 7 4 ‘ - ; ih ‘ + 4 me 1 * 4, - ~~ as h - # / ° - ne * : i _ : 2 es Sia i tore = . we! a aay = ce - - ty - a a a : : 2 yoo - ‘ ‘ a 3s 7 =: 7 7 7 m 2 : 7 ‘ - : ' : : 2 = 7 = Fara - 7 * a a 7 Se : ; . int = : at ¢ : ‘ 7 a - ant - 4 ° ~ - 7 oY Ses 7 4 % p= * a ~~ 7 - z - a >» © = > =O Ce a ‘ - * Paby i * A ‘ ‘ al on je ‘ i me iN . : Me ’ + s bat I= x F ‘ i et ’ i " 7 D , a : : 7 i > 3 ' = Ca ae — a 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. 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://members.aol.com/bsbipubs/watsonia. 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: Mr D. A. Pearman, The Old Rectory, Frome St Quintin, Dorchester, Dorset DT2 OHF Plant Records: the appropriate vice-county recorder, who should then send them to Dr C. D Preston, Biological Records Centre, Monks Wood, Abbots Ripton, Huntingdon, PEI7 21S. Obituaries: Mrs M. Briggs, 9 Arun Prospect, Pulborough, West Sussex, RH20 IAL. Back issues of Watsonia are available from the official agents for BSBI Publications, Summerfield Books, (John & Sue Atkins) Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX. Tel: 017683 41577. Fax: 017683 41687. E-mail: bsbipubs @beeb.net to whom orders for all issues prior to Volume 22 part 1 should be sent. Recent issues (Vol 22 part 1 onwards) are available from Mr M. Walpole, B.S.B.L., 68 Outwoods Road, Loughborough, Leicestershire, LE11 3LY. Watsonia August 2000 Volume twenty three Part two Contents BRIGGS, M. Presidential Address, 1999: The study of a local flora Box, J. D. Mistletoe Viscum album L. (Loranthaceae) on Oaks in Britain FOLEY, M. J. Y. A morphological comparison between some British Orobanche species (Orobanchaceae) and their closely-related non-British a from continental Europe: Orobanche reticulata Wallr. s. 1. JONSELL, B., NORDAL, I. & ROBERTS, F. J. Viola rupestris and its hybrids in Beian FOLEY, M. J. Y. & PORTER, M. S. Carex muricata L. (Cyperaceae) - a review of its present status in Britain OREDSSON, A. Choice of surveyor is vital to the Pane of flosighe. dines studies RICH, T. C. G. & SYDES, fo Rtecnniline andl te iechaes ei the Nationally Seatiee plants Ajuga pyramidalis L. and Melampyrum sylvaticum L. : : FOLEY, M. J. Y. Dactylorhiza incarnata (L.) So6 subsp. ochroleuca (Wiistnei ¢ ex Boll) P. F. Hunt and Summerh. laces A comparison of British and European plants a RIcH, T. C. G. Conservation oF nena S Ddiodiversity Hienatnan dase (Asteraceae), Welsh Hawkweed RICH, T. C. G. & HOUSTON, L. Conservation of Brita S biodiversity Miocene tavense (Asteraceae), Black Mountain Hawkweed RANDALL, R. D. & RICH, T. C. G. Conservation of Britain’s biodiversity Res trelleckensis (Rosaceae), Trelleck Bramble : : USHER, M. B. The nativeness and non-nativeness of species : MEIKLE, R. D. & ROBINSON, N. A. A new record for Salix; xX angusensis (Salicaceae) Rech. f. from Ainsdale Sand Dunes National Nature Reserve, S. Lancs. Vic. 39° ..: HANDLEY, R. J. & DAvy, A. i Discovery of faale plants of Naja marina ra (Hydrocharitaceae) in Britain ue son | oa ee ee NOTES Blunt, A. G. & Blunt, M. E. Corynephorus canescens (L.) P. Beauv. (Poaceae) at Kinver Edge, Staffordshire: a re-assessment Rich, T. C. G. A reanalysis of the Mistletoe (Viscum ibis | ‘Lemnihaeeas survey data from the 1970s and 1990s _... —... Rich, T. C. G., Jones, R. A. & Jebb, M. Three new Batish Sites fon Cares depauperata With. (Cyperaceae) represented in the Irish National Herbarium, Glasnevin oe Dines, T. D. & Preston, C. D Fleoeiaris maniite ficeorercan in ‘Scotland Rich, T. C. G. & Brown, N. Suaeda vera J. F. Gmel. (Chenopodiaceae), Shrubby Sea-blite, present in Anglesey (v.c, 52), Wales Piggott, C. D. A probably native and regenerating population of Tilia platyphyllos Scop. In Bedfordshire - Allen, D. E. Further British species of ueue L. (Reston in comhaed Fence subsp. muricata Allen, D. E. Rubus angusticuspis Sudre (Rosaceae) in Scilly BOOK REVIEWS oe ae ee ae OBITUARIES REPORT Published by the Botanical Society of the British Isles ISSN 0043-1532 Typeset by D. K. & M. N. SANFORD Printed in Great Britain by PALLADIAN PRESS, UNIT E, CHANDLERS ROW, PORT LANE, COLCHESTER, ESSEX CO1 2HG 231-235 237-256 257-267 269-278 279-286 287-291 293—297 299-303 305-310 311-316 317-322 323-326 327-330 331-334 335-338 338-339 340-341 341-342 343-344 344-345 346-347 347-348 349-358 359-372 373-375 i uae “WATSONT A Botanical Society of the British Isles Volume 23 Part 3 February 2001 Editors: M. Briggs, D. L. Kelly, D. R. McKean, D. A. Pearman, M.N. Sanford, D. A. simpson Botanical Society of the British Isles Patron: Her Majesty Queen Elizabeth the Queen Mother 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 2000—2001 President, Dr G. Halliday Vice-Presidents, Mr M. Walpole, Mr D. J. McCosh, Prof. C. A. Stace Acting Honorary General Secretary, Miss A. Burns Honorary Treasurer, Mr M. E. Braithwaite Editors of Watsonia Papers and Notes, D. L. Kelly, D. R. McKean, M. N. Sanford*, D. A. Simpson Book Reviews, D. A. Pearman Obituaries, M. Briggs *Receiving editor, to whom all MSS should be sent (see inside back cover). © 2001 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 23: 377—390 (2001) 6 iy On the status of the genus Koeleria Pers. (Poaceae) in Britain J. M. DIXON Dept. of Environmental Science, University of Bradford, Bradffrd BD7 1DP MAY 15 2001 LIBRARIES Since 1980 there has been confusion over the number of species in the genus Koeleria found in Britain. Prior to 1980, only two species - Koeleria macrantha (= cristata) and K. vallesiana - were reported; Flora Europaea added K. glauca to these. Results of an extensive programme to clarify the status of the Koeleria macrantha group (containing both K. macrantha and K. glauca) in Britain are presented here. Detailed morphological measurements and observations were made on plant material collected from 51 sites throughout Britain. The material was then grown under standard conditions; repeat measurements and observations were then made. Chromosome determinations were also carried out on samples of all populations. Whilst field specimens sometimes had morphological characteristics which might have identified them as either K. macrantha or K. glauca, all plants grown under standard conditions in the glasshouse converged in all their characteristics and were clearly referable to K. macrantha. The results are supported by cytological evidence: all plants were tetraploid (K. macrantha is reported as 2n = 28 and K. glauca as 2n = 14 in Britain). The results strongly suggest that only two species of Koeleria - K. macrantha and K. vallesiana - occur in Britain, and that confusion might have arisen as a result of the wide morphological variation inherent in K. macrantha. ABSTRACT KEYWORDS: Koeleria macrantha, Koeleria glauca, morphological measurement, chromosome determination. INTRODUCTION During the latter part of the 19th century and early part of the 20th, the genus Koeleria in Britain was divided into at least five species: K. vallesiana (Honck.) Gaudin, K. albescens DC., K. gracilis (eugracilis) Pers., K. britannica Domin and K. pseudocristata Domin. There were nine varieties and several forms distributed between them, together with the hybrid K. x mixta, considered to be the hybrid between K. vallesiana and K. britannica (Howarth 1933). Hubbard (1954) reduced the species to two (K. vallesiana and K. cristata), with a comment that a coastal sand dune variant had been named K. albescens. However, he noted that the British plant referred to as K. albescens was scarcely distinguishable from inland K. cristata, except perhaps by its whitish panicles. A further note stated that K. britannica is a dwarf variant with short dense panicles. The delimitation of two species continued to be adopted in British Floras throughout the 1960s and 1970s (Clapham et al. 1962, 1987; Keble-Martin 1965; Hubbard 1984). However, Humphries (1980) described the genus Koeleria in Flora Europaea in which species 5—13 of the genus are listed under the K. macrantha group. The author comments that the group is an imperfectly understood complex, with much variation as a result of polyploidy and adaptation to a wide range of ecological conditions. He goes on to state that many populations which have been given specific, subspecific and varietal rank might well represent only ecotypes and variants of different ploidy levels. As a result, the species in the K. macrantha group, which are based entirely on morphology, might only represent minor taxa (Tutin ef al. 1980). Within the K. macrantha group the species K. macrantha (Ledeb.) Schultes (a re-naming of the older K. cristata) and K. glauca (Schrad.) DC. are both recorded as occurring in Britain, along with K. vallesiana which in Britain is confined to a handful of sites in Somerset, and will not be discussed further here. British Floras and related publications from 1980 onwards have, for the most part, contained the three species: K. vallesiana, K. macrantha and K. glauca, the last-named also belonging to the macrantha group. Clapham et al. (1987) recorded the distribution of K. macrantha as “ in turf on 378 J. M. DIXON well-drained, base-rich soils particularly on chalk and limestone. Locally common” and K. glauca as “on sandy soils. Widely distributed throughout the British Isles”. Rose (1989) recorded K. macrantha as “frequent to locally common on dry sandy or calcareous grassland, dunes” and K. glauca as “common. On sandy soils, especially dunes”. Fitter & Fitter (1984) recorded the two species as occurring in Britain but commented that the distribution of K. macrantha is not yet distinguishable from that of K. glauca. Some authors were rather more cautious: Grime ef al. (1988) merely noted that “recently K. glauca has been separated from the more common taxon formerly called K. cristata’, Rodwell (1992) only referred to K. macrantha, while Stace (1997) under the sub-heading for macrantha noted that “despite several claims the diploid K. glauca has not been confirmed from BI’. This paper offers a detailed examination of the status of the Koeleria macrantha group in Britain. MATERIALS AND METHODS FIELD WORK Observations and measurements were made on material belonging to the macrantha group from 51 sites throughout Britain, using the B.S.B.I. distribution map as a guide (Perring & Walters 1962). The majority were coastal sites but a number of inland sites were also sampled. Site descriptions are given in Appendix | and Fig. | indicates their approximate geographical location; at three of the sites (numbers 5, 41 and 44 ) two or more different habitats were sampled. At each site estimates were made of the average height of stands of vegetation containing K. macrantha (apart from sites 1, 2 and 3), and an assessment was made of the status and habit of the species. Accompanying species were also recorded but are not presented here. An estimate of soil depth was made, and a soil sample collected from the top 10 cm for analysis. The soil was collected with a Stainless steel trowel, after removing any surface vegetation and litter, and placed in self-sealing polythene bags. Measurements of pH were made on fresh soil and the rest of the analyses, comprising loss on ignition, extractable Ca, K, Mg and PO, - P and total organic nitrogen, were carried out on air-dried soil: methods were according to Allen (1989). Soil analyses are presented in Appendix 2. Wherever possible about ten ramets of K. macrantha were collected from each site, far enough apart to be separate genotypes, and an additional number of inflorescences were gathered from the same population to ensure sufficient measurements were available. The ramets were wrapped in damp newspaper and placed in polythene bags, as collection took place over several days. LABORATORY AND GLASSHOUSE WORK The length and width of the three longest leaves of eight selected ramets were measured and the length of the three longest stems; their panicle lengths and widths (measured in the middle of the panicle) were also measured. Where a ramet did not have three inflorescences, measurements were included for that population from the extra ones collected. Thus for most populations a total of 24 measurements for each variable were obtained. Three populations could not be sampled extensively and these are indicated by an asterisk in Appendix 1. In addition, for each population six inflorescences were selected at random (from different ramets) and one or two spikelets were removed from the middle of the panicle and measured to the nearest 0-5 mm. Observations were also made on whether stem thickening was present, the shape of the glumes and lemmas, and the degree of hairiness present on both leaves and floral parts. Hairiness was subjectively assessed as glabrous, sparsely hairy, moderately hairy and densely hairy. After the measurements had been made all the ramets were potted up in 9 x 9 cm plastic pots containing John Innes No | compost and transferred to an unheated glasshouse. Each pot was moved one position horizontally and laterally every three weeks to ensure any variations in environmental conditions were evened out. The plants were watered as necessary, and in the spring were given four feeds of seedling strength commercial fertilizer “Tomerite’ to encourage flowering. . Sampling had taken place during the summers of 1997 (northern Britain) and 1998 (southern Britain). Each set of plants was grown in the glasshouse for approximately one year under standard conditions and then harvested. KOELERIA IN BRITAIN 379 | | | | | : | H | : | | | | | | i | | : | ! O Km _ 100 Miles 100 | | = | | | | | | | | 30 ! at | ~ ! | | 35 | | | | | | | | | | O | | | | 37 | | | | 38 oh 44 | ! 1 : , | | | o43 | ie Sire | | 47 g- a | _— 46 TA | | | | | j 49 | | | | } | - | | | | | | i he | | | | FIGURE |. Approximate locations of sites sampled for Koeleria. Measurements were then made, where possible, on each population for the same set of variables as had been made for the field samples. However, glasshouse populations from sites 15, 20 and 22 did not flower. In addition, chromosome determinations were carried out for each population. During the spring of 1998 and of 1999, when new roots were being produced, three pots were selected at random from each population and about six new, healthy white root-tips were snipped off from those emerging through the sides of the soil/root mass. These were fixed in 3:1 absolute ethanol / glacial acetic acid fixative and kept in the refrigerator for periods not exceeding one week until they could be processed. The root tips were washed in tap water and hydrolysed for 12 minutes at 65° C in MHCI. This extended hydrolysis was found to be necessary to ensure efficient squashing. The root tips were stained in Feulgen’s solution for 1—2 hours and then squashed in 45% acetic acid. 380 J. M. DIXON From the literature K. macrantha in Britain is described as tetraploid (2n = 28) and K. glauca as being diploid (2n = 14) (Clapham et al. 1987). Therefore, for the current study it was only necessary to ascertain whether 14 or 28 chromosomes were present. In view of the quantity of material being processed, to save time it was decided that any squashes with countable spreads of 20+ could safely be deemed to belong to the tetraploid and this method was adopted. For each population two out of three plants sampled were considered acceptable if good counts were obtained (not all plants had actively dividing roots); where necessary further plants were sampled. RESULTS MEASUREMENTS These are presented in Appendix 3. Leaf length is highly plastic and mean leaf length was longer in all the glasshouse plants with the exception of populations from site 39 (Broughton Bay), which had almost the same mean lengths for field and glasshouse plants, and from site 4 (Nairn) where the field population had a significantly longer (p = 0-05) mean leaf length (Nairn had the longest mean leaf length of any field population). The increase was not significant in only five populations - sites 5a (Branderburg), 6 (Portgordon), 21 (Allonby), 30 (Staxton) and 44a (Sandwich Bay). These were all populations with mean field leaf lengths >11 cm. All the shortest populations, that is with mean field leaf lengths < 6 cm, showed the greatest increase in leaf length in the glasshouse populations and were all significantly (p = 0-001) longer, increasing to at least twice the field length. Leaf width is not very variable in K. macrantha, only ranging from 1—4 mm. However, all populations increased their mean leaf width in the glasshouse, many of them significantly so. The populations which had not increased significantly in leaf length also tended to be those not showing significant increases in leaf width either. Stem length was more variable but showed an increase in all glasshouse populations apart from site 21 (Allonby) and site 24 (Coxhoe), with mean field lengths of 27-3 and 29-7 cm respectively. These two populations had the longest mean field stem lengths. Panicle length increased in most populations in the glasshouse; only one population had a mean measurement less than that of the field population site 13 (Mull of Galloway), but this was not significant. Some populations increased significantly in panicle length in the glasshouse while others did not. Panicle width also generally increased, but it is difficult to measure this with any degree of . accuracy and it depends on the age of the panicle, which becomes more contracted as it matures. Spikelet length, which in the field ranged from 3-5-6 mm, increased in all populations in the glasshouse apart from site 46 (Camber) which remained the same. However, unlike the leaf length measurements, there was no significant variation between the ratio of increase for any of the populations. Whatever the mean field length the mean increase in spikelet length was around 1:1-2, with the glasshouse spikelet lengths ranging from 4-5—7-5 mm. OBSERVATIONS Field populations varied considerably in their habit, with some very rigid, tufted populations such as from sites 2 (Barra), 12 (Ayr), 17 (Auchenmalg Bay), 23 (Silecroft), 27 (Jenny Brown’s Point); 43 (Cheddar) and 44 (Sandwich Bay - foreshore). Some were very dwarf, prostrate populations such as from sites 36 (Sizewell), 37 (Thorpeness Sluice) and 42 (Brean Down), some prostrate “trailing” populations such as from sites 39 (Broughton Bay) and 40 (Rhossili Cliff) and some tall, bushy populations as at sites 24 (Coxhoe) and 46 (Camber). These habit forms became much less pronounced in the glasshouse, but populations from sites 2, 12, 17 and 23 remained tufted and fairly rigid. Although the field populations varied in leaf colour from light blue-green to glaucous to dark green on the adaxial surface, and from mid-green to dark green on the abaxial surface, with coastal populations being generally more glaucous than inland populations, after a year in the glasshouse there was much closer correspondence in leaf colour, varying from blue-green to mid-green on the adaxial surface and a uniform mid-green on the abaxial surface. | KOELERIA IN BRITAIN 381 Leaf hairiness is extremely variable, with plants from the same population being glabrous to densely hairy and the degree of hairiness even varying on the same plant. Hairs can be minute and downy, to long (1-5 mm) on both surfaces and on the margins. Generally leaves are less hairy towards the tip, but in some populations, for example from site 24 (Coxhoe), the leaves were densely hairy on both sides almost to the tip, while in others, for example from site 27 (Jenny Brown’s Point) the leaves were almost entirely glabrous apart from a few hairs around the ligule, with not even any marginal teeth. In most populations the abaxial surface is more hairy than the adaxial. In general, leaves tend to be more hairy in southern populations than in northern, but extremes have been found in both the north and the south of Britain. None of the c. 1200 stems examined had any sign of thickening; some stems were very robust but there were no bulbous swellings at the base. All glasshouse inflorescences were greenish-purple until flowering, becoming silvery and losing their greenness as the panicles closed, and finally straw-coloured as the panicles matured. Panicles were variable in shape, and narrowly oblong, spiciform, pyramidal and some almost round panicles were observed; these variations were maintained in cultivation. The majority of plants had a moderately to densely hairy rachis, sometimes very fine and downy, sometimes with longer hairs to 0-5 mm. In only five populations were plants found which had a glabrous to sparsely hairy rachis - 13 (Ayr), 11 (Troon), 33 (Great Orme), 35 (Holme-next-the-sea) and 44 (Sandwich Bay - fixed dunes). Glumes varied from aristate to acuminate (most common) to mucronate; none were blunt when flattened. Glumes were densely hairy with some hairs to 0-5 mm, to minutely hairy on the keel and near the tip, to completely glabrous. Awned and very sharply-pointed glumes occurred more frequently in northern populations, with mucronate glumes more frequent in southern populations. Lemmas were also either aristate, acuminate (most common) or, rarely, mucronate. Lemmas also varied in hairiness with some being glabrous and some minutely hairy on the keel and near the tip: lemmas are never as densely hairy as glumes. CHROMOSOME COUNTS All populations were tetraploid with 2n = 28. No diploid plants were found. TABLE 1. DIFFERENTIAL CHARACTERISTICS FOR KOELERIA MACRANTHA AND K. GLAUCA Characteristic Koeleria macrantha Koeleria glauca 1) Habit Tufted Rigid tufted 2) Stems Not thickened, 10-40 cm Robust and with bulbous thickening towards the base, 10-45(—60) cm 3) Leaves Green or glaucous, flat or folded, Leaves of non-flowering shoots glaucous, more sometimes slightly rough, up to 15cm x __ or less inrolled and rough to touch, at least above, 1-2 mm densely silvery, not more than 5 cm 4) Leaf sheaths Persistent, with parallel veins Persistent, breaking down into filaments when old 5) Panicle 20—80(—100) mm x c. 10 mm. Narrowly 20-110 mm x c. 8 mm. Spiciform, interrupted oblong, silvery-green purplish below; whitish/silvery; rachis densely puberulent 6) Spikelets 2—5(-6) mm 4-5 mm 7) Glumes Acuminate,aristate or mucronate green, = Blunt when flat, pale green or brown brown or purplish with a silver hyaline margin 8) Lemma Acuminate and glabrous Subobtuse or mucronate, puberulent, at least towards the base 9) Chromosomes 2n = 28 (for Britain) 2n = 14 (for Britain) Sources: Humphries (1980); Clapham et al. (1987). 382 J. M. DIXON DISCUSSION Taxonomic differences between K. macrantha and K. glauca are listed in Table 1. If each character is examined in turn a comparison can be made between the samples collected in the field and the glasshouse material grown under standard conditions: 1) A number of the field populations of K. macrantha had a tufted habit but with flexible leaves. while several could be described as having a rigid tufted habit, which in four populations remained so in the glasshouse. 2) Examination of the stem did not reveal any thickening or bulbous swelling in these four populations nor indeed in any other population; not a single plant exhibited this characteristic. Stem length is similar for the two species, but with K. glauca extending less commonly to 60 cm. Most of the K. macrantha populations, both in the field and in the glasshouse, did not generally extend beyond 40 cm, but in three field populations and 19 glasshouse populations, stems 44-57 cm long were recorded. 3) None of the plants collected from the field had densely silvery leaves; a few were rough enough to be described as scabrid, but were not so in their glasshouse counterparts. Sixteen populations had mean leaf lengths equal to or less than 5 cm, but in the glasshouse, in all these populations, leaf length increased significantly, with minimal mean lengths of not less than 10 cm. Leaves of Koeleria plants fold or inroll (K. macrantha does both) when subject to water stress - a common occurrence on sand dunes and shallow soils over limestone, but are flat when not subject to a water deficit: all glasshouse plants, watered freely, had flat leaves. 4) Observations on leaf sheath breakdown did not reveal the presence of filaments in either field or glasshouse populations. 5) Panicles varied in size and shape as described in the results: none exceeded 9 cm either in the field or in the glasshouse, and the rachis in the majority of populations varied from moderately to densely hairy, only five populations had a sparsely hairy or glabrous rachis. These included three small populations, an intermediate one and a tall population (based on leaf length). 6) Spikelet length for the two species is an overlapping range, with K. macrantha extending to 6 mm and K. glauca stated to be not less than 4 mm. In the field only two populations had spikelets with measurements less than 4 mm, but these had increased to 5 mm in the glasshouse. Nine populations contained spikelets of 6 mm or over in the field and in the glasshouse this had increased to 41 populations containing spikelets measuring 6 mm or longer. 7) No plants had glumes that were blunt when flattened. Some had a well-developed hyaline margin and others much less so. Colouring was variable, depending on the state of maturity. 8) Lemmas varied in shape and also in hairiness as described in the results, and at maturity the thickened median rib was found to extend into the folded acute apex. 9) All plants were tetraploid. Koeleria glauca is described as a plant of sand dunes and sandy soils, habitats which are susceptible to drought and are often nutrient poor. Willis & Yemm (1961) investigated the nutrient status of three types of dune soils in Devon, and reported that while these had sufficient minor nutrients they were very deficient in nitrogen, phosphorus and often potassium. The vegetation of these dune soils produced some forms of several of the common species which were very depauperate, and which had even been recognised as dwarf varieties. Addition of the major nutrients led to an increase in height’and weight, often two to three times greater than the control plants. Ivimey-Cook & Proctor (1966) also commented that the sparse cover and dwarfing of plants on calcareous grey dunes in the Burren, Co. Clare (v.c. H9) may reflect nitrogen, phosphorus and potassium deficiencies in the skeletal soil. In the present study, although no sites were deficient in calcium and magnesium (see Appendix 2), several sites, 9 (Balmedie), 11 (Troon), 12 (Ayr), 19 (Druridge Bay), 23 (Silecroft) and 36 (Sizewell), all from sand dunes or raised beaches are very low in nitrogen and all the plants are of small stature. Site 36 is also deficient in phosphorus and potassium. Other short populations occurred in very shallow soils, some were heavily grazed and some occurred on exposed cliffs, usually in shallow soils. Plants in such stressed habitats are often dwarfed and very tufted relative KOELERIA IN BRITAIN 383 to plants from more favourable habitats. If the stature of the short populations was a genetic adaptation to such stressed habitats (as it is perhaps in K. glauca), then one would have expected the ratio of increase in leaf length brought about by the more favourable conditions of the glasshouse to be roughly the same regardless of the initial length of the leaf. Figure 2 shows that this is not so and that the shorter the field population in leaf length the greater the ratio of increase in the glasshouse, leading to the conclusion that the plants were suffering from environmental stress in their original habitat, as were those investigated by Willis & Yemm (1961). This is borne out by the fact that spikelet length, and thus the floral parts, which are generally much less plastic than the leaf and stem characteristics, did not increase in the linear manner demonstrated by the regression line in Fig. 2. The results show that K. macrantha is very variable in many of its morphological characteristics, and some plants collected in the field had a number of features which might lead them to being identified as K. glauca, such as leaf length, spikelet length, inrolled leaves and indumentum. However, after a year of growth under standard conditions all plants had converged in these (and other) features and were unmistakably K. macrantha, an identification supported by the cytological evidence of the constant tetraploid state. Wall (1988) investigated the role of B chromosomes in K. macrantha and sampled extensively from 27 sites (mainly coastal) throughout Britain, including two sites sampled by the present author. All plants that she examined were tetraploid and clearly K. macrantha. Thus sampling in depth from 78 sites covering a wide range of geographical locations and substrate types in Britain has yielded only tetraploid plants. Wall also commented that plants with high numbers of B chromosomes (K. macrantha can have up to 6) lack vigour and set fewer seeds. Thus it is possible that the shortest populations might be those with high numbers of these accessory chromosomes; for example, in the present study populations from sites 23 (Silecroft), 36 (Sizewell) and 41b (Porthcawl) all produced mean leaf lengths of less than 11 cm in their glasshouse populations compared with the mean overall glasshouse leaf length, for all populations, of 16-2 (+ 0-5) cm. Ratio field to glasshouse leaf length Uo 0 0.2 0.4 0.6 0.8 ] 122 1.4 Log) field leaf length (cm) FIGURE 2. Log. mean field leaf length vs. ratio of field to glasshouse leaf length for sampled Koeleria populations 384 J. M. DIXON wih (\ FIGURE 3. Comparison of spikelets and spikelet parts of Koeleria macrantha and Koeleria glauca. Top row K. glauca ; bottom row K. macrantha. G 1 = lower glume; G2 = upper glume; P = palea and pedicel; L = lemma. Entire spikelets on left. Indumentum not shown. Because many of the coastal sites are stressed habitats it is possible that some plants exhibiting some characeristics of K. glauca may well have been misidentified from field specimens, as a result of the inherent variation in K. macrantha, which is described in Flora Europaea (Tutin et al. 1980) as “ very variable in habit, leaf-size and indumentum”. Confusion may have arisen by the inrolling of leaves by K. macrantha where water deficits occur. To ascertain the degree of confusion which might arise in comparing K. macrantha and K. glauca an examination was made of European material of K. glauca held in K. Measurements and observations are summarised below. Stems of K. glauca were mostly very sturdy but in only two specimens was there any slight thickening which could be described as bulbous. Leaves of non-flowering shoots, which are presented as an identifying characteristic in Flora Europaea as being not more than 5 cm, surprisingly turned out to be in many cases considerably longer than this: mean leaf length was 6-1 (+ 0-2) cm, and the range for western Europe was 2-6—10-0 cm. Only about 37% of leaves had a length less than 5-5 cm. Thus leaf length does not appear to be a strong differentiating feature for the species. However, the maximum leaf length of 10 cm is considerably shorter than that recorded by the author for British K. macrantha (24-8 cm). The majority of leaves were inrolled, but some were folded and some flattened. However, this might have been an artefact of the preservation process. Leaves of K. glauca were all “felted” with minute rough teeth, giving the silvery, scabrid characteristic described for the species (see Table 1). This felting extends to the tip of the leaf and is in contrast to K. macrantha where the leaves have longer, true hairs rather than teeth (which are confined to the margins) and in most populations these do not extend to the tip. The leaf sheaths of K. glauca break down into a network of filaments, in a similar manner to those of K. vallesiana, in contrast to those of K. macrantha, where the leaf sheaths are persistent. = Panicle lengths to 115 cm were measured and there is less variation in the shape of the panicles in K. glauca compared with K. macrantha: 90% were spiciform with the rest exhibiting a slight pyramidal shape. Koeleria glauca is characterised as having the panicles interrupted below and KOELERIA IN BRITAIN 385 77% of the specimens examined were interrupted compared with only 14% in specimens of K. macrantha (from a much larger sample) with interrupted panicles. Spikelets of K. glauca ranged from 2-5—5-5 mm with a mean of 4-7 mm (median 4-0 mm) which was very little different from those of K. macrantha for British specimens (range 3-5—6-0 mm, mean 4-9 mm and median 4-8 mm). These figures do not agree with those quoted in Flora Europaea (see Table 1). Illustrations of the spikelets and the spikelet parts are provided in Fig. 3, and it can be seen that the glumes of the two species are different, with those of K. macrantha being pointed and K. glauca being blunt. The lemmas of K. macrantha are also more pointed than those of K. glauca and the median rib is thicker and more rigid than in K. glauca, and extends into the point. Some of the differential characteristics listed in Table 1, such as the rigid tufted habit of K. glauca and panicle width, were not possible to determine from herbarium specimens. However, these aspects notwithstanding, and in spite of the greater convergence of the two species than is indicated in their descriptions in Flora Europaea, there is a marked separation of the two species as far as the leaf sheath breakdown, leaf pubescence and floral characteristics are concerned, together with the more slender and uniform spiciform shape of the panicle, which even in herbarium specimens is more intensely silvery than those of K. macrantha. Thus the presented measurements and observations provided in this paper do not support the presence of K. glauca in Britain. In the most recent Floras, British authorities have more-or-less discarded K. glauca from the British list (see Cope 1998); the current paper affirms this view. ACKNOWLEDGMENTS My thanks are due to Dr A. D. Headley for providing material from Barra, Rhum and Skye, to Dr D. J. Hambler for drawing Fig. 3, and to Dr D. Comber, Dr W. H. G. Hale and Prof. M. R. D. Seaward for help and advice. REFERENCES ALLEN, S., ed. (1989). Chemical analysis of ecological materials, 2nd ed. Blackwell Scientific Publications, Oxford. CLAPHAM, A. R., TUTIN, T. G. & Moore, D. M. (1987). Flora of the British Isles, 3rd ed. Cambridge University Press, London. CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. (1962). Flora of the British Isles, 2nd ed. Cambridge University Press, London. Cope, T. A. (1998). Koeleria, in RICH, T. C. G. & JERMY, A. C., eds. Plant crib 1998: B.S.B.I., London. FITTER, A. & FITTER, R. (1984). Collins Guide to the grasses, sedges, rushes and ferns of Britain and northern Europe. William Collins, Glasgow. GRIME, J. P., HODGSON, J. G. & HUNT, R. (1988). Comparative plant ecology. Unwin Hyman, London. HOWARTH, W. O. (1933). Notes on Koeleria Pers. Reports of the Botanical Exchange Club 10: 37-41. HUBBARD, C. E. (1954). Grasses. Penguin Books, Harmondsworth. HUBBARD, C. E. (1984). Grasses. 3rd ed. Penguin Books, Harmondsworth. HUMPHRIES, C. J. (1980). Koeleria Pers., in TUTIN, T. G. et al., eds. Flora Europaea 5: 218-220. Cambridge University Press, Cambridge. IVIMEY-COoK, R. B. & PRocTorR, M. C. F. (1966). The plant communities of the Burren, Co. Clare. Proceedings of the Royal Irish Academy 64B: 211-301. KEBLE-MARTIN, W. (1965). The concise British flora in colour. Ebury Press & Michael Joseph, London. PERRING, F. H. & WALTERS, S. M.., eds. (1962). Atlas of the British flora. Thomas Nelson, London. RODWELL, J. S., ed. (1992). British plant communities. Vol. 3. Grasslands and montane communities. Cambridge University Press, Cambridge. ROSE, F. (1989). Grasses, sedges, rushes and ferns of the British Isles and north-western Europe. Viking, London. STACE, C. A. (1997). New Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. WALL, S. (1988). The role of the B chromosome in the genetic system of a tetraploid grass (Koeleria cristata L.). Ph.D. thesis, University of Manchester. WILLIS, A. J. & YEMM, E. W. (1961). Braunton Burrows: mineral nutrient status of the dune soils. Journal of ecology 49: 377-390. (Accepted July 2000) 386 J. M. DIXON APPENDIX 1. SITE DETAILS FOR KOELERIA SAMPLING. NUMBERS REFER TO APPROXIMATE LOCATION ON MAP HIG. 1 No Location NGR Alt. Slope Aspect 1* Nr. Dunvegan, Isle of Skye, Highland NG223549 3m flat : Koeleria frequent on basalt; also occasional on sandy beach with broken coral 2* Eoligarry, Barra, Western Isles NL695054 lm flat - Machair. Koeleria common to abundant in ungrazed areas 3. Fonchra, Rhum, Highland NG337003 3500s 2 Sa SE Base of steep, vegetated cliff over basalt. Koeleria frequent 4 Nairn, Highland NJ866569 2m flat - Pebbly track Koeleria dominant, also in longer vegetation at side of track 5 Nr Branderburgh, Grampian NJ204714 15m ZO? N Dune-heathland with mix of calcicoles and calcifuges, also on rocks at edge of sea 6 Portgordon, Grampian NJ401645 3m flat . Sandy soil over pebbles, Koeleria scattered but frequent 7 St. Combs, Grampian NK059630 15m 12° NE Raised beach,vegetation to 40 cm high, Koeleria in isolated clumps 8 Cruden Bay, Grampian NK096358 15m 40° WSW All plants scattered in pockets across face of granite cliff, Koeleria frequent 9 Balmedie, Fife NJ977182 15m 10° NNE Short grazed turf, Koeleria common, plus longer ungrazed turf, with scattered Koeleria 10 Lunan Bay, Tayside NO691502 10m BD NW Lightly grazed, fixed dunes, vegetation 3-6 cm. Koeleria scattered but frequent 11* Troon, Strathclyde NS324305 3m flat - Fixed dunes, Koeleria scattered but frequent 12 Ayr, Strathclyde NS328205 10m 20° W Edges of fixed dunes (top was mown), Koeleria very scattered 13. Dunure, Strathclyde NS254158 2m flat - On rocks (hornfels) at top of beach in shallow pockets of soil, isolated plants 14 Girvan, Strathclyde NX160943 10m 90° N In shallow pockets of soil on rocks (phyllite) at top of shore, also on flat tops of rocks, isolated plants, an abundance of ants here 15 Labarax, Dumfries & Galloway NX972601 5m flat - Koeleria scattered on fixed dunes, isolated plants also in rock crevices (slate) 16 Mull of Galloway, Dumfries & Galloway NX159307 50m ee E On grassy slope and also on flat cliff top. Koeleria frequent but scattered 17 Auchenmalg Bay, Dumfries & Galloway NX243515 6m flat - Very short, prostrate turf on track, vegetation 10-15 cm. Koeleria dominant here 18 Alnmouth, Northumberland NZ256115 3m flat - Fixed dunes, vegetation 30-45 cm high. Koeleria frequent but scattered 19 Druridge Bay, Northumberland NZ292936 10m flat - Fixed dunes, vegetation 10-15 cm high, grazed (horses). Koeleria frequent 20 Silloth, Cumbria NY090493 10m flat - Fixed dunes, vegetation to 45 cm high. Koeleria frequent 21 Allonby, Cumbria NY080440 7m flat - Partially fixed dunes,. Koeleria on top but not sides of dunes, vegetation to 45 cm 22 St. Bees, Cumbria NY957118 50m 5° E Short, but ungrazed, turf on cliff top. Koeleria frequent to abundant 23 Silecroft, Cumbria SD123805 10m flat - Top of bank of consolidated pebbles, short, ungrazed turf. Koeleria common, also on track, trampled and very short 24 Coxhoe, Durham NZ326364 132m _- flat - Dolomite quarry spoil heap. Koeleria in scattered small patches : 25. Far Arnside, Cumbria SD453760 12m 20° SW Grassy slope over Carboniferous limestone, just above beach. Koeleria frequent in small patches 26 Thrang End, Yealand Redmanyne, Cumbria SD494766 70m 10° NNW Carboniferous limestone pasture, sheep grazed. Koeleria frequent * = only small samples available for measurements and observations 44 45 46 47 48 49 50 51 KOELERIA IN BRITAIN 387 Location NGR Alt. Slope Aspect Jenny Brown's Point, Silverdale, Lancashire SD463734 5m 20" S Sandy bank over Carboniferous limestone, sloping into sea. Koeleria scattered among rocks Bishopdale, North Yorkshire SD943803 AL9m== 20° S Grassy bank over Carboniferous limestone, ungrazed vegetation to 45 cm high. Koeleria frequent Langcliffe, North Yorkshire SD833656 366m fiat . Koeleria isolated but frequent in pockets of soil on clints of limestone pavement, and in shallow grikes Staxton, North Yorkshire TAO11782 LOZ mse 5° E Koeleria scattered among tall vegetation on overgrown path of disused chalk pit Wormhill, Derbyshire SK107756 372m: 18: S Isolated plants in pockets on Carboniferous limestone outcrop. Turf 3-12 cm high Monyash, Derbyshire SK160661 274m fiat - On top of Carboniferous limestone outcrop. Koeleria scattered in small clumps. Some also present on more or less vertical rock face. Great Ormes Head, Gwynedd SH780836 76m ise N On Carboniferous limestone sea cliff, above tidal limits but within splash zone. Koeleria abundant, some on vertical face nearer the sea Little Ormes Head, Gwynedd SH8 13823 130m,” “45° W Very shallow soil over Carboniferous limestone. Koeleria abundant Holme-next-the-Sea, Norfolk TF708448 2m flat Partially fixed dunes, long loose vegetation to 60 cm. Koeleria occasional and very scattered, musty at edge of board-walk Sizewell, Suffolk TM475628 3m flat : Raised beach, just above sea level. Sparse, very short turf on almost pure sand. Koeleria scattered but frequent Thorpeness Sluice, Suffolk TM472593 240 flat - On inland side of a shingle bar, shingle mixed with sand. Koeleria abundant Aldeburgh, Suffolk TM465575 3m 6° NW Sandy bank at side of car park. Koeleria dominant Broughton Bay, West Glamorgan SN416928 30 m oy SE Early fixed dunes, dense tall vegetation to 60 cm. Koeleria restricted to path edges where vegetation was rather shorter Rhossili Cliff, West Glamorgan SN405878 50m 25° N Sloping cliff top, shallow soil over Carboniferous limestone. Koeleria scattered Nr. Porthcawl, Mid Glamorgan ST804782 6m 5° NE Fixed dunes, dense tall vegetation to 60 cm. Koleria very scattered, mostly on path edges. Also on limestone outcrop nearby Brean Down, Somerset ST286587 45m flat - Dwarf population on exposed Carnoniferous limestone cliff top; very shallow soil. Koeleria abundant Nr Cheddar, Somerset ST504517 245 m 15% S Koeleria frequent on Carbonifeous limestone outcrop, occasional in pasture Sandwich Bay, Kent TR364578 3m 10° WSW Roadside bank, also on flat fixed dunes and on flat sandy/pebbly foreshore. Koeleria common on bank and in dunes, more scattered on foreshore St. Margaret's At Cliffe, Kent TR372447 80 m jae ESE In short to moderately long vegetation (12-40 cm) on top of chalk cliff. Koeleria scattered in small populations Camber Golf Course, East Sussex TR946194 5m 25% ESE Partially fixed dunes of deep sand. Koeleria on edges of banks, frequent Nr. Beachy Head, East Sussex TQ592973 154m 90° WwW Koeleria virtually limited to roadside vertical bank, not in the pasture behind it Durlston Head, Dorset SZ025769 61m 22. S Very shallow (c. 5 cm) soil over limestone. Very short (2—7 cm) grazed turf. Koeleria frequent but scattered Carbis Bay Golf Course, Cornwall ST294588 95m 12° W Fixed dunes, Koeleria mainly on edges of banks and paths Mullion Cove, Cornwall SW666106 61m 29° N In short turf on top of cliff over serpentine. Koeleria scattered but frequent Kynance Cove, Cornwall SW686135 64 m 10° SW Koeleria abundant in very short turf over serpentine 388 J. M. DIXON APPENDIX 2. SOIL ANALYSES FOR SITES SAMPLED FOR KOELERIA Site No pH % Losson Camg 100g' Mg mg 100g1 Kmg100g-' PO,-Pmg_ Total N (%) Ignition 100g-' ] 5:8 56-1 205 88 39 1-12 1-92 Z 6-4 5-5 292 4] ANG 1-12 0-23 3 6-2 30-8 198 66 ZS 0-88 0-84 + 6-2 Res. 268 99 28 1-44 0-66 5a 5-9 6:8 205 52 14 0-56 0-02 5b 6-0 5-0 190 a 14 1-12 0-03 6 5:8 4-5 759 4] 16 0-90 0-20 yi 5-7 10-4 365 ik i) 0-92 0-39 8 6-1 17-1 348 119 34 0-92 0-56 9 6-0 =| 183 4] 16 0-90 0-10 10 5-9 4.] 198 40 ibs: 0-88 0-21 11 6-5 1:8 A Kos 38 14 12 0-06 2 6:0 1-7 190 4] 15 0-80 0-04 i 6-2 3-4 210 61 17 1-20 0-06 14 5-9 24-2 185 83 28 0-68 0-80 15 6-1 8-4 203 60 21 0-76 0-31 16 soil analysis not available 17 5:8 4-0 188 51 HIS 1-12 0-22 18 6-0 3-9 255 44 14 1-12 0-11 19 6:3 3-2 205 58 14 0-76 0-10 20 ao | 5-4 250 a2 Wy 0-60 0-21 21 6-0 7:8 360 56 18 0-70 0-24 Ze 6:5 8-4 150 60 24 0-88 0-27 23 6-4 5-4 210 iB 17 1-12 0-15 24 6:3 44 255 50 15 0-79 0-10 ZA 6:3 11-8 448 39 20 0-60 0-48 26 5-9 23-0 490 40 24 0-90 0-97 27 6:5 7:9 320 58 22 0-88 0-36 28 6:3 19-2 650 58 26 0-76 0-69 20 6-9 n/a 278 38 16 0-80 0-36 30 7-0 5-4 364 59 LF 0-68 0-47 a1 6:6 32°1 625 13 20 0-96 0-99 a2 6:6 11-9 330 14 2 2-80 0-57 33 6:2 17-2 550 56 25 0-56 0-33 34 6:3 7:6 383 39 19 0-68 0:27 35 5-5 1-7 69 13 5 0-48 1-00 36 5:8 0-3 56 16 5 0-12 0-10 3) 6-4 8-6 75 35 14 3-20 0-37 38 6-5 4-5 200 25 8 0-96 0-24 39 6-2 5-4 144 16 8 0-58 0-18 AO 6-4 16-4 460 55 40 0-46 0-75 Ala 6:6 6-6 356 25 11 0-48 0-25 Alb 6-3 32-9 _ 431 n/a 39 1-12 1-18 42 6-3 17-4 420 |W 2s) 39 1-00 0-77 43 6:0 17-7 363 20 13 0-48 0-61 44a 6-7 8-3 356 26 19 0-90 0-19 44b 6-8 4-0 119 24 8 iheesD 0-12 44c 6:8 ai 194 20 10 0-48 0-24 45 7-0 28-8 750 44 54 1-44 0-16 46 6-6 4.] Lol 23 9 0:90", 0-16 47 68 | 72 Zoi 2 11 0-94 0-26 48 7:0 16-7 500 100 54 1-00 0-59 49 7-0 7-4 a13 29 2) 0-46 0-29 50 6-1 31:3 150 388~ ». 50 1-58 1-05 51 6-0 35-2 106 200 46 0-31 0-88 5a = dune heathland; 5b = rocks; 41a = dunes; 41b = outcrop; 44a = roadside; 44b = foreshore: 44c = fixed dunes KOELERIA IN BRITAIN 389 APPENDIX 3. MEAN FIELD AND GLASSHOUSE MEASUREMENTS (+ SE) AND SIGNIFICANT DIFFERENCES FOR KOELERIA SAMPLES Site 1 Field Glass. 2 Field Glass. 3. Field Glass. 4 Field Glass. 5a Field Glass. 5b Field Glass. 6 Field Glass. 7 Field Glass. 8 Field Glass. 9 Field Glass 10 Field Glass. 11 Field Glass. 12 Field Glass. 13 Field Glass. 14 Field Glass. 15 Field Glass. 16 Field Glass. 17 Field Glass. 18 Field Glass. 19 Field Glass. 20. Field Glass. 21 Field Glass. 22 Field Glass. 23 Field Glass. 24 Field Glass. 25 Field Glass. 26 Field Glass. Leaf length (cm) 5-7 (0-5) 19-8 (0-6) 3 6-0 (0-4) 16-4 (0-9) 3 8-5 (2:3) 16-5 (0-8) 3 16-3 (1-7) 14-7 (0-5) 1 5-9 (0-3) 16-5 (0-7) 3 12-7 (0-7) 14-0 (0-5) N 13-6 (0-8) 15-3 (0-5) N 11-9 (0-7) 23-8 (0-4) 3 9-2 (1:2) 15-5 (0-7) 2 4-5 (0-4) 16-5 (0-5) 3 7-1 (0-5) 18-0 (0-6) 3 3-9 (0-2) 16-5 (0-6) 3 3-6 (0-4) 18-5 (0-5) 3 10-8 (0-6) 22°1.(1-7) 3 7:5 (0-6) 11-1 (0-3) 3 3-3 (0-3) 16-2 (0-6) 3 8-6 (0-9) 17-6 (0-4) 3 4-6 (0-2) 13-9 (0-9) 3 8-1 (0-5) 21-1 (0-9) 3 3-9 (0-2) 20-7 (0-7) 3 11-7 (0-9) 16-4 (0-2) 3 13-7 (0-6) 15-3 (0-5) 1 5-5 (0:3) 15-3 (0-4) 3 3-8 (0-2) 10-0 (0-4) 3 15-1 (0-9) 25-8 (0-9) 3 8-4 (0-6) 17-9 (0-4) 3 4-9 (0-3) 17-7 (0-7) 3 Leaf width (mm) 1-2 ©1) 2-9 (0-1) 3 1-5 (0-1) P7T(O1)3 1-4 (0-1) 2-8 (0-1) 3 2.2 (0-1) 2.7 (0-2) N m2) 2.4 (0-1) 1 1-8 (0-1) 22 (0.0) 2.0 (0-1) 2-5 (0-1) 1 1-9 (0-1) 2-6 (0-1) 3 2.3 (0-1) 2-9 (0-1) 3 1-6 (0-1) 2-4 (0-1) 3 1-8 (0-1) 2-5 (0-1) 3 1-6 (0-1) 21 (0:1)3 1-7 (0-1) 2301) 2-8 (0-1) 2.9 (0-2) N 1-9 (0-1) 2-4 (0:2) 2 1-8 (0-1) D301)3 2.0 (0-1) 2-6 (0-1) 3 2.0 (0-1) 2-6 (0-1) 3 2.3 (0-1) 2.9 (0-1) 3 1-8 (0-1) 2-4 (0-1) 3 2.0 (0-1) DIOLS 2-1 (0-1) 2-5 (0-1) 2 Dna) 3-1 (0-1) 3 2.1 (0-1) 2.7 (0-2) 1 2.0 (0-1) 2-5 (0-1) 3 2.3 (0-2) 3-0 (0-1) 3 1-5 (0-1) 2-5 (0-1) 3 Stem length (cm) 9-7 (0-5) 24-3 (1-3) 3 18-2 (0-9) 33-5 (1-2) 3 22:6 (1-2) 32:1-°C1-3)3 PA eA Tp) 25-0 (1:0) N 10-9 (0-9) 20-8 (0-9) 3 17-4 (1-7) 19-1 (0-7) N 24-9 (1-7) 36-5 (1-5) 3 24-2 (1-2) 22-2 (0-7) N 18-6 (1-6) 30-7 (1-9) 3 17-5 (1-9) 28-0 (1-1) 3 9-6 (0-6) 31-0 (1-2) 3 3-9 (0-4) 25-4 (1-0) 3 11-4 (0-6) 24-9 (0-3) 3 17-0 (0-9) 23-4 (1-3) N 13-2 (1-3) 17-1 (1-0) 1 5-9 (0-7) n/a 14-9 (2-5) 22-9 (1-6) 3 12-3 (1-2) 29-3 (0-9) 3 14-8 (0-6) 26-4 (1-9) 3 11-4 (0-7) 24-7 (0-2) 3 24-3 (1-5) n/a 27-3 (0-9) 20-9 (0-9) 3 3-8 (0:3) 11-0 (0-8) 3 29-7 (2:9) 25-5 (0-8) N 15-5 (0-7) 33-9 (1-3) 3 21-4 (1:3) 37-6 (1-6) 3 (cm) 2:7 (0-2) 5-0 (0-3) 3 4.0 (0-1) 5-0 (0-2) 3 3-4 (0-2) 6-5 (0:2) 3 4.7 (0-3) 6-0 (0-2) 2 2-8 (0-2) 6-2 (0-1) 3 3-1 (0-2) 5-6 (0-2) 3 4-0 (0-2) 7-0 (0-2) 3 3-5 (0-2) 3-5 (0-1) N 3-1 (0-2) S=1(0-1) 3 3-2 (0-2) 5-6 (0-3) 3 3-1 (0-3) 5-7 (0-2) 3 1-7 (0-2) 4-7 (0-2) 3 3-8 (0-3) 6-9 (0-1) 3 5-9 (0-4) 6-0 (0-3) N 3-0 (0-2) 4-1 (0-1) 3 2S (0-1) n/a 3-6 (0-3) 3-3 (0-2) N 2-7 (0-2) 4-5 (0-1) 3 2-8 (0-2) 5-1 (0-2) 3 2-6 (0-1) 4-5 (0-2) 3 5-2 (0-2) n/a 3-9 (0-5) 4-6 (0-2) N 3-4 (0-2) n/a 2-3 (0-05) 4-7 (0-2) 3 5-4 (0-4) 7-1 (0-1) 3 3-6 (0-3) 7-0 (0-1) 3 3-5 (0-2) 5-6 (0-1) 3 Panicle length Panicle width Spikelet (mm) length (mm) 3-6 (0-3) 5-4 (0-1) 5-5 (0-4) 1 6-8 (0-2) 3 5-5 (0-6) 5-0 (0-3) 6-4 (0-2) N 6-8 (0-1) 3 4-3 (0-2) 5-2 (0-1) 6-3 (0-2) 3 6-3 (0-2) 3 4-0 (0-2) 5-1 (0-1) 4-5 (0-1) 1 5-8 (0-2) 3 3-7 (0-2) 4-8 (0-1) 5-5 (0-1) 3 5-2 (0-1) 1 3-3 (0-3) 4-8 (0-1) 5-0 (0-2) 2 5-9 (0-2) 3 3-9 (0-1) 4.7 (0-1) 5-2 (0-2) 3 6-9 (0-1) 3 4.9 (0-1) 5-2 (0-2) 5-2 (0-2) 3 6-0 (0-3) 1 4-5 (0-2) 5-0 (0-3) 5-8 (0-2) 3 5-6 (0:2) N 3-9 (0-2) 5-0 (0-1) 5-6 (0-3) 3 5-7 (0-1) -2 3-2 (0-2) 4-6 (0-2) 4-6 (0-2) 3 6-2 (0-2) 3 3-4 (0-2) 4-8 (0-1) 4.7 (0-2) 3 5-3 (0-1) 1 4-5 (0-2) 4.4 (0-2) 5-0 (0-3) N 5-6 (0-2) 3 7-1 (0-5) 5-5 (0-2) 4-6 (0-2) 3 5-8 (0-1) N 4-6 (0-2) 4-9(0-2) 5-1 (0-2) 2 5-5 (0-2) 1 4-1 (0-2) 4-3 (0-1) n/a n/a 4-9 (0-3) 4.7 (0-2) 5-0 (0-1) N 6-1 (0-2) 3 4-8 (0-2) 5-3 (0-2) 5-8 (0-2) 2 5-4 (0-2) N 4-5 (0-2) 4-6 (0-2) 5-7 (0-2) 2 5-7 (0-2) 2 4-4 (0-2) 4-8 (0-2) 5-0 (0-2) 1 5-3 (0-2) N 4-5 (0-2) 5-2 (0-2) n/a n/a 5-0 (0-3) 4.7 (0-1) 5-0 (0-2) N 5-2 (0-2 1 4-6 (0-2) 5-5 (0-2) n/a n/a 4.2 (0-2) 5-3 (0-1) 4-4 (0-2) N 5-4 (0-2) N 4-8 (0-3) 4-8 (0-2) 4-1 (0-3) N 5-7-(0:2) 2 4-0 (0-2) 4.9 (0-2) 5-7 (0-4) 3 5-3 (0:2) N n/a 4-7 (0-3) 4-5 (0-1) 5-8 (0-3) 2 n/a = not available 5a = dune heathland; 5b = rock. Numbers in bold indicate levels of significance for t-tests between the field and glasshouse populations. N = not significant, 1 = p at 0-05, 2 = p at 0-01 and 3 =p at 0-001 390 Site 27 Field Glass. 28 Field Glass. 29 Field Glass. 30. —~ Field Glass. 31 Field Glass. 32 Field Glass. 33 Field Glass. 34 Field Glass. 35. Field Glass. 36 —- Field Glass. 37 Field Glass. 38 Field Glass. 39 Field Glass. 40 Field Glass. 41a Field Glass. 41b Field Glass. 42 Field Glass. 43 Field Glass. 44a Field Glass. 44b Field Glass. 44c Field Glass. 45 Field Glass 46 Field Glass. 47 Field Glass. 48 Field Glass. 49 Field Glass. 50 Field Glass. 51 Field Glass. Leaf length (cm) 5-5 (0-7) 17-8 (0:6) 3 10-4 (1-1) 18-3 (0-4) 3 5-9 (0-3) 17-9 (0-9) 3 13-2 (0-7) 13-6 (1-1) N 6:3 (0-7) 13-7 (0-6) 3 6-6 (0-6) 14-2 (0-8) 3 4-4 (0-2) 18-4 (0-8) 3 4-4 (0-2) 19-6 (0-5) 3 9-6 (0-6) 22-9 (1-0) 3 4-6 (0-3) 10-0 (0-2) 3 5-0 (0-3) 13-3 (0-5) 3 7-0 (0-3) 13-5 (0-3) 3 12-1 (1-0) 12-0 (0-4) N 5-0 (0-3) 17-7 (0-8) 3 6-5 (0-3) 12-3 (0-3) 3 5-0 (0-3) 10-6 (0-4) 3 2-7 (0-2) 12-2 (0-3) 3 5-7 (0-3) 14-1 (0-7) 3 14-1 (0-6) 14-9 (0-3) N 13-6 (0-5) 16-7 (0:6) 3 11-0 (0-8) 15-4 (0-4) 3 9-4 (0-6) 18-1 (0-8) 3 9-2 (0-6) 19-0 (0-7) 3 8-8 (0-7) 17-5 (0-6) 3 4-6 (0-3) 13-7(0:5)3 8-4 (0-6) 16-2 (0-9) 3 9-1:(0-4) 13-8 (0-4) 3 6-9 (0-4) 11-9 (0-3) 3 Leaf width (mm) 1-8 (0-1) 2-5 (0-05) 3 F001) 2-8 (0-1) 3 1-4 (0-1) 2-4 (0-05) 3 1-5 (0-1) 2:3 (0-1) 3 1-8 (0-1) 2101)2 1-9 (0-1) 2-0 (0-1) N 1-7 (0-1) 2-6 (0-1) 3 1-7 (0-1) 2-4 (0-1) 3 1-9 (0-1) 23 O4)2 1-8 (0-1) 2:3 (0-1) 3 2-0 (0-1) 2-1 (0-1) N 2-1 (0-1) 2.3 (0-1) N 2-2 (0-1) 2-4 (0-1) 1 L001) 2-3 (0-1)3 1-8 (0-1) 2-4 (0-1) 3 1-7 (0-1) 2:3 (0-1) 3 1-5 (0-05) 2-0 (0-1) 3 1-5 (0-1) 2-1 (0-1) 3 2-3 (0-1) 2.5 (0-1) N 2-4 (0-1) 2-6 (0-1) 1 2-0 (0-1) 2-4 (0-1) 3 2-0 (0-1) 2-5 (0-1) 3 2-1 (0-4) 2-5 (0-1) N 1-8 (0-) 2-4 (0-1) 3 1-5 (0-1) 2-3 (0-1) 3 1-8 (0-1) 27 O13 2101) 2.3 (0-2) N 2-1 (0-1) 2-6 (0-1) 3 J. M. DIXON Stem length (cm) 20103) 26-9 (1-7) 3 175 23) 274A03)3 16-2 (1-6) 29.6 (1-1) 3 22.5 (1-2) 25-9 (1-8) N 10-2 (1-1) 20-5 (2:6) 2 17-7 (1-1) 20-8:(1-2)3 8-5 (0-8) 23-6 (1-2) 3 8.5 (0-8) 30-9 (1-7) 3 17-5 (1-2) 34-4 (2-4) 3 11-8 (0-8) 28-0 (1-1) 3 12-0 (0-9) 44-1 (1-9) 3 19-3(1-0) 39-4 (1-8) 3 13-8 (0-6) 19-6 (1-1) 2 7-9 (0-9) 22-3 (1-4) 3 14-7 (0:5) 25-1 (1-4) 3 10-4 (0:6) 28-0 (0-9) 3 8-2 (0-8) 25-4 (0-9) 3 21713) 42-9 (2-1)3 29-9 (2:8) 30-9 (0-8) N 26-4 (2:2) 41-4 (0-8) 3 30-4 (1-4) 39-1 (1-0) 3 20-3 (1-0) 32-8 (2:6) 3 28-1 (0-9) 28-7 (1-6) N 14-8 (0-7) 23-5 (0-2) 3 10-8 (1-0) 25-2 (1-9) 3 15-5 (0-8) 16-4 (1-1) N 19-7-1-0) 20-8 (0-8) 3 10-8 (0-7) 14-5 (0-4) 3 Panicle length (cm) 3-5 (0-3) 6:20-1)3 3-9 (0-2) 6-2 (0-2) 3 3-1 (0-2) 6-0 (0-1) 3 3-1 (0-2) 5-5 (0-2) 3 2-5 (0-2) 4-5 (0-3) 3 3-5 (0-2) 5-5 (0-2) 3 2-8 (0-3) 5-5 (0-1) 3 2-8 (0-3) 5-4 (0-2) 3 4-1 (0-2) 6-8 (0-1) 3 3.2 (0-2) 5-5 (0-2) 3 2-8 (0-1) 6-9 (0:2) 3 4-1 (0-1) 6:5 (0:2) 3 3-9 (0-2) 4-4 (0-2) N 2.3 (0-1) 5-0 (0-2) 3 3.3 (0-1) 5-8 (0-3) 3 2-9 (0:3) 4-6 (0-2) 3 2.2 (0-1) 5:7 (0-2) 3 3-5 (0-2) 6-2 (0:2) 3 5-9 (0:3) 6:8 (0:2) N 4-8 (0-2) 6-6 (0-13 5-4 (0:3) 6-9 (0-2) 3 4:3 (0-2) 6-6 (0:3) 3 5-6 (0:3) 5-6 (0-1) N 3-1 (0-2) 5.3 (0-2) 3 2.7 (0-2) 5-4 (0-2) 3 3.3 (0-2) 4-1 (0-1) 2 4-1 (0-2) 5.0 (0-1) 2 3.3 (0-2) 4-4 (0-1) 3 41a = dunes; 41b = outcrop; 44a = roadside; 44b = foreshore; 44c fixed dunes Numbers in bold indicate levels of significance for t-tests between the field and glasshouse populations. N = not significant, 1 = p at 0-05, 2 = p at 0-01 and 3 = p at 0-001 Panicle width (mm) n/a 4-6 (0-2) 3-8 (0-2) 4-4 (0-2) N 3-4 (0-2) 5-0 (0-2) 3 3-7 (0-1) 4-9 (0-2) 3 3-6 (0-2) 4-6 (0-3) 2 3-6 (0-2) 4-2 (0-2) 1 4-2 (0-2) 4-9 (0-2) 2 4-2 (0-2) 4-4 (0-2) N 3-3 (0-1) 3-9 (0-2) 2 2-9 (0-1) 4-6 (0-1) 3 2-9 (0-1) 4-9 (0-2) 3 3-8 (0-2) 4-7 (0-2) 1 3-7 (0-2) 5-7 (0-2) 3 3-1 (0-2) 6-0 (0-7) 3 3-9 (0-2) 5-8 (0-3) 3 3-4 (0-3) 4-7 (0-1) 3 2-9 (0-1) 4-2 (0-1 )3 3-3 (0-1) 3-9 (0-3) 1 3-9 (0-1) 5-1 (0-2) 3 3-8 (0-2) 4-7 (0-1) 3 3-7 (0-2) 4-8 (0-2) 2 3-9 (0-2) 4-5 (0-3) N 4-1 (0-1) 4-6 (0-2) N 3-3 (0-1) 4-4 (0-1) 3 3-0 (0-2) 4-5 (0-3) 3 4-1 (0-2) 5-6 (0-3) 3 4-3 (0-2) 5-4 (0-2) 2 4-9 (0-2) 6-1 (0-3 )2 Spikelet length (mm) 4-9 (0-2) 5-8 (0-2) 3 5-1 (0-2) 5-6 (0-2) N 4-8 (0-2) 5-4 (0-1) 2 4-8 (0-1) n/a 4-3 (0-2) 5-9 (0-3) 2 4-8 (0-3) 5-6 (0-1) 1 4-8 (0-2) 6-1 (0-2) 3 5-2 (0-2) 5-8 (0-2) N 5-2 (0-2) 6-0 (0-1) 3 4-4 (0-2) 5-3 (0-1) 2 4-2 (0-1) 5-2 (0-2) 2 4-3 (0-2) 5-7 (0-3) 2 4-8 (0-1) 5-1 (0-2) N 4-8 (0-1) 5:3: (0-1). 4 4-9 (0-1) 6-4 (0-1) 3 4-3 (0-1) 5-3 (0:2) N 4-3 (0-1) 5-3 (0-2) 2 4-8 (0-2) 5-4 (0-2) 1 4-5 (0-2) 5-4(0-1)N 4-7 (0-2) 5-0 (0-1) 2 4-9 (0-1) 5:3 (0-2) N 4-9 (0-2) 5-4(0-1)N 5-3 (0-3) 5-3 (0-3) N 4-3 (0-3) 5-0 (0:2) N 4-8 (0-2) 5:7 (0-1) 1 4-8 (0-2) 5-3 (0-2) N 4-8 (0-1) 5-9 (0-3) 2 5-0 (0-2) 5-6 (0-1) 1 Watsonia 23: 391-400 (2001) 391 The decline of Spartina alterniflora (Poaceae) in the British Isles L. C. MASKELL and A. F. RAYBOULD Centre for Ecology and Hydrology, Winfrith Technology Centre, Dorchester, Dorset DT2 8ZD ABSTRACT Spartina alterniflora was introduced from the U.S.A. in the 1820s. It is one of the parents of the fertile amphidiploid Spartina anglica and is an important resource for the study of polyploid speciation. Throughout the 20th century the distribution of S. alterniflora has declined and it is currently found at only one site, in Southampton Water. This study compares present and past distributions and discusses reasons for its decline, both throughout the harbour and at the one remaining site. Industrial and marine developments have resulted in loss of habitat as well as causing changes in tidal action, sediment deposition and increased pollution. These have all contributed to the decline of Spartina alterniflora. Competition with Spartina anglica has also played a role. KEYWORDS: decline, distribution, polyploid speciation Spartina anglica. INTRODUCTION Spartina alterniflora Lois. is a perennial grass native to saltmarshes of the eastern seaboard of North America. Around 1820 it was unintentionally introduced to mudflats on the River Itchen, Hampshire (the earliest confirmed record of S. alterniflora is a herbarium specimen collected by Borrer from the Itchen in 1829), from where it spread throughout Southampton Water (Marchant 1967). In some areas, S. alterniflora grew beside S. maritima (Curtis) Fernald, a plant of Old World saltmarshes, which is now extinct in Southampton Water (Gray et al. 1999). Fertilisation of S. alterniflora by S. maritima pollen (Ferris et al. 1997) produced a sterile hybrid now named Spartina x townsendii Groves & J. Groves. The first record of this plant is from Hythe in 1870 (Groves & Groves 1880). Chromosome doubling in S$. x townsendii formed Spartina anglica (Marchant 1968; Raybould er al. 1991a), a fertile amphidiploid species. Spartina anglica spread naturally along the south coast of England, the Isle of Wight and the north coast of France by seed and by rhizome fragments (Oliver 1925). It was also introduced into many saltmarshes in Britain and overseas for coastal defence and land reclamation (Ranwell 1967). Its rapid spread, and in the south of England a rapid decline, raised concerns about changes to shipping channels and effects on populations of wading birds and other species occupying intertidal mudflats (Gray et al. 1991; Raybould 1998). In addition to economic and nature conservation relevance, S. anglica has immense scientific importance. Because the species formed so recently, and because both its parental species are extant, it provides a superb opportunity to study the ecological and genetic consequences of polyploid speciation, a key process driving plant evolution. However, throughout the 20th century S. alterniflora has declined in Southampton Water. It is extinct in the River Itchen and River Hamble, and since the early 1960s the only remaining population of Spartina alterniflora in the British Isles has been at Bury Marsh, Marchwood, Southampton Water. The map in Fig. | shows the location of the River Hamble, River Itchen and the Marchwood site. This population is the last remnant of the S. alterniflora material involved in the origin of S. anglica and is therefore a unique resource to evolutionary biologists. THE HISTORY OF SPARTINA ALTERNIFLORA IN THE BRITISH ISLES Marchant (1967) gives a detailed account of the spread of S. alterniflora. At the height of its distribution, S. alterniflora grew as a more-or-less continuous sward on the east shore of the Itchen. There were also two smaller patches on the west shore of the Itchen. In Southampton 392 L. C. MASKELL AND A. F. RAYBOULD River Itchen 6 0 Marchwood site 0 FIGURE |. Map of southern Britain with inset of Southampton water showing the locations of the River Itchen, R. Hamble and the last remaining site for Spartina alterniflora; Bury Farm, Marchwood. Water/River Test upstream of the Itchen, S$. alterniflora occurred as several discontinuous swards from Hythe Marina to Cracknore Hard, and as a large sward on the Bury Farm marshes as far upstream as Eling Churchyard. The species also occurred on the eastern shore of the Test from Redbridge to Millbrook. Downstream of the Itchen, S. alterniflora grew at Hythe in a large patch about half a mile south of the pier. S. alterniflora was also common in the Hamble at Burlesdon, Lincegrove, Hamble-le-Rice and Warsash and occurred as far downstream as Hillhead at the mouth of the Meon. It is likely that S. alterniflora spread through Southampton Water by vegetative reproduction only. Bromfield (1836) commented that S. alterniflora in the Itchen “seldom perfects its fruit” and Marchant (1968) states that “British” S. alterniflora always “fails to set seed”. This is also the case in New Zealand (Partridge 1987). This may be due to low pollen fertility (Marchant 1968). It is possible that the introduced S. alterniflora is a single clone (Raybould et al. 1991b) and because some clones of S. alterniflora are self-incompatible (Daehler 1998) even if viable pollen is produced, self-incompatibility may prevent any seed set in “British” S. alterniflora. Recession of S. alterniflora began around 1920 (Marchant & Goodman 1969) and by the 1930s it was extinct in the Hamble and at Hythe. By 1924, the populations at Eling and Redbridge to Millbrook were extinct. A small population at Dibden was destroyed by reclamation in 1963; this left the Bury Farm marshes at Marchwood as the final site for the S. alterniflora material involved in the evolution of S. anglica. The main objectives of this work were to map the current distribution of S. alterniflora on the Bury Farm marshes and to survey areas in Southampton Water and the Rivers Test, Itchen and Hamble where S. alterniflora once occurred to assess why the populations had declined. DECLINE OF SPARTINA ALTERNIFLORA 393 SS yo IN, Transect 1 NY Cara “.\, Transect 2 Nee /\ Transect 3 aye NS MS only Spartina altemifiora Ao Mixed; Spartina alterniflora and S.anglica oy = <4 9 uo 200 Meters — v4 = ge mee vi FIGURE 2: The distribution of Spartina alterniflora at Marchwood, Bury marsh, Southampton Water. MATERIALS AND METHODS BURY MARSH: MAPPING SPARTINA ALTERNIFLORA Patches of Spartina alterniflora were identified from the morphological characters in Marchant (1964) and Hubbard (1984). Each patch was mapped using a Magellan promark X hand held GPS unit, taking measurements of dimensions and recording associated vegetation. The data were imported to ARCVIEW to create a digital map of the marsh. This survey was carried out in October/November 1999. BURY MARSH: PHYSICAL NICHE AND ASSOCIATED VEGETATION The topography of the marsh was surveyed in October/November 1999 using a Pentax Total Station electronic theodolyte. Transects were set up across the marsh (Fig. 2). Transect 1 was placed in the middle of the marsh and passed through several patches of Spartina alterniflora. Transect 2 was established to the western side of the marsh and crossed a particularly low-lying patch of S. alterniflora which lies at the back of the marsh near to an equally low-lying patch of S. anglica. Transect 3 was placed to the eastern side of the marsh where there is far less S. alterniflora present. Measurements were taken at each point where the vegetation or topography changed significantly and these have been plotted onto profile diagrams. The plant species found have been related to the height (Ordnance Datum) at which they were found. SURVEY OF PREVIOUS SITES In addition to the above, the sites around Southampton water where Spartina alterniflora had been recorded previously were visited to assess possible reasons for its loss. RESULTS BURY MARSH: MAPPING SPARTINA ALTERNIFLORA Figure 2 shows the distribution of S. alterniflora on the Bury Farm Marsh and the extent of the 394 L. C. MASKELL AND A. F. RAYBOULD taj eee a “ oa. t a Ui i = FIGURE 3a. Spartina alterniflora at Bury Marsh, Marchwood. This photo shows the topography of the marsh; there is a large creek system. FIGURE 3b. Spartina alterniflora growing at the front of the marsh, which is being visibly eroded. DECLINE OF SPARTINA ALTERNIFLORA 395 marsh in relation to neighbouring coastal features. Figure 3a shows a view across the marsh with S. alterniflora in the foreground and a part of the extensive creek system in the background. Only two patches appear to be completely mixed stands of S. alterniflora and S. anglica, although five other patches have S. anglica beginning to invade at the edges. The “mixed” and “pure” patches are marked in Fig. 2. It is noticeable that S. alterniflora is now absent from areas described as “mixed” by Marchant (1964), indicating that S. anglica has replaced it. Atriplex portulacoides L., Limonium vulgare Miller, Aster tripolium L. and Triglochin maritimum L. are all present in the patches. Atriplex portulacoides and Puccinellia maritima (Huds.) Parl. are particularly abundant. Only one stand appears to be purely S. alterniflora with no other species and this is found at the front of the marsh to the north west of the map. This stand can be seen in Fig. 3b. The area in which the S. alterniflora is growing 1s visibly eroding. Most patches of S. alterniflora are on high parts of the marsh in areas with high species diversity. The species tends not to occur right on the edges of deep creeks where Atriplex portulacoides grows in pure stands. It is also found growing in low-lying areas of the marsh. The map in Fig. 4 shows the distribution of S. alterniflora at the site in 1964. This map has been created using information in Marchant’s thesis (1964). The distribution shown in Marchant’s map has been superimposed onto the shape of the marsh as it is today. The marsh has changed in shape and extent since the original map so this new version is a pictorial representation rather than a distribution map. It does, however, enable comparisons to be made between 1964 and the present. The abundance of S. alterniflora has decreased. In 1964, the eastern side of the marsh contained several mixed stands of S. anglica and S. alterniflora. These were restricted to the back of the marsh, while the front of the marsh was dominated by S. anglica. Now there are only a few small patches of S. alterniflora on the eastern side of the marsh and two of these are still mixed stands today. Spartina alterniflora is concentrated on the west side of the marsh. In 1964, it was 7 Spartina alterniflora only M07 2 iNiixed: Spartina anelica and S.akemificea FIGURE 4. Map based on Marchant’s 1964 map showing the distributions of Spartina alterniflora, S. anglica and mixed swards. 396 L. C. MASKELL AND A. F. RAYBOULD distributed throughout this part but has definitely contracted. Marchant (1964) describes some of the stands as degenerate. He noted that although the largest pure stands of S. alterniflora grew in the low-lying areas, these swards were less vigorous and had lower pollen fertility. On the higher sites, although other plant species are present, the individuals of S. alterniflora appear to be much healthier. BURY MARSH: PHYSICAL NICHE AND ASSOCIATED VEGETATION The three marsh profiles are quite similar and can be seen in Figs. 5a, 5b and 5c. The height at the top of the marsh was between 2-2-5 m OD. The raised parts of the marsh were all c. 1-5 m OD and the creeks c. 0-5 m OD. Puccinellia maritima tended to be found on higher elevations around 1-5 m OD or above and consequently was more abundant towards the back of the marsh. There appeared to be no particular pattern to the distribution of S. alterniflora. In Transect 1 S. alterniflora was found from 1-58—1-6 m OD. It occurs in Transect 2 from 1-1—1-59 m OD with S. anglica and at 1-67 m OD in Transect 3 with S. anglica. Spartina anglica was found from 1-2-1-87 m OD in Transect 1, from 1-2—1-59 m OD in Transect 2 and from 1-42—1-88 m OD in Transect 3. SURVEY OF PREVIOUS SITES Many of the sites where S. alterniflora was recorded in the past (before 1910) have now been developed for docks and port facilities such as the river Itchen and the area between Hythe and Cracknore Hard. The river Test has also been developed, the eastern shore now consists of docks and port facilities whilst the Marchwood site on the western shore is quite isolated. S. alterniflora used to be found along a stretch of Southampton Water east of Weston point. However, the shoreline here is now bare mud and shingle; the S. alterniflora habitat has completely disappeared, presumably due to erosion. There is a similar situation at the Hook (near to the College of Maritime Studies, on the shore near Newtown (SU4905) from where S. alterniflora was last recorded in 1907. There is no longer any salt marsh here and the substrate is composed of shingle and sand rather than mud. All that remains of the marsh is a small patch of S. anglica. S. alterniflora was common in the Hamble at Burlesdon, Lincegrove, Hamble-le-Rice and Warsash. There has been less development of the river Hamble than the river Itchen and there are Height OD (m) 25 - Festuca Mixed species, rubra, Puccinellia od Aster maritima, Spartina tripolium Spergularia media, alterniflora Aster tripolium, ii ‘i some Spartina anglica f Bare mud at the bottom of the Atriplex portulacoides at the edges of creeks, | creeks, Atriplex portulacoides ae Aster tripolium, Puccinellia maritima on the _on the edges. top, mud at the bottom of creeks. ib: ge 0.5 [- ; bed 0) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Distance (m) DECLINE OF SPARTINA ALTERNIFLORA 397 Height OD (m) 25 : : Spartina anglica, Puccinellia maritima, Aster ; tripolium, | Spartina Limonium vulgare | : alterniflora, Puccinellia maritima, : Spartina 2 Spartina alterniflora, : anglica, Spartina anglica and :Puccinellia Atriplex : maritima and : ; ; portulacoides : Aster tripolium Spartina alterniflora and i i bare mud Atriplex portulacoides and bare mud Atriplex portulacoides and Spartina anglica 0.5 0 10 20 30 40 50 60 70 80 90 100 Distance (m) Height OD (m) 2.2 ; : Spartina anglica, Puccinellia _ Spartina anglica, maritima, Aster : Puccinellia B | tripolium, ; eta | maritima, Aster Atriplex Spartina anglica, tripo lium, portulacoides, Spartina alterniflora Atriplex Puccinellia i | id 1.8 _maritima, Aster 'P ORUIaCOlaes, :tripolium, Spartina anglica, 1.6 : 2 P ae :Puccinellia maritima, : Aster tripolium, Atriplex :portulacoides, 1.4 Puccinellia maritima, Aster j 12 | tripolium, aaa vet Triglochin i j maritimum, Creek, Bare mud 1 |Spergularia media 0.8 0.6 5c 0 20 40 60 80 100 120 140 160 180 200 220 Distance (m) FIGURES 5 a., b., & c. Topographic profiles of the transects which were established at Bury Marsh, Marchwood. The most abundant plant species have been added to the profile. 398 L. C. MASKELL AND A. F. RAYBOULD Height OD (m) 3.5 Juncus | Spartina anglica, ae i effusus, | Puccinellia maritima, | Spartina anglica, Holcus _ Spergularia media, Mud in creek: Puccinellia maritima, 3 |lanatus, _ Salicornia spp. and Atriplex Atriplex _ Festuca portulacoides portulacoides VP"? Holcus pheidee : Spartina 25 lanatus, Spartina anglica, anglica Puccinellia | Puccinellia maritima, | age maritima Atriplex portulacoides 2 Salicornia spp. Puccinellia maritima, " Spergularia media, Festuca ‘Bare mud | rubra, NY Salicornia | an) SPP... | 2 : _ | Bare mud (0) 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 Distance (m) FIGURE 6. The topographic profile of the transect established at Hacketts marsh on the River Hamble. still areas of mature marsh at Burlesdon (Lincegrove and Hacketts), that have been given protected status. S. alterniflora was last recorded from these marshes pre-1910. The reasons for decline of S. alterniflora here are unclear. Unlike the river Itchen it is not due to loss of habitat through development. We established a transect on Hacketts marsh to compare the topographic profile and vegetation with Bury Farm Marsh, Marchwood. This profile can be seen in Fig. 6. The top of the marsh is around 2 m OD in height. There is a flat area containing a lot of Puccinellia maritima which has been quite heavily grazed between 1-5 m OD and 2 m OD. The bottom of the creek is approximately 1-7 m OD. The other raised marsh areas are at a height of 1-5 m OD and contain small amounts of Atriplex portulacoides, Salicornia spp. and Spartina anglica. S. anglica is found throughout the marsh, growing between heights of 0-99-1-91m OD. At the front it grows on top of cliffs approximately | m high. It is not found on the low-lying bare mud at the front of the marsh and does not form as dense stands as it does at Marchwood. Puccinellia maritima is abundant particularly at the back of the marsh. Other marsh species include Aster tripolium, Limonium vulgare, Atriplex portulacoides, Juncus maritimus Lam., Salicornia spp. and Triglochin maritimum. DISCUSSION REASONS FOR THE DECLINE OF SPARTINA ALTERNIFLORA As well as the direct impacts of industrial and marine developments such as loss of saltmarsh habitat there are indirect impacts such as changes in wave action and tidal currents, changes in sediment deposition and pollution which have implications for the whole of the estuary system, including the existing site at Marchwood. There has been much erosion of saltmarshes and many areas where S. alterniflora used to grow are now low-lying mud or shingle, the saltmarsh having completely disappeared. DECLINE OF SPARTINA ALTERNIFLORA 399 It is not clear why S. alterniflera has persisted at Marchwood, while becoming extinct on saltmarshes in the Hamble. Marchant (1964) suggested that Marchwood is unique in its uneven topography; however, this was not based on topographical levelling and our study suggests that Hacketts Marsh is not dissimilar. At Marchwood, the S. alterniflora population has declined significantly since 1964. This is probably due to an interaction of factors. Spartina anglica has played a role, as it has completely replaced S. alterniflora in areas where mixed swards once occurred. The degree of tidal inundation is considered to be a dominant factor influencing plants in the intertidal zone; it affects a number of factors, including degree of anoxia, salinity and build-up of chemicals. Determination of the level above Ordnance Datum at which a species occurs should give some idea of the ecological conditions that it prefers or can tolerate and suggest factors that limit its distribution. The topographic levelling showed little difference in height above Ordnance Datum between the upper and lower limits of the two. However, competition between the two species is not the only factor in the decline of S. alterniflora on uneroded marsh at Marchwood. Marchant (1964) noted that pure stands of S. alterniflora were showing signs of decline even though S. anglica was not present. Pollution may have affected vegetation generally by weakening the plants. Marchant (1964) suggested that reclamation work downstream altered tidal movement and scour in the estuary which reduced the size of the marsh and deepened the channel. The regular dredging of the deep- water channel may be important. The soft mud at the surface of the marsh may have crept laterally into the deep channel and lowered the level of the overlying marsh. The dredgings are taken to the reclamation area so the supply of water-borne silt has been reduced, which causes the marsh accretion rate to be reduced. The lowering of the marsh increases waterlogging and tidal immersion of sites. In low-lying waterlogged sites the sediment that is deposited is also more easily removed by re-suspension than at sites of higher elevation where it may have a chance to dry out and become part of the marsh surface so the process is cyclical and self-enforcing (Reed & Cahoon 1992). Finally, there may be an age-related decline in vigour because there is no sexual reproduction of S. alterniflora in Britain. Clones of S. anglica decline in vigour with age (e.g. Thompson et al. 1991) and it is likely that a similar process occurs in S. alterniflora. Marchant (1967) noted that herbarium specimens of S. alterniflora collected in the 19th century are larger than plants growing at Marchwood in the 1960s. CONCLUSIONS The loss and decline of Spartina alterniflora from many of the sites in Southampton water appears to be due to changes in land use. This includes direct impacts from construction as well as indirect effects from changes in tidal action, channel topography and increased pollution. Spartina anglica may have out-competed S. alterniflora in other areas. At some of the sites such as Hacketts and Lincegrove marsh in the Hamble it is difficult to determine an obvious impact and loss may be due to a combination of factors. We consider the best strategy for conservation of S. alterniflora to be preservation of its habitat at Marchwood. Periodic monitoring of the site would determine whether erosion at the front of the marsh is continuing and whether S. alterniflora populations are being invaded by other species. If the dieback of S. alterniflora on the remaining sections of marsh since 1964 is due to the loss of sediment, it would perhaps be appropriate to consider whether the marsh can be protected. Future research should examine the physical processes taking place at this site and determine the rate of erosion, size of particles and types of sediment which are involved in marsh construction. Possible actions for protection may include deposition of dredging spoil which is already being removed from the deepwater channel nearby. The low-lying areas at the western end of the marsh might be most suitable for deposition. Control of S. anglica may also be needed. ACKNOWLEDGMENTS This work was commissioned and supported by Ian Pearson from English Nature. We would like to thank Duncan Hornby and Kathy Hodder for help and advice on GIS. 400 L. C. MASKELL AND A. F. RAYBOULD REFERENCES BROMFIELD, W. A. (1836). ‘A description of S. alterniflora of Loiseleur, a new British species’. J. D. Hooker’s companion to Botanical Magazine 2: 254. DAEHLER, C. C. (1998). Variation in self-fertility and the reproductive advantage of self-fertility for an invading plant (Spartina alterniflora). Evolutionary ecology 12: 553-568. FERRIS, C., KING, R. A. & GRAY, A. J. (1997). Molecular evidence for the maternal parentage in the hybrid origin of Spartina anglica C. E. Hubbard. Molecular ecology 6: 185-187. GRAY, A. J., MARSHALL, D. F. & RAYBOULD, A. F. (1991). A century of evolution in Spartina anglica. Advances in ecological research 21: 1-61. GRAY, A. J., RAYBOULD, A. F & HORNBY, D. (1999). Solent Spartina survey: A survey of the non-S. anglica species in the Solent. /.7.E. Report for English Nature. GROVES, H. & GROVES, J. (1880). Report of the Botanical Exchange Club 37: 1880-1881. HUBBARD, C. E. (1984). Grasses. Penguin Books Ltd, Harmondsworth. MARCHANT, C. (1967). Evolution in Spartina (Gramineae). I. History and morphology of the genus in Britain. Botanical journal of the Linnean Society 60: 1-24. MARCHANT, C. (1964). Cytonomic studies in the genus Spartina Screb. Ph.D. thesis, University of Southampton. MARCHANT, C. (1968). Evolution in Spartina (Gramineae). I. Chromosomes, basic relationships and the problem of the S. x townsendii agg. Botanical journal of the Linnean Society 60: 381-409. MARCHANT, C. J. & GOODMAN, P. J. (1969). Spartina alterniflora Loisel. Biological Flora of the British Isles. Journal of ecology 57: 291-295. OLIVER, F. W. (1925). Spartina townsendii; its mode of establishment, economic uses and taxonomic status. Journal of ecology 13: 75. PARTRIDGE, T. R. (1987). Spartina in New Zealand. New Zealand journal of botany 25: 567-575. RANWELL, D. S. (1967). World resources of Spartina townsendii (sensu lato) and economic use of Spartina marshland. Journal of applied ecology 4: 239-256. RAYBOULD, A. F., GRAY, A. J., LAWRENCE, M. J. & MARSHALL, D. F. (1991a). The evolution of Spartina anglica C. E. Hubbard (Gramineae): origin and genetic variability. Biological journal of the Linnean Society 43: 111-126. RAYBOULD, A. F., GRAY, A. J., LAWRENCE, M. J. & MARSHALL, D. F. (1991b). The evolution of Spartina anglica C. E. Hubbard (Gramineae): genetic variation and status of the parental species in Britain. Biological journal of the Linnean Society 44: 369-380. RAYBOULD, A. F. (1998). The history and ecology of Spartina anglica in Poole Harbour. Proceedings of the Dorset Natural History and Archaeological Society 119: 147-158. REED, D. J. & CAHOON, D. R. (1992). The relationship between marsh surface topography, hydroperiod and a growth of Spartina alterniflora in a deteriorating Louisiana saltmarsh. Journal of coastal research 8(1): 77-87. THOMPSON, J. D., MCNEILLY, T. & GRAY, A. J. (1991). Population variation in Spartina anglica C. E. Hubbard. II. Reciprocal transplants among three successional populations. New phytologist 117: 129- 139. (Accepted August 2000) Watsonia 23: 401-411 (2001) 401 Conservation of Britain’s biodiversity: Carex depauperata With. (Cyperaceae), Starved Wood-Sedge iC. G- RICH Dept. Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP and C.R. BIRKINSHAW Missouri Botanical Garden, BP 3391, Antananarivo 101, Madagascar ABSTRACT Fifteen years conservation work on the rare, statutorily protected species Carex depauperata With. (Cyperaceae), Starved Wood-sedge in Britain are summarised. In the British Isles it has been recorded historically from 14 sites in v.cc. 6, 9/11, 16, 17, 52, 83, 90 and H5. About 60 plants are currently known in two native sites in England and one in Ireland, and it has been transplanted to two more sites. It is widespread in Europe and Asia though rare in many countries, with the greatest frequency of records in France and Spain. It is a distinct species showing little variation. It is a polycarpic perennial that usually grows in dry, deciduous woodlands on a range of soil types, and is often associated with tracks. It is a woodland gap plant that responds to opening of woodland canopies through coppicing and disturbance, and declines if management is neglected and shade increases. Periodic (c. every five years) disturbance of its woodland habitat by coppicing or scrub clearance is recommended as management for conservation. KEYWORDS: Rare species, habitat management, critically endangered species, distribution, ecology. INTRODUCTION Carex depauperata With. (Cyperaceae), Starved Wood-sedge, is a very rare plant in the British Isles. It is currently known from two sites in Britain and one in Ireland, all with small populations. In Britain it is listed as Critically Endangered (the highest I.U.C.N. threat level), and is protected by Schedule 8 of the Wildlife and Countryside Act 1981 (Wigginton 1999). In the Republic of Ireland it is listed on the Flora (Protection) Order, 1999. It is also reputed to be rare and scattered across much of its European range, and often occurs in tiny populations. It is extinct in Germany and Luxembourg, and is described as rare in Belgium, Corsica, Italy, Switzerland and Turkey (Blab et al. 1984; Delvosalle et al. 1969; G. Aymonin, pers. comm., 1989; Landolt 1989; Nilsson 1985). Although apparently once reduced to a single plant and on the verge of extinction, British populations of C. depauperata are now recovering and increasing through a combination of restocking, research, appropriate habitat management and pure chance. In this paper, we summarise the data from about 15 years detailed work so that other conservation projects may benefit from our experience. Full details are given in Birkinshaw (1990, 1991), Rich (1992, 1994a, b), Rich & Fairbrother (1994, 1995, 1996) and McDonnell (1997). DISTRIBUTION AND CONSERVATION IN THE BRITISH ISLES Data have been abstracted from the literature, herbaria (ABD, BM, BRISTM, BTN, CGE, CLE, DBN, DEE, E, GLAM, K, LIV, LTR, NUW, OXF, PTH, RNG, SLBI and TCD), field survey and correspondence with botanists. 402 T. C. G. RICH AND C. R. BIRKINSHAW 1. Axbridge, North Somerset (v.c. 6) Plants were first found in a wood between Axbridge and Cheddar in June 1860 by T. B. Flower (BM) but the locality was later reported to have been destroyed (White 1912). In 1911 H. W. Pugsley re-discovered it by a lane in the same area (but possibly in a different site) occurring thinly on an open grassy bank over 100 yards long (White 1912). H. S. Thompson found more, with an abundance of fruiting spikes, in 1915 (E). The site has since been visited many times and its changes in population are well documented. A photograph taken by J. E. Lousley in 1937 shows C. depauperata on an open, flowery bank without scrub or hedges (RNG). The site gradually became overgrown, and by the late 1950s one last plant remained on a shaded bend of the lane which was continually being damaged by farm vehicles (R. M. Harley, pers. comm., 1981). The number of plants increased slightly during the next two decades, and in 1974 R. S. Cropper found six plants. Concerned for their conservation, he began to monitor them annually and initiated restocking, possibly single-handedly maintaining the species in the wild in Britain for the next decade. Cropper’s detailed monitoring notes were kindly provided to Rich & Fairbrother (1994), and are briefly summarised, with unpublished notes by R. A. Jarman, as follows. By 1977, the six plants were reduced to two due to damage from tractors, and there was no further change until 1981. There was light clearance of Rubus around the plants in the winters of 1977/78 and 1978/79, and the hedge opposite the plants was cut back in winter 1978/79. In October 1981, several plants grown from a cutting taken a few years earlier were re-introduced. Between 1982 and 1990, between four and six plants were counted annually, and in the winter of 1991/1992 or 1992/1993 a second transplant took place, with nine plants being present in 1994. The hedge on the east side of the lane was cut again in 1990. In 1993 following the success at Ockford Wood (see below), Plantlife became involved in management of the site in consultation with English Nature and Somerset Wildlife Trust. In 1994, the large evergreen Quercus ilex which was shading the plants was cut down and a carpet of Hedera helix and invasive Prunus spinosa suckers removed along a 40 m stretch. One year later the average diameter of the plants had increased by c. 50% and plants were much more floriferous. C. Hancock noted heavy grazing of inflorescences, so a small deer track was blocked resulting in a ten-fold increase in the number of inflorescences in 1996. By 1999, there were at least 55 plants with many seedlings with several dense patches (one patch c. 7 x 1 m dominated by C. depauperata alone), and a new plant occurred about 50 m further down the lane. 2. Leigh Woods, North Somerset (v.c. 6) A single immature specimen was gathered 1886 by H. S. Thompson, but was not identified until a few years later by A. Bennett, by which time Thompson could not remember the exact place he had gathered it (BRISTM; White 1912). This is still the only record for a very well-recorded National Nature Reserve. 3. Templecombe, North Somerset (v.c. 6) Murray (1896) reported it from Templecombe but it has not been seen again and no further information is available. 4. Cranborne to Damerham, Dorset (v.c. 9)/South Hants (v.c. 11) A herbarium specimen collected in the 1920s by H. H. Haines from hedgebanks between Cranborne and Damerham was discovered in LTR in 1984. The area concerned has been searched without success, though the exact site is not known (Pearman 1994). 5. Charlton Wood, Greenwich, West Kent (v.c. 16) Carex depauperata was described as new to science from Charlton Wood after being found by Mr Woodward in 1787. It was last recorded there by R. R. Hutchinson in 1830 (SLBI). This area has now been built over. 6. Effingham chalk pit, Surrey (v.c. 17) “Two or three tufts only” were first seen by W. W. Reeves in 1874 (Reeves 1874) and last recorded in 1881 by F. W. Ward (RNG). This is presumed to be the same site as “Bookham”. CONSERVATION OF CAREX DEPAUPERATA 403 7. Ockford Wood (Westbrook Copse), Godalming, Surrey (v.c. 17) Plants were first found at Godalming by W. Borrer in 1807 (E), though the exact site is not known. It was first reported by J. D. Salmon at “Westbrook Woods” in 1843 (CGE) and was well- documented until its demise in the 1970s. E. C. Wallace photographed the plant on an open flowery bank in 1936 (RNG). In the 1940s, the plant was locally frequent in the wood, along the track side, on the bank, at the top of the bank and on the slopes above (F. Rose, pers. comm., 1992). J. E. Lousley saw “five fine plants by lane” in 1949 (RNG). R. W. David and O. Polunin saw five plants “one plant in gutter, two on bank above, and two more just into the wood at the top of the bank and 10 yards downhill” in 1961. By 1970, one plant remained on the edge of a precarious bank which Mrs J. E. Smith and W. E. Warren took measures to shore up. F. Rose and E. C. Wallace last saw one vegetative plant inside the wood in 1972. This plant seems to have disappeared from the site about this time under a landslide (J. E. Smith, pers. comm., 1993). J. E. Lousley and Mrs J. E. Smith searched all the known sites in 1973 without success (Lousley 1976) and R. W. David failed to find the plant in 1974. In 1975 the Surrey Flora Committee again searched all the old localities and placed a request for information with a picture of the sedge in a local paper (one Carex sylvatica was the only response). Thorough searches by Mrs J. E. Smith with other botanists in 1986 and 1987 of many suitable areas around Godalming failed to find any plants. C. Birkinshaw failed to find it in 1989. It was thus something of a surprise when one flowering plant was found in 1992 by F. Rose, P. Marren and T. C. G. Rich in exactly the same spot as where it was last seen. During the Great Storm of 1987, a large branch had been blown down from a Tilia x vulgaris tree directly above the old site which had opened up the canopy and probably stimulated buried seeds to germinate (Rich 1994b). The plant had remains of inflorescences from 1991, suggesting that it was in at least its third year. In November 1992, the canopy and hedge were thinned and the Hedera helix and Rubus fruticosus in the ground layer removed taking care not to damage the plant. A local botanist Mrs A. Fairbrother was recruited to keep an eye on the plant and keep the surrounding vegetation down, which has proved very successful for growth of the plant to the present day. 8. Frith Hill, Godalming, Surrey (v.c. 17) Plants were found in c.1842 by J. D. Salmon (Brewer 1863) and by many botanists up to 1891. 9. Charterhouse (Milton Wood), Godalming, Surrey (v.c. 17) This site was first reported in c. 1843 as “copse near Hurtmore Cottages” by J. D. Salmon (Brewer 1863). Lousley (1976) recorded that he last saw it in 1938 and that it disappeared during the war, probably after timber extraction. 10. Farnham, Surrey (v.c. 17) There is a specimen in DBN collected by W. Mclvor from “near Farnham” in August 1846 (Rich et al. 2000). 11. Porth-y-felin, Holyhead, Anglesey (v.c. 52) In 1936 A. G. and W. Holder discovered C. depauperata on rocks below the sea wall at Porth-y- felin, Holyhead (LIV; correspondence in NMW). In 1967 W. E. Hughes visited the site using the Holder’s directions and found one plant. This site appears to have been erroneously reported as on the edge of a quarry by Ellis (1983) which has misled many botanists. Searches in 1998 and 1999 by T. C. G. Rich, S. Ellis and R. A. Jones of the exact site have failed to refind the plant. Aerial photographs taken by the Luftwaffe in 1940-41 show the area around the rock to have changed little, though there is now an invasion of Fallopia japonica and a small concrete ramp has been constructed. Clearance and disturbance might regenerate the plant from the seedbank. 12. Bonaly, Edinburgh (v.c. 83) A specimen from “Wood above Bonelly” (assumed to be Bonaly) was collected by W. R. McNab probably sometime between 1860 and 1869 (DBN; Rich et al. 2000). This area, now largely a Regional Park, was searched by Mr C. Dixon without success in 1999 (pers. comm., D. McKean, 1999). 404 T. C. G. RICH AND C. R. BIRKINSHAW 13. Forfar (v.c. 90) There are two records which require confirmation, but which could refer to the same site. Hooker (1821) cited “woods near Forfar, G. Don”, which was accepted by Gardiner (1848) but rejected as doubtful by Ingram & Noltie (1981) as the origin could have been Don’s garden (the source of several other confusing records). Another specimen from “Forfarshire” collected by T. McFarlane (ex herb. Rev. H. G. Carroll, DBN) appears to have been collected independently suggesting the species has indeed occurred in the county (Rich et al. 2000). 14. Killavullen, East Cork (v.c. H5) Plants new to Ireland were discovered in scrubby open woodland on a rocky Carboniferous limestone outcrop by T. O’Mahony in 1973 (O’Mahony 1976), and a total of twelve plants were found on more detailed searches in 1974 and 1975. In 1984, these original plants were found to have gone extinct, but a new sub-population of six plants was found to the west. Two further sub- populations with two and three plants respectively were found in 1992. By July 1994, the sub- population of two plants had gone (despite being large vigorous plants in 1992), but the other two sub-populations survived though fluctuating in numbers. The site has a dense field layer of Hedera helix through which there are many small animal tracks with which the plant is associated. Some limited canopy clearance work took place in 1982. In 1989, although the canopy was beginning to close, 20 plants were recorded. The plant has since declined and by 1998 the population had declined to five small tussocks of which only three were fruiting (O’Mahony 1999). The site has been monitored regularly, and a full account of this site is in preparation by T. O’ Mahony. Errors or probable errors A specimen in DBN labelled “Leith Hill, Surrey, 30 June 1890, per J. Leitch” may be a corruption of Frith Hill (Rich et al. 2000). There is a second Frith Hill in Surrey at SU9058. It appears that the Godalming Frith Hill records have been erroneously plotted in SU95 in Perring & Walters (1990) and Wigginton (1999) (A. Lockton, pers. comm., 2000); all records are assumed to belong to the one at Godalming (SU9744). There is a specimen from Ranworth, Norfolk collected by W. Curnow in 1878 in E whose status is currently in doubt. The soils in this area are predominantly coarse loams or unsuitable wet fen land. There is currently no ancient woodland of more than 2 ha size in the parish, and the woods that are present are predominantly semi-natural woods managed for pheasant shooting, though there is some hazel coppice and oak woodland on the valley sides (R. Handley, R. Ellis, pers. comms., 2000). Curnow was an acute observer of natural history and specialised in lower plants, and made some outstanding collections (Davey 1909), but he seems to have botanised predominantly, though not exclusively, in Cornwall. Letters he wrote in 1876 indicate he was trying to obtain sedges and grasses for his collection from other botanists, and it may be that the specimen is of cultivated origin and Curnow confused the original locality (R. J. Murphy, pers. comm., 2000). It has thus been recorded from 14 sites in v.cc. 6, 9/11, 16, 17, 52, 83, 90 and HS. A distribution map summarising the records is given in Fig. 1. It has a somewhat disjunct distribution in the British Isles, characterised by a few long-known and persistent sites contrasting with the nearly half which are represented by one historical herbarium specimen only. In 1999 it was only known as a native in two sites in England and one in Ireland. Population counts for the last ten years for these sites are summarised in Table 1. | TRANSPLANT SITES 1. Cheddar Wood On 30 October 1977 C. E. D. Smith and R. A. Jarman, as part of a project with the Somerset Trust for Nature Conservation and Bristol Conservation Corps, transplanted 15 plants from Bristol and Cambridge University Botanic Gardens (both grown from seed collected at Axbridge in 1965 by D. E. Coombe) into each of two 10 m x 10 m quadrats, one cleared of ground, understory and overhanging vegetation, and the other under closed overgrown canopy nearby. The plants were reported heavily grazed soon after they had been planted. On 22 July 1979, all 15 plants had CONSERVATION OF CAREX DEPAUPERATA 405 FIGURE 1. Distribution of Carex depauperata in the British Isles. ® 1990+. O pre-1990. ? exact location uncertain. x error or probable error. 406 T. C. G. RICH AND C. R. BIRKINSHAW TABLE 1. NUMBER OF MATURE PLANTS AT CAREX DEPAUPERATA NATIVE SITES 1990-199] Site Godalming Axbridge Cork 1989 0 6 20 1990 i 4 - 1991 | 5 - 1992 1 6 ifs) 1993 3 - 17. 1994 3 14+ 7, 1995 3 14+ 10 1996 3 15+ uf 0d + 18+ i 1998 + 51+ 5 1999 + 55+ - Data for Godalming by extrapolation, counts for Axbridge include transplants at native site only; data for Cork courtesy of T. O’ Mahony. survived in the cleared plot, but only five small vegetative plants remained in the overgrown plot. One plant, probably a survivor from the original transplants was seen in the clearing in the mid- 1980s (R. Corns, pers. comm., 1995), but the exact location of the other plot is not now known. No plants were found by C. Birkinshaw in 1988 or 1989, or by T. C. G. Rich in 1993 and 1994 but, surprisingly, two plants were found in the clearing in 1995 by T. C. G. Rich, four in 1996, and nine in 1999, 2. Edge of Cheddar Wood In April 1988, 100 divisions of C. depauperata from Cambridge University Botanic Garden (origin as above) were planted inside and outside an exclosure in ungrazed pasture/woodland edge on the edge of Cheddar Wood to investigate the light, grazing and disturbance requirements. Eighteen months later all plants were still alive, though varying in performance. Plants subject to grazing were less fecund than those protected from grazing, plants placed in disturbed vegetation were much larger than those in undisturbed vegetation suggesting a benefit from reduced competition, and optimum fecundity occurred on the woodland edge (Birkinshaw 1990). By October 1991, 86 plants were still alive (the losses mostly due to competition from Hedera), and the main trends observed by Birkinshaw had continued (Edgington 1991). By 1993 plants in the enclosure were shaded by dense Rubus fruticosus, and fewer plants survived within it than outside. The fence was removed in 1998. At least 17 plants survived in 1999, mainly those planted outside the original enclosure indicating that competition affected survival. WORLD DISTRIBUTION There is little recent information available on its distribution or frequency in Europe. Data were compiled from herbarium specimens and literature to give a basic assessment of its distribution. Information was abstracted from non-British specimens at the following herbaria: BM, COI, FI. K, L, LTR, LIV, MA, NMW, OXF; P and RNG. Information was also abstracted from relevant Floras held in the library at the National Museum of Wales, although the Flora holdings are patchy and incomplete. The European records traced are mapped in Fig. 2. It has been recorded from Albania, Belgium, Bulgaria, Corsica, Croatia, England, France, Germany, Greece, Hungary, Ireland, Italy, Luxembourg, Romania, Scotland, Sicily, Spain, Switzerland and Wales. In addition, it has been recorded from Iran, Iraq, Turkey and the former U.S.S.R. (Caucasus, Ukraine, Pamir, Tien-Shan) (Nilsson 1985), but the records are too fragmented and unlocalised to merit plotting. The records traced show it is widely scattered across Europe and near Asia, though these are likely to be biased towards western European sources. France and Spain appear to be its main strongholds, and it is certainly still widespread in the latter. CONSERVATION OF CAREX DEPAUPERATA 407 FIGURE 2. European distribution of Carex depauperata compiled from herbarium specimens and the literature (all dates). ECOLOGY LIFE CYCLE Carex depauperata is a distinctive sedge, instantly recognisable across its range by its few flowers and large utricles. Vegetatively it is less distinct, and can be similar to C. sylvatica when young, but usually differs in forming dense clumps with more erect leaves which droop at the tips to form a “fountain”. The red coloration of the basal sheaths is distinctive when present; it is often more pronounced on herbarium specimens than in live material. Vegetative plants can be very difficult to find, and young plants are difficult to distinguish from other Carex species. No hybrids are known. Material from the two British, one Irish and one French site (the latter originally collected in southern central France by P. A. Harmes) was cultivated in an open herbaceous border on clay soils in a garden in Surrey by Mrs J. E. Smith between 1995 and 1999. After two years in cultivation, no significant morphological differences could be detected between the plants. Similarly, examination of a range of herbarium material from the western part of its range shows that it varies little morphologically, except a little in size due to local growth conditions. Davies (1956) reported a haploid chromosome number of n = 22 from a plant from Axbridge. Carex depauperata is a polycarpic hemicryptophyte. The diameter of clumps and the number of inflorescences can increase or decrease between years depending on growth conditions. Leaves are usually green though the winter. In cultivation and in the wild plants have been observed to flower in their second or third year. Although some plants in cultivation in Britain have lived for at least 25 years, wild plants may have shorter lives, e.g. in the Cork population the maximum life-span recorded was 13 years and many individuals survived for only 2-5 years (T. O’Mahony, pers. comm., 2000). 408 T. C. G. RICH AND C. R. BIRKINSHAW Large tussocks in well-lit sites may produce up to 200 inflorescences, each consisting of a terminal male spike and 1—4(—5) female spikes. Heavily shaded plants may not flower. Flowering begins in April and usually continues until June. The inflorescences are protandrous and wind- pollinated. Isolated plants have been observed to set fruit, and it is thus presumably self- compatible. Observations at Cambridge Botanic Gardens found an average of 4-96 + 0-8 (s.d.) nutlets per spike (n = 40), one of the lowest figures for a British sedge outside the Subgenus Primocarex. The nutlets are large (c. 4-5 x 2-1 mm) and heavy (average weight air-dry nutlets excluding utricles 0-0099 mg, n = 125). Fruit set is usually high; it sets abundant fertile seed in both extant British sites and many ripe fruits have been observed on herbarium specimens. Fruiting stems tend to droop around the plant late in the summer onto the soil, and may persist with fruits intact through the winter. It is usually possible to find at least some of the previous year’s inflorescences at the next flowering. The method of seed dispersal is unknown. Wind dispersal is unlikely because the nutlets are heavy and lack adaptations normally associated with this method of dispersal (e.g. wings). However, it is possible that the nutlets have evolved for dispersal by graminivores such as rodents (that could collect and cache the nutlets, and then sometimes fail to exploit these caches; for examples of species dispersed by rodents see Janzen 1983), or by herbivores that could swallow the nutlets when eating the foliage and at least sometimes pass them in a viable condition (for discussion of this method of dispersal see Janzen 1984). In addition, the nutlets will certainly sometimes be dispersed by the surface flow of water following rain. Failing all else, the long infructescence culm (up to 1-5 m) allows nutlets to be deposited some distance from the parent plant. Fresh nutlets may take a long time to germinate. Experiments showed samples took 18-29 months after sowing to germinate, suggesting an after-ripening requirement like many other sedges (Birkinshaw 1990; Jermy er al. 1982). Dormancy appears to be partly controlled by the pericarp; 30% germination (n = 20) occurred within 3—5 months when the pericarp was nicked. Germination is also almost certainly controlled by environmental conditions such as light. Seeds may remain dormant for long periods. This has been demonstrated experimentally by Birkinshaw (1999), who showed that at least some seeds in samples buried deeply in the soil germinated when brought to the surface after an interval of nine years. Observations of the Godalming site suggest dormancy of nearly 20 years between the last plant being seen and the new one re-appearing. HABITATS Birkinshaw (1991) summarised its habitat in Britain. [t usually occurs in dry, deciduous woodland, often in south-facing sites, but also occurs along the base of an old hedge in Somerset, and has been recorded from scrubby rocks by the sea on Anglesey. It is often associated with rocky woodland in Europe, but not exclusively so. Carex depauperata will prosper in conditions of full sunlight (provided that it is not out- competed by more vigorous heliophytes) to semi-shade. It grows well in full sunlight in the “Rare Plants Flower Bed” outside the English Nature Headquarters in Peterborough, and in gardens elsewhere. It is tolerant of shade, but plants may become moribund and cease to flower in deep shade. However, this sedge is probably best considered as a woodland gap plant adapted to the cycle of gap formation followed by secondary succession that is a feature of natural forests. In particular, it is hypothesised that the seeds of C. depauperata in the soil seed bank are stimulated to germinate by environmental changes associated with the formation of a gap in the forest; these seedlings grow rapidly to maturity in the conditions of high illumination and low competition within the gap and replenish the seed bank with new seeds; then as woody plants colonise the gap and close the forest canopy once again, the sedge declines and finally disappears leaving seeds waiting for the formation of a new gap, maybe several decades in the future. In support of this hypothesis is the long dormancy of C. depauperata seeds and various observations showing that the plant increases in abundance following coppicing or localised felling (woodland management operations that simulate gap formation; e.g. Lousley 1976; pers. comm., T. O’ Mahony 1989). At Fontainebleau, it was formerly plentiful in woods near the River Seine but the population declined to a handful of plants until coincidentally the French Forestry Department felled the Fagus trees and churned up the soil during forestry operations, after which hundreds of plants re-appeared (F. Rose, pers. comm., 1992). CONSERVATION OF CAREX DEPAUPERATA 409 The soils associated with C. depauperata range from acid podzols to calcareous brown earths (full soil analyses from five sites are given in Birkinshaw 1990). In Somerset it occurs on a calcareous brown earth (pH 7-4), in Surrey on weakly calcareous sandy soils, in Cork on a Mull rendzina (pH 6-2-6-6), at Fontainebleau, France it occurs on a podzol (pH 3-44-5), at Chantilly, France on sandy brown earth (pH 5-8—7-2) and at Ohrid, Yugoslavia on a clay-rich terra fusca. It is also reported from schistes in France. Soil type per se does therefore not seem to be critical provided they are free-draining (Birkinshaw 1991), though oddly Hegi (1980) also records it associated with damper soils with trickles of fresh water and notes a preference for nutrient-rich soils in central Europe. The populations at Axbridge, Godalming and Cork occur in rather patchy W8d Fraxinus excelsior - Acer campestre - Mercurialis perennis woodland Hedera helix sub-community of the national vegetation classification (Rodwell 1991). The site at Anglesey was probably in W24 Rubus fruticosus - Holcus lanatus underscrub. At Fontainebleau is occurs in managed beech woods equivalent to the NVC community W15 Fagus sylvatica - Deschampsia flexuosa woodland and at Chantilly in woodland equivalent to the W8 Fraxinus excelsior - Acer campestre - Mercurialis perennis woodland, Primula vulgaris - Glechoma hederacea sub-community. In general these equate to the order Fagetalia and it is typically recorded from such woodlands and in Quercus pubescens woodland of the order Quercetalia pubescenti-petraeae through most of Europe (Hess et al. 1967; Oberdorfer 1979; Hegi 1980; Ellenberg 1988). In Spain it occurs in Pinus pinaster woodland, Quercus pyrenaica woodland, Castanea woodland and Corylus avellana scrub (data from herbarium labels in MA). It is often associated with tracks ranging in size from small animal tracks to rough vehicle tracks (Birkinshaw 1991). This may be due to slightly enhanced light levels and its ability to tolerate crushing (it can tolerate even quite heavy trampling and even some crushing by vehicles) which competitive heliophytes may not be able to do. Observations at all three extant sites indicate grazing damage, especially to the inflorescences, presumably by rabbits, deer and/or horses. M. and C. Kitchen have observed deer eating plants in their garden in Gloucestershire (pers. comm., 1994). The exclosure experiment on the edge of Cheddar Wood showed plants subject to grazing by rabbits and deer were significantly less fecund than those protected from grazing (Birkinshaw 1990). It is tolerant of at least light grazing, but the number of inflorescences and consequently seed production may be significantly reduced. In Britain it is a lowland plant, being recorded from + sea level on Anglesey to c. 50 m at Axbridge and Godalming. In Europe it also occurs from near sea level in the Aegean (R. M. Harley, pers. comm., 1981) to at least 500-950 m in Spain and 1260 m in the mountains of central Europe (Hegi 1980). CONSERVATION Carex depauperata is a fully protected species under Schedule 8 of the Wildlife and Countryside Act 1981. The Cheddar Wood site is part of a S.S.S.I. but the Ockford Wood site is not. In Ireland it is protected under the Flora (Protection) Order, 1999. Plantlife monitor the species annually, and have prepared a species action plan (Davis 1999). Of the 14 sites from where it has disappeared or declined, there is no information on the causes of the decline for ten. The main cause of decline at three others appears to be lack of appropriate woodland management and increase in shade, with one recorded as probably being lost due to forestry operations. The main current threats to the plant are lack of appropriate management which may result in either dense shade or competition from other species, damage by vehicles, potentially too much grazing, and, at Godalming, landslides which have caused temporary extinction in the past. Whilst there are many specimens in herbaria, collecting by botanists has probably been only a very minor part of the decline, and has not been responsible for extinction at any sites as far as we are aware. Further visits to known historic sites may result in it being rediscovered, as at Godalming. Both native populations at Godalming and Axbridge have increased following conservation management of opening up the canopy and/or coppicing with disturbance of the ground flora. Such management, repeated perhaps every five years or so, is recommended for all sites, depending on 410 T. C. G. RICH AND C. R. BIRKINSHAW growth of the canopy and competing species. Where populations are critically small, more regular visits to clear competing vegetation around individual tussocks may be required and proved very successful at Godalming. Some protection from heavy grazing may be required at some sites. The potential for restoration of old sites in England from the seed bank has been investigated by McDonnell (1997), who concluded that there was insufficient detailed historic data to merit disturbance to regenerate it from the seedbank at any sites, and proposed reintroductions at Milton Wood, Godalming and Effingham. The recently rediscovered locality details suggests that restoration of the Anglesey site would be worthwhile. Material from Somerset, Surrey and Cork is held in cultivation in several botanic gardens and private gardens, and seeds from both English sites have been deposited in the Millennium Seed Bank. It grows readily in cultivation and increases from seed in at least some gardens, contrary to the statement by White (1912). ACKNOWLEDGMENTS We are grateful to David Allen, Ms J. R. Andrews, Richard Brinklow, Arthur Chater, Bob Corns, P. Cuccuini, Dick David, Mike Edgington, John Edmondson, Bob Ellis, Sally Ellis, Anne Fairbrother, Lynne Farrell, Ro Fitzgerald, Sam Hallett, Cif Hancock, Richard Handley, Stephen Hewitt, W. E. Hughes, George Hutchinson, Matthew Jebb, Andy Jones, Stephen Jury, Mark Kitchen, Gerald Legge, Peter Marren, Jim McGregor, Andy McVeigh, Douglas McKean, Rose Murphy, Henry Noltie, Donald Pigott, Francis Rose, Alan Showler, Joyce Smith, Peter Stovold, Sybil Tucker, Keith Watson, the Royal Botanic Gardens Wakehurst Place for help and information, and especially Tony O’ Mahony for details of the Cork population and Liz McDonnell for help at Axbridge. We would like to thanks the Keepers of the herbaria for access to specimens and libraries. The work was funded in various stages by the British Ecological Society, English Nature’s Species Recovery programme, Nature Conservancy Council, National Museum & Gallery of Wales and Plantlife. REFERENCES BIRKINSHAW, C. R. (1990). The ecology of Carex depauperata and its reinforcement at Cheddar Wood, Somerset. CSD Report, no. 1152. Nature Conservancy Council, Peterborough. BIRKINSHAW, C. R. (1991). The habitat preferences of Carex depauperata Curtis ex With.. Bulletin of the British Ecological Society 22: 26-31. BIRKINSHAW, C. R. (1999). Germination of buried Carex depauperata seed after nine years. BSBI News 82: 39-40. BLAB, J., NOWEK, E., TRAUTMANN, W. & SUKOopP, H. (1984). Rote liste der Gefahrdeten Tiere und Pflanzen in der Bundesrepublik Deutschland. Greven. DAVEY, F. H. (1909). Flora of Cornwall. F. Chegwidden, Penryn. DAVIES, E. W. (1956). Some new chromosome numbers in the Cyperaceae. Watsonia 3: 249-250. DAVIS, R. (1999). Species Action Plans for Plants. Starved wood-sedge. Unpublished working document, Plantlife, English Nature and World Wide Fund for Nature, London. DELVOSALLE, L., DERMET, F., LAMBINON, J. & LAWALREE, A. (1969). Plantes rares, disparues ou menacées de Disparition en Belgique: L’Appauvrissement de la Flora Indigéne. Ministére de 1’ Agriculture/Service de Réserves Naturelles Dominales et de la Conservation de la Nature. DRUCKER, G. R. F., WHITBREAD, A. & BARTON, J. (1988). Surrey inventory of ancient woodlands (Provisional). Nature Conservancy Council, Peterborough. EDGINGTON, M. (1991). Species recovery programme. Carex depauperata transplants - Cheddar Wood. Monitoring report 3 October 1991. Unpublished report, English Nature, Taunton. ELLENBERG, H. (1988). Vegetation ecology of central Europe, 4th ed. Cambridge University Press. Cambridge. GARDINER, W. (1848). The flora of Forfarshire. Longman, Brown and Green, London. HEGI, G. (1980). Ilustrierte flora von Mittel-europa 2(1): 230-231. Paul Parey, Berlin. HEss, H. E., LANDOLT, E. & HIRZEL, R. (1967). Flora der Schweiz. Birkhauser Verlag, Basel. HOOKER, W. J. (1821). Flora scotica. Archibald Constable & Co, Edinburgh. INGRAM, R. & NOLTIE, H. J. (1981). The flora of Angus (Forfar, v.c. 90). Dundee Museums and Art Galleries, Dundee. CONSERVATION OF CAREX DEPAUPERATA 411 JANZEN, D. H. (1983). Dispersal of seeds by vertebrate guts, in FUTUYMA, D. J. & SLATKIN, M., eds. Coevolution, pp. 232-262. Sinauer Associates Inc., Sunderland, Massachusetts. JANZEN, D. H. (1984). Dispersal of small seeds by big herbivores: foliage as fruit. The American naturalist 123: 338-353. LANDOLT, E. (1989). Plantes vasculaires menacées en Suisse (listes rouges fédérales et régionales). Office Fédéral de 1’Environne, des Foréts et du Paysage, Berne. LOUSLEY, J. E. (1976). Flora of Surrey. David & Charles, Newton Abbot. MCDONNELL, E. J. (1997). Carex depauperata (Starved wood-sedge) 1997 report. Back from the brink project report no. 99. Plantlife, London. NILSSON, O. (1985). Carex L., in DAVIS, P. H., ed. Flora of Turkey and the eastern Aegean Islands. p. 124. Edinburgh University Press, Edinburgh. OBERDORFER, E. (1979). Pflanzensoziologische Exkursionsflora. Eugen Ulmer, Stuttgart. O’MAHONY, T. (1976). Carex depauperata Curt. in NE Cork (H5) a sedge new to Ireland. /rish naturalists’ journal 18: 296-298. O’ MAHOoNy, T. (1999). A report on the flora of Co. Cork (v.cc. H3-—5), 1998. Irish botanical news 9: 28-32. PERRING, F. H. & WALTERS, S. M. (1990). Atlas of the British flora, 3rd ed. BSBI, London. REEVES, W. W. (1874). Carex depauperata, Good. Journal of botany 12: 205. RICH, T. C. G. (1992). Starved Wood-sedge (Carex depauperata) at Godalming. Back from the brink project report no. 1. December 1992. Plantlife, London RICH, T. C. G. (1994a). Starved Wood-sedge (Carex depauperata) at Godalming: Monitoring report for 1993—- 1994. Back from the brink project report no. 26. February 1994. Plantlife, London RICH, T. C. G. (1994b). The rediscovery of starved wood-sedge Carex depauperata Curtis ex With. at Godalming, Surrey after the Great Storm of 1987, in KIRBY, K. J. & BUCKLEY, G. P., eds. Ecological responses to the 1987 Great Storm in the woods of south-east England. pp. 105-108. English Nature Science no. 23. English Nature, Peterborough. RICH, T. C. G. & FAIRBROTHER, A. (1994). The status of starved wood-sedge (Carex depauperata) in Britain in 1994. Back from the brink project report no. 39. December 1994. Plantlife, London. RICH, T. C. G. & FAIRBROTHER, A. (1995). How starved wood-sedge (Carex depauperata) got its name? Back from the brink project report no. 57. November 1995. Plantlife, London. RICH, T. C. G. & FAIRBROTHER, A. (1996). Starved wood-sedge (Carex depauperata) in 1996. Back from the brink project report no. 80. November 1996. Plantlife, London. RICH, T. C. G., JONES, R. A. & JEBB, M. (2000). Three new sites for Carex depauperata With. (Cyperaceae) represented in the Irish National Herbarium, Glasnevin. Watsonia 23: 340-341. RODWELL, J. S., ed. (1991). British Plant Communities, Volume 1. Woodlands and scrub. Cambridge University Press, Cambridge. WHITE, J. W. (1912). The flora of Bristol. John Wright and Sons, Bristol. (Accepted June 2000) * po Watsonia 23: 413-419 (2000) 413 A morphological comparison between some British Orobanche species (Orobanchaceae) and their closely-related, non-British counterparts from continental Europe: Orobanche rapum-genistae Thuill. s.1. MEN Ye FOLEY Department of Biological Sciences, Institute of Environmental and Natural Sciences, University of Lancaster, Lancaster, LAl 4YQ ABSTRACT A morphological comparison has been made between British plants of O. rapum-genistae and plants referable to subsp. rigens from Corsica, which is endemic to that part of the Mediterranean. Characters which best separate the two taxa are indicated and it is concluded that the latter is more appropriately placed at species level as O. rigens Loisel. KEYWORDS: Orobanche rapum-genistae subsp. rigens, Corsica, taxonomy. INTRODUCTION BACKGROUND Orobanche rapum-genistae sensu lato comprises a small but widespread group of western European taxa thought to be distributed eastwards as far as Germany and Italy (limit at Val Vestino, south Tirol (Gilli 1965)) and southwards to Spain and Portugal (Beck 1926-28); it is also recorded for Algeria (Battandier 1890) and other north African countries, although its presence in some may be doubtful. The current northern limit is in Scotland. Fournier (1937) recognised three taxa at subspecific level. These were the type (subsp. rapum- genistae), subsp. benthamii (Timb.-Lagr.) P. Fourn., differing from the former by its extremely long bracts and a distinctly bilobed upper lip to the corolla, which had been recorded from Spain, Portugal and northern Italy, and subsp. rigens (Loisel.) P. Fourn., apparently similar to the type but almost glabrous, from Corsica and Sardinia and the associated small off-shore islands. It is not definitely known whether the distributional ranges of the latter two taxa overlap with subsp. rapum-genistae. More recent treatments (Chater & Webb 1972; Pignatti 1982) have supported Fournier's approach. The host plants are usually leguminous shrubs (Fabaceae), especially the genera Cytisus, Ulex and Genista, but with subsp. rigens apparently restricted to species of Genista (Chater & Webb 1972; Camarda 1981; Pignatti 1982). Throughout most of its European range O. rapum-genistae is represented by subsp. rapum- genistae. In the British Isles this occurs southwards from southern Scotland and is widespread but local, having substantially declined during the past century (Foley 1994) and receded into a predominantly south-westerly distribution pattern, but with substantial populations remaining in Wales; it also occurs in south-east Ireland. In the British Isles the host is invariably either Cytisus scoparius (L.) Link or Ulex europaeus L., although there is an old record from Kent on Genista tinctoria L. The procumbent, maritime subspecies of C. scoparius (subsp. maritimus) is parasitised on sea cliffs in the Channel Islands (Rumsey & Jury 1991). In France, from where the type (O. rapum-genistae Thuill.) was originally described, it is widely distributed up to subalpine levels (Rouy & Foucaud 1909) and records for Switzerland are also mainly from montane and (rarely) subalpine localities (Hess et al. 1972). In Germany it is restricted to the west, known from the Rheinland and Schwarzwald eastwards to Westphalia (Gilli 1965); it is also present (as subsp. rapum-genistae) in Belgium and Holland (Chater & Webb 414 WE... ¥; FOLEY. 1972). In Italy, Pignatti (1982) gave all three subspecies: the type, widespread throughout; subsp. benthamii from the alpine foothills of Piedmont and Lombardy; and subsp. rigens from Corsica, Sardinia and Sicily. Also for Corsica, Gamisans & Jeanmonod (1993) recorded all three taxa but the presence of two of them (the type and subsp. benthamii) is doubtful, whilst Camarda (1981) also gave localities for subsp. rigens (as O. rigens) in Sardinia. Both the type and subsp. benthamii are known from the Iberian peninsula (Willkomm 1893; Merino 1906; Coutinho 1913; Cadevall 1932; de Lima 1947; Franco 1984). For Algeria only the type was recorded by Battandier (1890) and Quezel & Santa (1963) although there is a record for subsp. benthamii by Beck (1930). O. rapum-genistae Ss. l. is apparently present in Morocco (Greuter et al. 1989) but doubtful in Tunisia (Bonnet & Barratte 1896). BASIC TAXA RELATED TO OQ. RAPUM-GENISTAE AND SUBSEQUENT TREATMENTS O. rapum-genistae was described from near Paris, France (Thuillier 1799) where it parasitised Cytisus scoparius and flowered in June: “caule crasso, conspicue pubente et angulato: spica longissima: confertiflora, bracteis calycibusque villosissimis: corollis rufescentibus, brevi- tubulatis’. The description is barely adequate, although the reddish (but shortly tubular) corollas give some indication of its appearance. The type specimen is possibly in LE. O. rigens was first described from Corsica (Loiseleur 1807) from specimens seen in Richard’s herbarium “caule simplicissimo glabro, squamis lanceolatis rigentibus imbricatis, corollis quadrifis, staminibus inferne nudis, stigmate 2-lobo, lobis distantibus, stylo glaberrimo, simillima praedenti [O. major L.], sed discrepare videtur squamis rigentibus subpungentibus, et glabritie omnium parte, exceptis bractaeis quae leviter pubescunt. Flores rubiginosi ut planta ipsa’. According to this description, one of the main characters relates to the more or less glabrous nature of the stem, style and lower part of the filaments - and indeed of most parts other than the lightly pubescent bracts. The imbricate, lanceolate scale-leaves are rigid, the stigma lobes apparently well separated and the flowers and the whole plant are of a reddish coloration. No specimen was mentioned by Loiseleur but his description indicates that it was based on Corsican material collected by Richard and retained in the latter’s herbarium; however such potential type material appears now to be lost or destroyed. Loiseleur’s herbarium at Avignon (AV) also contains no relevant material collected by Richard (Camarda 1981) but the Paris herbarium (P) has a Corsican specimen gathered by Soleirol in 1826 and presumably studied by Loiseleur “au mont d’Oro prés des neiges eternelles”. This has been proposed as the neotype for O. rigens (Camarda 1981). O. benthamii Timb.-Lagr. was based upon the earlier O. rapum var. bracteosa which was described by Reuter (1847) as “spicd valde elongata, bracteis flores superantibus apice comos@”’. Reuter’s diagnosis referred to plants collected by Bentham at Collioure in southern France and by Moris in the subalpine region of Piedmont, Italy. There is a fuller description of the same plant collected at les Basses Corbiéres hills, near Durban, southern France as O. benthamii (Timbal- Lagrave 1874). Again the bract character is referred to: “bractées.....dépassant de beaucoup les fleurs, ce qui rend l’inflorescence chevelue” - i.e. making the inflorescence “hairy”. The corolla is stated to be very large, bright red - is the stem - and with lobes with fimbriate teeth. The filaments are glabrous and inserted at the base of the corolla and the stigma is white or yellow. Beck (1930) placed all these taxa in his Grex Arcuatae, members of which were characterised as follows. The corollas were broad and more or less ventricose above the point of filament attachment and of similar internal and external colour, their dorsal line was strongly curved over the whole length, the margins of the corolla lobes were glandular-ciliate, the filaments inserted very low near the corolla base and the stigma lobes yellow. Within the Arcuatae he recognised three species: | O. rapum-genistae, O. rigens and O. anatolica Reuter (1847) - the last is non- European, being from the Caucasus and Iran. As regards O. rapum-genistae itself, Beck (1930) recognised two varieties in addition to the type. One of these was var. bracteosa (= O. benthamii Timb.-Lagr.) described above, the other, var. insolita, was based upon Guimaraes’ description of O. insolita from Bussaco, Portugal (Guimaraes 1904) where it apparently parasitised Eryngium campestre. This is the usual host for O. amethystea and it is possible that O. insolita may be a variant of this, although filament, bract and inflorescence characters and the presence of abundant white, glandular hairs (Guimaraes 1904) are consistent with O. rapum-genistae s. l. Unfortunately, no material of O. insolita that is known to have been seen by Guimaraes has been located. OROBANCHE RAPUM-GENISTAE 415 TABLE 1. COMPARISON OF MORPHOLOGICAL CHARACTERS OF POPULATIONS OF OROBANCHE RIGENS AND O. RAPUM-GENISTAE CORSICA Locality Tassenita valley Asco Number 10 Date 10.6.1997 Species rigens Grid Ref./Coordinates 42 26'N; 8 58'E Plant height 213-5 + 16-7 Stem width 10-1 + 1-0 pubescence 0-8 +0 colour pink-mauve - yellow Inflorescence length 133-0 + 9-8 density jy c= 3) 05 Leaf width 6:9 + 0-4 pubescence 0-8+0 shape ovate-tapering colour pink-slt. mauve Bract length 16-7 + 0-7 shape ovate-lanceolate pubescence 1-5+0-1 colour darkish mauve Corolla length 23:2 + 0-6 height 9-1+0-3 approximate shape subcampanulate attitude 1-1+0-1 dorsal line 2:5+0-1 pubescence 1-:0+0 colour pink-mauve Upper lip divided 0-5 + 0-2 Lower lip ciliate 0-2+0 denticulate 0 shape of lobes 0 relative size 10+0 (central to lateral) Bracteole 0 Calyx length 13-6 + 0-3 segment division 1-8 + 0-4 pubescence 14+0-1 shape of teeth elongated colour mauve Filaments insertion 2:9+0-1 height pilosity above 0 pilosity below 0 Anthers pilose 0 Stigma lobes separation Daly) colour pale yellow Apparent Genista host salzmannii var. lobelioides CORSICA BRITAIN BRITAIN BRITAIN Haut Asco Haydon Bridge Cyttir Mawr Amisfield Asco Northumberland Anglesey Dumfries i 10 10 10 10.6.1997 126.1993 5.61993 12 & 19.6.1993 rigens rapum-genistae rapum-genistae rapum-genistae 42 26'N; 8 58'E NY8565 SH5774 NY0083 165 £13-9 415-5 + 18-7 360 + 14-9 3799+ 20:5 8-9 + 0-5 10-8 + 0-3 11-6+0-3 13-14 0-8 0-9+0 Ian 1:7+0-1 1-8+0-1 pink. pink-yellow, pink-yellow __ pink-pale brown _ pale brown-pink yellow 105-7 + 10-1 224-8 + 18-0 195-7 + 15-8 175-0 + 14-8 1-:8+0-1 1-3+0-1 1:9+0-1 1:-6+0-1 6-8 + 0-4 6-7 + 0-2 6-7 + 0-2 8-0 + 0-4 0:9+0 0-9 + 0-1 1-0+0 2 20:1 ovate-lanceolate linear-lanceolate triangular- tapering lanceolate pink-yellow brown medium brown fawn-dark brown 18-4 + 0-8 27-6 + 0-5 21:9: 11 25-1 + 1-0 ovate-lanceolate lanceolate lanceolate lanceolate 14+0-1 2:0+0 1-6+0-1 1:8+0-1 pink-mauve pale brown pink-brown dark brown 21-6 + 0-6 24-4 + 0-3 222205 25:6 + 0-6 7-71-05 10-0 + 0-3 10-4 + 0-4 10-3 + 0-2 subcampanulate + infundibuliform + infundibuliform + infundibuliform 1:0+0 1-3 +0°1 1-6+0-1 1-4+0-1 24+0-1 2:2 + 0-1 29 + 0-1 2-6 + 0-1 0-8 + 0-1 1540-1 1:2+0-1 1-4+0-1 pink-mauve, pink-fawn pink-pale fawn pink-fawn pink-yellow 0 0 0-1+0-1 0 0-1+0-1 10+0 0-3 + 0-1 0-7+0:1 0 0 0 0 0 0-8 + 0- 0-8 + 0-1 0-8 + 0-1 10+0 1-4 + 0- 1-3+0-1 1:3+0-1 0 0 0 0 13-0 + 0-9 15-0 + 0-5 13-0 + 0-4 15-8 + 0-6 1040 1:0+0 1-4+0-2 12+0-1 1-2+0-1 2:0+0 1:2+0-1 1:7+0-1 triangular- tapering + acuminate tapering lanceolate mauve-pink pale fawn pale pink-fawn pale fawn 26+ 0-1 3-4+0-2 2:0+0:1 3-8 + 0-2 0 Oe 0 0-9 + 0-1 1-0+0 0 0 0 0 0 0 0 0-1+0 2:0+0 3-022,0 2:9+0-1 2:9 +0-1 pale yellow bright yellow bright yellow bright yellow Genista salzmannii var. lobelioides Cytisus scoparius Ulex europaeus Cytisus scoparius Mean values and standard errors (in mm) 416 M. J. Y¥. FOLEY Beck (1930) maintained O. rigens as a separate species from O. rapum-genistae on the basis of the former’s more or less glabrous corolla, filaments and style. In contrast, Fournier (1937) in his French Flora placed all the principal European Arcuatae taxa at subspecific level under O. rapum- genistae. Such a treatment has been maintained by several later workers (Chater & Webb 1972: Pignatti 1982) but modified by others. For example O. rigens was ranked at species level by both Camarda (1981) and Uhlich ef al. (1995) despite having much earlier been considered a variety by Fiori (1926). In other instances some of the taxa have been simply omitted or ignored completely. METHODS Three British populations of subsp. rapum-genistae (two parasitic upon Cyfisus scoparius and one upon Ulex europaeus) were morphometrically compared with two Corsican populations of subsp. rigens parasitic upon Genista salzmannii var. lobelioides (see Table 1 for locality details). Characters measured or assessed and the methods used were as described in a prior paper (Foley 2000). Mean values and standard errors of the relevant characters are summarised in Table 1 and the quantifiable numerical elements of these data have been subjected to PCA (Figure 1) in a similar manner to that previously decribed (Foley op. cit.). By this technique, each specimen measured is located in a multidimensional array, in which the number of dimensions is equal to the number of characters measured, and where the most similar specimens are placed closest together. This technique extracts the axes of greatest variation from the multidimensional space, thereby simplifying the space to a few (usually 2 or 3) dimensions, so allowing the location of each specimen to be visualised. RESULTS AND DISCUSSUION In Figure 1 it can be seen that the Corsican plants are clearly separated from the British on the PC1 axis. Within the British plants there is only slight overlap between plants at Cyttir Mawr, Anglesey (on Ulex europaeus) and those of the other two populations on Cytisus scoparius (at Amisfield and Haydon Bridge). Whether this difference is significant, possibly due to the influence of the different hosts on the plant’s morphology, is open to question but this could additionally be investigated by cultivation experiments. When the characters of the two taxa (subsp. rapum-genistae and subsp. rigens) are compared (Table 1), it can be seen where the principal differences lie. It is true that there is quite a close similarity in many of them but relatively intangible morphological differences, readily apparent in living plants in the field, are reflected in at least some of these. Plants of subsp. rigens appear to be less robust than those of subsp. rapum-genistae (overall height, inflorescence length, stem width etc.) - but this may in part be due to their occurrence at a higher average altitude. In many plants the inflorescence occupies a higher proportion of the overall height of the plant and often reaches close to ground level. Leaf and bract shape is distinctly ovate (rather than lanceolate) and the corolla is subcampanulate rather than ventricose- infundibuliform as in subsp. rapum-genistae (the latter is also somewhat carinate dorsally), with the lobes of the lower lip much more rounded and of equal size. Corolla colour (and general plant coloration - stem, leaves, calyces) differs, being markedly mauve-pink in subsp. rigens rather than pink-fawn. Perhaps the clearest character difference lies in filament pubescence. Whilst the filaments are inserted at similar heights, they are glabrous throughout their entire length in subsp. rigens whereas those of subsp. rapum-genistae are very distinctly glandular-pilose in their upper portion. The style and ovary are also distinctly pilose, being provided with colourless, few-celled, glandular hairs; in contrast subsp. rigens is glabrous in this respect. Stigma lobe colour is also much paler in this taxon and the stigma lobes are noticeably less separated. Another character difference which is very apparent is the overall level of pubescence, this being very much reduced in subsp. rigens compared to O. rapum-genistae. Since subsp. rigens appears to be a much more montane-subalpine plant than subsp. rapum-genistae, it might be expected that if the former were.merely a close variant of the latter then the pubescence, as in other montane plants, would be much greater (not less) in the higher altitude taxon. OROBANCHE RAPUM-GENISTAE 417 6 5 -4 3 2 = 0 1 2 3 4 5 6 . rigens (Tassineta, Stranciacone, Massif du Cintu, Corsica) . rigens (Haut Asco, Massif du Cintu, Corsica) . rapum-genistae (Amisfield, Dumfries-shire, Scotland) . rapum-genistae (Cyttir Mawr, Anglesey, Wales) . rapum-genistae (Haydon Bridge, Northumberland, England) Qoeta S49 S495 FIGURE 1. First two axes of Principal Components Analysis for individual plants of Orobanche rigens and O. rapum-genistae. PC1 & PC2 jointly account for 43-6% of the total variance. CONCLUSIONS It is clear that whilst both subsp. rapum-genistae and subsp. rigens are quite similar, they also possess several distinctive characters. In contrast to subsp. rapum-genistae, the following characters are diagnostic for subsp. rigens: the very low level of overall pubescence, the differering corolla shape and coloration, the more ovate leaf and bract shape, the glabrous filaments and the differing stigma lobe characters. In the absence of evidence of intermediates between the two, it is concluded that these differences are sufficient for subsp. rigens to be separated at species level as Orobanche rigens Loisel. Other work, based mainly upon herbarium specimens of Sardinian O. rigens (Camarda 1981), reached a similar conclusion; both those findings and the present ones contrast with the approach taken by most other workers subsequent to Fournier (1937), who have maintained O. rigens at subspecific level. The situation regarding subsp. benthamii is less easy to resolve as it is a little-known taxon of limited and relatively unknown distribution; it is not recorded for Britain. During this work no living plants and only three herbarium specimens (out of many hundreds of this group examined) could reasonably be referred to this taxon and even those differed from subsp. rapum-genistae only in their significantly longer bracts. Other differences indicated in Chater & Webb (1972), e.g. an erect, distinctly bilobed upper corolla lip in subsp. benthamii, were not evident in these specimens. It is concluded that the latter should be considered to be merely a form of O. rapum- genistae. 418 M. J. Y. FOLEY TAXONOMIC DESCRIPTION OROBANCHE RAPUM-GENISTAE THUILL. (syn.: O. major L. pro parte; O. palatina Schultz; O. rapum var. bracteosa Reuter; O. bracteosa (Reuter) Nyman; O. elatior Horn; O. foetida Duby, O. sarothamnophyta St Lager, O. benthamii Timb.-Lagr., O. rapum-genistae subsp. benthamii (Timb.-Lagr.) P. Fourn.) Described from near Paris, France. Type possibly in LE. Typically 250-450 mm tall, with a quite dense, long, many-flowered inflorescence. Stem: 8—15 mm wide immediately below the inflorescence, distinctly glandular-pubescent, pink-yellow to pink-brown. Leaves: maximum width 6-8 mm, + lanceolate, fawn-brown. Bracts: 20-28 mm long, linear-lanceolate, light brown, distinctly glandular-pubescent. Calyx: 13-16 mm long, segments fairly equally divided, tapering, pale pink-fawn, glandular-pubescent. Corolla: 22-26 mm long, sub-erect, + infundibuliform, ventricose, somewhat carinate dorsally, pink-fawn, glandular- pubescent, dorsal line distinctly curved; upper lip + entire, margins of lower lip somewhat ciliate, not denticulate, rather acute, the central lobe often larger than the laterals. Filaments: inserted 24 mm above the corolla base, glabrous below, distinctly glandular-hairy above. Stigma lobes: deep yellow, well separated. Parasitic mainly upon species of Ulex and Cyfisus. This is a plant of scrub-invaded areas such as rocky hillocks, field margins and headlands, roadside verges and other habitats typical of its hosts. Within Britain, observations have suggested a possible west-east cline in host preference with Ulex europeaus apparently being preferred in the west (especially Wales) and Cytisus scoparius the preferred host further east. Flowers April—July. The species is apparently widespread but often very local in western Europe extending north to southern Scotland and east to the Italian south Tirol. The long-bracted form (syn.: O. rapum-genistae var. bracteosa Reuter; O. benthamii Timb.- Lagr.) appears to differ from the above only in its exceptionally long bracts. Suggested differences in corolla lip characters have not been confirmed. It is apparently of very limited occurrence and recorded only very locally in southern France, Italy, parts of the western Mediterranean and the Iberian peninsula. OROBANCHE RIGENS LOISEL. (syn.: O. condensata Moris pro parte; O. thyrsoidea Moris; O. rapum var. glabrescens Gren. & Godr., O. rapum-genistae subsp. rigens (Loisel.) P. Fourn.) Described from Corsica. Type thought to be lost (Richard herb.), neotype: “au mont d’Oro [Corsica] prés des neiges eternelles, 1826, leg. Soleirol” (P!). Typically 150-230 mm tall, with a fairly dense, many-flowered inflorescence which sometimes extends to most of the length of the spike. Stem: 8-10 mm wide immediately below the inflorescence, slightly glandular-pubescent to nearly glabrous, pink-yellow to somewhat mauve. Leaves: maximum width 6—7 mm, usually + ovate, pink-yellow to pink-mauve. Bracts: 16-20 mm long, + ovate, pink-mauve, somewhat pubescent. Calyx: 13-14 mm long, segments fairly equally divided, teeth elongated, triangular-lanceolate, mauve-pink, somewhat glandular-pubescent. Corolla: 21-24 mm long, erect to suberect, + subcampanulate, not noticeably carinate dorsally, pink-mauve sometimes tinged yellowish, only slightly glandular-pubescent, dorsal line distinctly curved. Upper lip sometimes noticeably divided, margins of lower lip sometimes slightly ciliate, not denticulate, rounded, lobes of + equal size. Filaments: inserted 2—3 mm above the corolla base, glabrous throughout their length. Stigma lobes: pale yellow, only moderately separated. Parasitic upon species of Genista, in Corsica usually only on Genista salzmannii DC. var. lobelioides (or occasionally var. salzmannii) and in Sardinia on Genista corsica. There is a suspicion that Anthyllis hermanniae may also be parasitised (D. Jeanmonod, pers. comm. 1996). In Corsica, this is a plant of montane-subalpine habitats mainly found in the Thymus-Genistetum association of dwarf spiny shrubs at an altitudinal range of (600—) 900—c.2000 m, but in Sardinia it occurs down to sea-level (Camarda 1981). Flowers (March—) May—June (—July at high altitudes). It is endemic to Corsica and Sardinia and the associated off-shore islands. It is widespread but local in suitable habitats but may be under-recorded. Records for Sicily are probably errors for other taxa. Although both O. rapum-genistae and its long-bracted form (var. bractosa) have been recorded for Corsica, it is possible that these are errors and that O. rigens is the only member of the group present. OROBANCHE RAPUM-GENISTAE 419 ACKNOWLEDGMENTS I am most grateful to the trustees of the Welch Bequest Fund of the Botanical Society of the British Isles and to the Botanical Research Fund (Kew) for financial assistance during this project. I am also grateful to D. Jeanmonod (Geneva) and I. Guyot (Bastia) for providing me with information regarding Corsican localities for O. rigens, and to all those vice-county recorders who have similarly helped with information on O. rapum-genistae in Britain. I also wish to thank the curators of the following herbaria for their loan of specimens: B, BM, E, LIV, MA, P, SEV. REFERENCES BATTANDIER, J-A. (1890). Orobanche in BATTANDIER, J-A. & TRABUT, L. Flore de l’Algérie, 4: 659-664. F. Savy, Alger & Adolphe Jourdan, Paris. BECK, G. (1926-28). Orobanche in Die Pflanzenareale, | Reihe, Heft 7, Karte 64. Gustav Fischer, Jena. BECK, G. (1930). Orobanchaceae, in ENGLER, A., ed. Das Pflanzenreich IV. 261: 1-348. Willhelm Engelmann, Berlin. BONNET, E. & BARRATTE, G. (1896). Exploration scientifique de la Tunisie. Imprimerie nationale, Paris. CADEVALL, J (1932). Flora de Catalunya 4: 301. Palau de Generalitat, Barcelona. CAMARDA, I. (1981). Le piante endemiche della Sardegna, 106-109. Bollettino della Societa Sarda di Scienze Naturali 21: 373-382. CHATER, A. O. & WEBB, D. A. (1972), Orobanche in Tutin, T. G. et al., eds. Flora Europaea 3: 286-293. Cambridge University Press, Cambridge. COUTINHO, A. X. P. (1913). Flora de Portugal, 565-571. Aillaud, Lisboa. DE LIMA, A.P. (1947). Flora Portuguesa, 496-500. Imprensa Moderna, Porto. Fiori, A. (1969). Nuova flora analitica d'Italia, 2: 389. Edizioni Agricole, Bologna. FOLEY, M. J. Y. (1994). O. rapum-genistae, in STEWART, A. et al., eds. Scarce Plants in Britain. Peterborough. FOLEY, M. J. Y. (2000). A morphological comparison between some British Orobanche species (Orobanchaceae) and their closely-related non-British counterparts from continental Europe: Orobanche reticulata Wallr. s. 1. Watsonia 23: 257-267. FOURNIER, P. (1937). Les Quatre Flores de la France, 796. Paul Lechevalier, Paris. FRANCO, J. do’A. (1984). Nova flora de Portugal, 2. Sociedade Astoria, Lisboa. GAMISANS, J & JEANMONOD, D. (1993). Catalogue des plantes vasculaires de la Corse, 11-258. Conservatoire et Jardins botaniques de la ville de Geneve, Geneva. GILLI, A. (1965). Orobanche, in HEGI, G., ed. Illustrierte Flora von Mittel-Europa, 6(1): 470-505. Carl Hanser Verlag. Miinchen. GREUTER, W, BURDET, H. M. & LONG, G. (1989). Med-Checklist, 4: 256-265. Conservatoire et Jardins botaniques de la ville de Geneve. Geneva. GUIMARAES, J. d’A. (1904). Monographia das Orobanchaceas Portuguezas. Broteria, 3: 5-207, Lisboa. HEss, H. E., LANDHOLT, E. & HIRZEL, R. (1972). Flora der Schweiz und angrenzender Gebiete. Bd. 3: 250— 260. Birkhatiser Verlag, Basel & Stuttgart. LOISELEUR, J. L. A. (1807). Flora gallica, 2: 384-385. Lutetiae, Paris. MERINO, R. P. B. (1906). Flora Descriptiva é Ilustrada de Galicia, 3: 42-52. Tipografia Galaica, Santiago. PIGNATTI, S. (1982). Flora d'Italia, 1-3. Edagricole. Bologna. QUEZEL, P. & SANTA, S. (1963). Nouvelle Flore de l’Algérie et des Régions Désertiques Méridionales, 2: 852-859. Centre national de la Recherche Scientifique, Paris. REUTER, G. F. (1847). Orobanchacaeae in DE CANDOLLE, A. P., ed. Prodromus Systematis Naturalis Regni vegetabilis 11: 1-45. Victoris Masson, Paris. ; Rouy, G. & FOUCAUD, J. (1909). Flore de France, XI. Les Fils d’Emile Deyrolle, Paris. RUMSEY, F. J. & Jury, S. L. (1991). An account of Orobanche L. in Britain and Ireland. Watsonia 18: 257- Day, THUILLIER, J. L. (1799). Flore des environs de Paris, 2nd ed., 317. Gazette nationale, Paris. TIMBAL-LAGRAVE, P. M. E. (1874). Mémoires. Academie des sciences, inscriptions et belles-lettres de Toulouse, 7(6): 643-644. UHLICH, H., PUSCH, J. & BARTHEL, K-J. (1995). Die Sommerwurzarten Europas: Gattung Orobanche, 1-235. Die Neue Brehm-Bucherei Bd. 618, Westaerp Wissenschaften. Magdeburg. WILKOMM, M. (1893). Supplementum Prodromi Florae Hispanicae, 185-191. Sumptibus E. Schweizerbart, Stuttgart. (Accepted July 1999) . rT. ye Watsonia 23: 421-435 (2001) 42] Brambles (Rubus L. sect. Rubus and sect. Corylifolii Lindley, Rosaceae) of the Channel Islands Dees ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ ABSTRACT The history of the investigation of the predominantly apomictic genus Rubus L. in the Channel Islands (Les Iles Anglo-Normandes) is outlined and the records made to date are listed and critically assessed. Of species currently admitted to the British Isles list, 39 inclusive of two naturalised horticultural taxa are considered reliably recorded: 26 from Jersey, 23 from Guernsey, 14 from Sark and 11 from Alderney. Though these totals roughly correlate with island size, they are each made up of different species in substantial part. They include one apparent endemic, R. cordatifolius (Rogers ex Riddelsd.) D. Allen, shared by Guernsey and Sark. Many unnamed entities, however, also occur, one or two in each main island relatively widely, but the group has been too little investigated on the French mainland to tell how far any extend there sufficiently to merit being described should they not prove to belong to existing taxa. Dispersal and other factors determining the presence or influencing the frequency of certain species in the islands are also examined. KEYWORDS: apomictic species, Britain, France, Normandy, Brittany, distribution, bird dispersal. INTRODUCTION By virtue of their being Crown possessions in British home waters the Channel Islands - Les Iles Anglo-Normandes to the French (to whom one group of islets belong) — have traditionally been treated as part of their territorial responsibility by students of the British-cum-Irish flora. From the strictly geographical point of view, though, it is to Continental Europe that they manifestly belong, as confirmed by the Precambrian geology they share with Brittany and the nearest part of Normandy as well as by a general floristic character broadly different from that on the other side of the English Channel. That very degree of difference has tended to deter specialists in major critical groups such as Rubus, who have been conscious of the extra dimension of difficulty these islands are likely to present without some reasonable measure of familiarity with the representatives of those groups in the nearby parts of the mainland. The Rubus flora is particularly richly diverse, as it chances, in the Cotentin Peninsula of Normandy immediately to the east (Allen 1996) and only small portions of that, or of the whole of Brittany for that matter, have been investigated even cursorily by batologists until the last twenty years. Though knowledge of that wider context is still very incomplete, the islands themselves have by now been sufficiently well worked to justify putting together up-to-date lists of the taxa that seem to have been reliably recorded from each of the largest and to examine how far the presence or absence of some of the members of this predominantly apomictic group has extra insights to provide about the floristic affinities of the individual main islands — especially to one another — and of the archipelago as a whole. PREVIOUS WORK Study of the islands’ Rubus flora dates from Babington’s pioneer exploration of 1837-8. Though that was before he had begun to specialise in the group, by then he was already sufficiently intrigued by its complexity to collect the odd specimen and seek the assistance of Borrer in naming them. Not surprisingly, though, given the primitive state of Rubus taxonomy at that period, their joint efforts proved on the whole inconclusive (Babington 1839); disappointingly, too, not all of the specimens appear to have survived. There is equally slight surviving evidence in herbaria of a 422 D. E. ALLEN visit made by H. C. Watson to Guernsey in 1852 and of one by A. Ley to both there and Jersey in 1885, while two species listed by Marquand (1892) from Guernsey were under names that cannot be referred with confidence to any recognised today. The islands were thus effectively still virgin territory batologically when W. Moyle Rogers, the then leading British specialist, devoted a visit of almost a fortnight in 1897 expressly to closing what had come to seem a glaring gap in his knowledge of the distribution in the British Isles of the various taxa he was about to recognise in his forthcoming revision of the group (Rogers 1900). Accompanied by his son, F. A. Rogers, who did some collecting as well, he concentrated on Jersey and Guernsey, forgoing Alderney and managing a few hours only on Sark. In the subsequent published reports (Rogers & Rogers 1898; Derrick 1898; Rogers 1899) he was to claim, with evident relief, that the vast majority” of the forms encountered were “practically identical” with ones with which he was familiar in Britain. The specimens collected, however, now mostly in BM, provide notably less support for that conclusion than he had persuaded himself was the case, for those that are of entities to which a name cannot be put today almost match in number the 16 taxa in his list that he proves to have identified with British ones correctly. Had he written, rather, that most of the widespread taxa encountered were also British, that would have been near the truth — and is perhaps what he intended to convey. Considering how many more of Britain’s more widespread Rubus entities had yet to be discriminated and describea at that period, the extent to which Rogers mastered the group in the islands on that single visit is nonetheless impressive. Unfortunately, for half a century after his death there were not to be any further Rubus specialists, British or French, with direct experience of the islands and attempts at determining material from there had a poor record of success by comparison. When W. C. R. Watson was invited in 1930 to pronounce upon random specimens G. C. Druce had collected on various visits, his touch was so uncertain that not only are one or two of his determinations patently wide of the mark, but in two cases he changed his mind in later years - but not for the better. He would really have been wiser not to have hazarded names at all, for Druce’s material consists in the main of scarcely enlightening scraps. The mere handful now to be found in OXF are evidently by no means the full number that Druce amassed, for some records of his based at least partly on determinations by Rogers were published by him at an earlier date (Druce 1907). Most of Druce’s specimens are undated, but one from Jersey in 1923 shows that his collecting just overlapped with the period of activity there of Fr. Louis-Arséne, whose herbarium, now at JSY, contains a number of Rubus species allegedly from various named localities in that island. As shown by Le Sueur (1982), however, the stated provenance of all Louis-Arséne’s botanical specimens is suspect and it would be unsafe to do other than ignore the examples of Rubus in their entirety. RECENT WORK Around 1970 a limited resumption of collecting took place in connection with the new Flora of Guernsey then in preparation. Material of a few plants met with by D. McClintock and Lady A. Brewis on different occasions was submitted to B. A. Miles for determination, with mixed success, before being deposited in the herbarium of the Société Guernesiaise in St. Peter Port (STP) and the few records resulting published in the eventual book (McClintock 1975). Conscious of the inadequacy of the treatment of the group in that work, McClintock prevailed upon the present writer to pay a week’s visit to Guernsey in July 1978 with a view to a more intensive survey. The opportunity was taken to spend a day and a half in Sark as well and three days in Jersey after that, with sufficiently productive results to warrant the publishing of new lists for all three islands (Allen 1981; Le Sueur 1984). Those lists have subsequently been considerably modified and extended in the light of four further visits by the writer to Jersey and Guernsey for a week each (including a day on Herm) in 1990; to Jersey and Sark, similarly for a week each, in 1992; to Alderney for four days in 1993; and finally to Jersey again for a week, with a day’s return to Sark, in 1994 (the updated Sark list, without localities, was subsequently published by Marsden (1995)). All material collected has been deposited in BM except for a few specimens donated to JSY and STP. BRAMBLES OF THE CHANNEL ISLANDS 423 As a result Alderney, Sark and Herm can be claimed to have been investigated more or less exhaustively; Jersey and Guernsey, however, have been covered only incompletely — possibly important omissions in Jersey are Les Grands Vaux and parts of the north coast - and those larger two have undoubtedly more to yield up. ISLAND LISTS The records for each of the main islands are considered below in turn, in the geographical sequence of these from south to north. As no non-British taxa have as yet been detected, the species order in each list is that in Kent (1992), to which reference should be made for sectional and series names. Many specimens cannot be matched at present with any entity known from the rest of the British Isles or mainland Europe whether described or not. Unfortunately, authentic material of many taxa recorded from northern France is not in any British herbarium and the location of type specimens is all too often unknown. By no means all of those, however, would now be rated worthy of taxonomic recognition, for names have frequently been conferred in the past on the unique morphotypes (or “individua”’) that the peculiar reproductive system found in the Rubus fruticosus ageregate not uncommonly gives rise to, or on entities with a distribution too narrowly local to have any significance geographically. In each of the main islands, however, there are one or two “innominati’ that are relatively widespread, some at least of which may qualify for description should they turn out to occur widely outside the Channel Islands as well. Details of these select few have been thought worth appending at the end of the respective lists. Now that UTT-gridded Ordnance Survey maps have become available in recent years for each of the four largest islands, it is at last possible to pinpoint Channel Islands records with six-figure references, and these are provided below in appropriate instances. Numbers cited for some of Rogers’ specimens are not his own (he did not use collection numbers) but those of Herb. Barton & Riddelsdell, now in BM, into which were incorporated many specimens of particular interest extracted from Rogers’ herbarium by Riddelsdell after it passed into his possession. JERSEY The most southerly of the four main islands and the one with the warmest climate, Jersey is also much the largest, its 115 km* making it almost twice the size of Guernsey. Though 24 km from the mainland, not nearly as close as Alderney and Sark, it has much the shallowest intervening sea, a principal reason for thinking that it was the last of the four to be severed from Normandy (Sinel 1912). While it resembles Guernsey in gradually sloping one way, it does so in the opposite direction from north to south. The north and south-west coasts are fringed with steep cliffs with a narrow belt of heath on top which widens in places into distinct commons and is one of the two habitats in the island particularly productive of Rubus diversity. The other is the series of long wooded valleys running down the centre of the island to the south coast. The lengthiest of these, the St. Lawrence Valley (since the building of two large reservoirs better known as Waterworks Valley), preserves at its north tip, close to the museum at Hamptonne, a fragment of rocky woodland with an exceptional assemblage of unusual brambles, almost all of them so far unmatched with any known from Britain or France. This is the one piece of ground encountered in the whole of the Channel Islands that most gives the impression of a portion of Normandy surviving little changed from the period before it was cut off by the sea. R. briggsianus (Rogers) Rogers Recorded by Rogers from one spot near Waterworks Valley and by Druce from La Corbiére. In neither case does a specimen seem to have been kept. Still not certainly known on the European mainland, the species may yet survive in the coastal heath. R. couchii Rilstone ex D. E. Allen Common and widespread in the north-east corner, rising to abundance in Rozel Woods. A specimen of Druce’s in OXF from Flicquet Bay (where the species is prominent today) was erroneously referred by Watson (Druce 1931) to “R. cariensis” sensu Rogers (1.e. R. altiarcuatus W. C. Barton & Ridd.). R. laciniatus Hayne & Willd. Mont du Vallet, L’Etacq, since the 1950s (Le Sueur 1984). 424 D. E. ALLEN [R. lindleianus Lees Druce (1907), who was familiar with this common English species, claimed to have seen it “in several places” on his recent visit, leaving it ambiguous whether Rogers had confirmed his identification. None can be found in OXF today, no sign of the species has been discovered in Jersey since and it appears to be absent from north-west France. Presumably one or more other white-flowered, eglandular brambles were mistaken for it.] [R. macrophyllus Weihe & Nees Rogers may have been correct in recording this from a deep lane on the west side of Bouley Bay, but his specimen (BM) is not satisfactory enough for certainty. Though occasional in woods in central Normandy, this species has only once been recorded in the Cotentin Peninsula. | . questieri Lef. & P. J. Mull. Common to abundant in several wooded tracts, but absent from large areas. Rogers believed he saw this only in the Vallée des Vaux, where Ley had collected it earlier, but there are specimens of his in BM (4900, 7459) from two other localities which he failed to recognise, suggesting that his concept of the species was too narrow. . cardiophyllus Lef. & P. J. Mull. Rare, as in Normandy, and nowhere more than a patch. On roadsides in La Ville des Quennevais and Waterworks Valley but on wood margins in the Mourier Valley and two places in St. Peter’s Valley. Earlier records are dubious: the surviving specimen (BM, 8093) of what Rogers took to be a small-leaved form of this, at Rozel and Pont Marquet, is of some other bramble, while Druce’s material from Le Portelet and St. Aubin’s (Druce 1931) was so determined by Rogers in an aggregate sense only. . davisii D. E. Allen Open heath south of Gros Nez, extending thence along the cliff-top towards Plémont. . dumnoniensis Bab. Rogers recorded a supposedly strong form of this “especially abundant at Gorey Bay”, but his specimen (K) is of some quite different (but unknown) bramble. Another specimen of his from Rozel (BM) he also incorrectly referred to this species. His claim to have seen it in abundance in Waterworks Valley must accordingly be treated with reserve; though certainly present, it is no more than occasional there today at best. Apart from a bush in the next valley to the west (on the wooded margin of La Rue de I|’Aleval) and several on the upper common at Le Ouisné it has been noted otherwise only on the north coast, at Le Marionneux (a colony), Le Jardin d’ Olivet and La Chrétienne. . polyanthemus Lindb. Occasional. Mostly among bracken on the north coast heaths, but in woodland in the Vallée de Bellosanne. . prolongatus Boulay & Letendre Rare. Several patches at top of Waterworks Valley and a bush on a wood margin near Tesson Mill in St. Peter’s Valley. Rogers’ specimen (BM) of the “R. micans” he listed from Anne Port proves to be this and probably his Gorey record of that belongs here too. [R. rhombifolius Weihe ex Boenn. “South of England incurvatus”’, the name by which Rogers mostly understood this species, was recorded by him from St. Aubin’s Bay. No specimen has been traced and the determination seems unlikely to have been correct. ] [R. rubritinctus W. C. R. Watson Very rare. Two patches on lane at rear of Le Jardin d’Olivet. R. Recorded by Watson (1958) on unknown evidence and almost certainly in error. ] sprengelii Weihe Very local. Apart from a find by Druce in 1908 vaguely localised as “St. Helier” (OXF), where it must surely have been obliterated by building, all records are from the north-east of the island, where it is plentiful especially around Bouley Bay. . armeniacus Focke “Himalayan Giant” An increasingly widespread escape, though in colony strength as yet only in La Rue de I’ Aleval. This cultivar, so very extensively grown in English gardens since the Second World War, has found little or no favour in France and is conspicuously absent from Normandy and Brittany. . lamburnensis Rilstone Specimens (BM, 8043-4) of a bramble found by Rogers “in good quantity” in Waterworks Valley match well Cornish material of this species in BM authenticated by Rilstone himself. These Cornish specimens were initially labelled by him as R. macrostemon Focke, to which Focke had independently referred, rather hesitantly, Rogers’ Jersey specimens earlier. . ulmifolius Schott Very common. . adscitus Genev. Very local. Abundant in wood along east bank of Queen’s Valley Reservoir; also a patch or:two in St. Peter’s Valley and its Rue de I’ Aleval branch. A specimen of Druce’s from Gros Nez (OXF), misdetermined by Focke as the Anjou species R. andegavensis Bouvet, was initially referred to R. adscitus by Watson (Druce 1931), a determination, however, he later withdrew; its identity remains unestablished. a if eo = eee BRAMBLES OF THE CHANNEL ISLANDS 425 . boraeanus Genev. Very common and (as Rogers noted) generally distributed, including on the fixed dunes of Les Quennevais. . caesarius D. E. Allen Locally abundant. Especially plentiful in the coastal districts, on banks, roadsides and open heathland, but scarce in the central valleys. This is clearly the “abundant” bramble which Rogers passed as the species now known as R. vestitus Weihe, although, as he admitted, “usually untypical”. . corbieri Boulay ex Corbiére Very rare. Single bushes at Portinfer and Bouley Bay. . leightonii Lees ex Leighton Locally common at several points on the north coast, very thinly scattered elsewhere. . percrispus D. E. Allen & R. D. Randall Extinct? Old quarry, Le Jardin d’Olivet, one clump 1978, gone by 1990. . dentatifolius (Briggs) W. C. R. Watson Locally plentiful on wooded slopes around Bouley Hay; otherwise seen only in St. Peter’s Valley (including a colony in its Rue de |’ Aleval branch) and upper part of Waterworks Valley. There is, however, a specimen in BM collected on the south-west cliffs of La Moye by R. P. Murray in 1900, misdetermined as R. dumnoniensis. [R. adenanthoides Newton Listed for Jersey by Watson (1958) under the name mostly applied by R. him to this species, R. pseudadenanthus W. C. R. Watson. As none of Druce’s material received this name from him, it appears he was repeating Focke’s determination of one of Rogers’ specimens as R. adenanthus Boulay & Gillot. This was a very handsome” plant found in some quantity in a very sunny spot in Gorey Bay. Rogers subsequently provided a description of it under that name in his Handbook (Rogers 1900: 53), indicating that it was a very hairy, considerably glandular-aciculate plant intermediate between R. radula Weihe ex Boenn. and R. vestitus. At least one specimen was formerly in his herbarium, but none can now be traced. ] bloxamii (Bab.) Lees Rare. Except for a population in oakwood at the upper end of the Vallée des Vaux seen only on the north coast cliffs: at intervals from Plémont to Gros Nez and on the east side of Bouley Bay. [R. echinatus Lindley A record of this from heathland by La Téte d’Ane (Le Sueur 1984) is R. R. suspect: a similar-looking unnamed bramble was all that could be found in 1994 at this locality. | peninsulae Rilstone Common to abundant in and near most stretches of woodland. A species not described till 1950 and known elsewhere only in Devon, Cornwall and one wood on the northernmost part of the Brittany coast, this was initially queried by Rogers as R. oigoclados P. J. Mull. & Lef. (9169, 9170) or R. micans Godr. (7458) but eventually (Rogers 1900: 78) regarded by him as a form allied to R. botryeros (Focke ex Rogers) Rogers. . conjungens (Bab.) Rogers Rogers’ record from Bouley Bay of the species then known as “R. corylifolius var. cyclophyllus” is confirmed by his specimen in BM. This widespread British bramble has as yet been detected on the European mainland only in Belgium. It is frequent near the coast in Devon and Cornwall. nemorosus Hayne & Willd. Very local. Bouley Bay district and in and between the central valleys, but nowhere more than a patch. [R. pruinosus Arrh. Rogers’ sole record of this (Bouley Bay) must be treated with reserve in the R. absence of a specimen. | transmarinus D. E. Allen Correctly identified by Rogers as the member of sect. Corylifolii Lindley familiar to him round Bournemouth as “R. dumetorum var. ferox’, this is occasional, rather than “generally distributed”. Since the species was described (Allen 1994) it has proved to occur widely in Normandy and plentiful in the district of Brittany between St-Brieuc and St- Malo. Innominati. Two unnamed brambles occur quite widely. One of these, as yet unmatched with any known outside the Channel Islands, is a member of ser. Rhamnifolii (Bab.) Focke with pink flowers and many long, subulate rachis prickles. This is found mainly in the Bouley Bay - Rozel area and the central valleys (up to colony strength beside the All in St Peter’s Valley at WV605518), was twice collected by Rogers (BM, 10082, 10083) and has been detected in Guernsey also, on a hedgebank near St. Saviour Reservoir. 426 DE. ALLEN Unlike that, a robust member of ser. Mucronati (Focke) H. E. Weber has been encountered on the mainland immediately to the east: round a small wood on the D42 | km north-east of St- Maurice-en-Cotentin, near Barneville. This has been noted in Jersey in three different areas: Waterworks Valley, in some quantity north of Millbrook Reservoir and again east of Hamptonne; Vallée des Vaux at WV651514; and Val de La Mare Reservoir, by the dam. In the third of these the population is wholly white-flowered, whereas elsewhere only pink petals have been seen. In and near an oak-holly wood near Trois Bois (Coétil de la Qualité nature reserve, WV623525) are patches of another robust bramble with large pink flowers, often red-based styles and incised leaflets, which, despite a sometimes amply glandular rachis and slightly aciculate stem, seems best in ser. Sylvatici (P. J. Mull.) Focke. This has also been collected in the Cotentin Peninsula, but in more distant parts, and is common in woods in the South Hams district of S. Devon, v.c. 3, where it is well-known to R. W. Gould (as “the Andrews Wood Bramble’) and was collected by Briggs from 1872. GUERNSEY Covering 63 km*, Guernsey has much less ground conducive to Rubus diversity than Jersey, lacking wooded valleys at all comparable in scale and length and surviving expanses of suitable heathland of a similar size and number. The flat, rather exposed northern one-third is largely bereft of interest for the batologist, in marked contrast to the south, where plantations, massively augmented by self-sown trees, clothe the cliffs and extend some way inland in places. That some natural scrub may have escaped on these cliffs the otherwise wholesale removal of the island’s original tree cover is suggested by the present-day restriction to those of such characteristic ancient woodland indicators as Euphorbia amygdaloides L., Ruscus aculeatus L. and Luzula sylvatica (Huds.) Gaudin (McClintock 1975: 19). That might account for the presence on them too of several Rubus species known nowhere else in the island — though these might equally well have been brought by frugivorous migrant birds, for which the fringe of trees and shrubs along the cliff- tops, especially its profusion of bullace, Prunus domestica subsp. insititia (L.) Bonnier & Layens (which fruits heavily in some years at least (N. Jee, pers. comm., 1990)), must constitute a favourite landfall. R. briggsianus Seen by Rogers in several places (and “in considerable quantity’, according to a postcard he sent to E. F.Linton in 1914, now in BM), including St. Peter, Petit Bot Bay, Cobo and quarries between St. Sampson and L’ Ancresse, from the last two of which specimens of his are in BM. Extensive draining of boggy ground since may have rendered it extinct. R. laciniatus Several recent records for this naturalised escape are cited by McClintock (1975). Rogers met with it only in Petit Bot Bay quarry. [R. oxyanchus Sudre Cited for the Channel Islands by Edees & Newton (1988) as a result of the patch of R. daveyi, q.v., having initially been determined as this. ] R. plymensis (Focke) Edees & Newton Common in plantations round Belvedere House, Fort George. R. questieri Very local. Rogers said he had searched for this in Guernsey in vain, but as in Jersey his over-narrow conception of the species is shown by his having collected it under other names: his “R. villicaulis’ (STP) and “R. rhombifolius’ (BM, 7211; K), both from Fermain, where it is abundant in woodland today, have proved to be this. Scattered bushes extend to plantations on the south edge of St. Peter Port and occur in one or two other spots as well. R. cardiophyllus Rare. Of the two stations recorded by Rogers, Fermain Bay yielded just one bush in 1978 while Petit Bot Bay may have been a loose localisation of a colony that grows among gorse on the cliff-top 0-5 km to the south. A bush has also been noted at Jerbourg. R. cordatifolius (Rogers ex Ridd.) D. E. Allen Supposed by Rogers at the time of his visit to be merely a luxuriant state of what he understood as R. dumnoniensis, this very distinct bramble still hangs on at Petit Bot Bay, the source of his specimen (BM, 9823) that now constitutes the lectotype. Elsewhere in the south it has been noted in coastal scrub at Icart and in Silbe Nature Reserve in St. Peter; its presence, however, also on the north edge of St. Peter Port, in two spots east of Le Hougue des Quartiers, suggests it could turn out to be fairly widespread. It is still not known in any of the other islands apart from Sark or for certain anywhere outside them. A Belgian bramble that has been passing there under the names of “R. nemoralis” and “R. argenteus subsp. incarnatus var. leventii’” (H. Vannerom, pers. comm., 1993) has been BRAMBLES OF THE CHANNEL ISLANDS 427 identified with it by Vannerom (1998), but the one specimen of that seen by the writer, though very similar, lacks the pecuhar inflorescence structure of the Channel Islands plant; the resemblance may therefore be fortuitous. R. daveyi Rilstone One patch in a bullace thicket beside the cliff-top lane midway between Portelet Bay and Les Fontenelles at WV300745 and a further bush nearby among gorse. Hitherto believed endemic to Cornwall and a few places in Devon. [R. dumnoniensis Rocquaine Bay was identified by Rogers (1899) as among the “one or two other localities” apart from Petit Bot Bay (his locality for what was to prove R. cordatifolius) in which he claimed in the initial paper to have seen this. Though its presence in the other three main islands makes it a likely Guernsey species, the lack of later records and the over-broad interpretation of the taxon by Rogers, revealed by material from Jersey and elsewhere misdetermined as this species, make it necessary to treat his claim with reserve in the absence of any specimen. ] [R. herefordensis Sudre The basis for a claim by Watson (1958) to have seen a specimen of this from St. Andrew has not been ascertained. The species as now understood is unknown outside Hereford, v.c. 36.] R. polyanthemus Local, but markedly more plentiful than in the other islands. Particularly large populations in the south-west corner, as in Silbe Nature Reserve and round Pleinmont. R. prolongatus Rare. Recorded from six localities in the south, half of them on the coast, but in quantity only on Petit Bot cliffs. [R. riddelsdellii Rilstone A specimen in STP collected by McClintock west of Chateau des Marais was thought by Miles to be this, but the material is so poor that any determination would seem unwise. ] rubritinctus Listed by Watson (1958) on unknown evidence. Rogers remarked on its absence. sprengelii Collected by Rogers at Fermain Bay (BM, 6073), but perhaps now extinct. armeniacus “Himalayan Giant” As in Jersey, becoming a widespread escape. pydarensis Rilstone Very local or overlooked. Common in hedges about 1 km inland in the district south of Les Arquets, St. Peter (conf. L. J. Margetts). Specimens in BM collected by Rogers at Fermain Bay (4898) and La Valette, referred by him to an aggregate taxon he called “R. argentatus’, also turn out to be this species, hitherto believed endemic to Cornwall and a few places in Devon. R. ulmifolius Very common. [R. adscitus Believed to have been seen in 1978 in Silbe Nature Reserve, but requires confirmation]. R. boraeanus Occasional. Noted more especially in Castel and St. Andrew. Surprisingly, Rogers appears not to have seen this. Of the only two localities given in the original paper (Rogers & Rogers 1898), one is omitted from his 1899 paper, suggestive of second thoughts, while the specimen from the other locality (BM, 10258) has proved to be a misdetermination. Possibly the species has colonised the island subsequently to his visit. [R. conspersus W. C. R. Watson Listed by Watson (1958) on the strength of his determination of one of Rogers’ specimens in BM from L’ Ancresse Common as this in 1951. The specimen, however, is R. corbieri.] R. corbieri Mainly in the north, where it is frequent on hedgebanks and abundant in old quarries, having probably colonised Guernsey from Sark subsequent to that island’s sixteenth-century resettlement or, alternatively, direct from the Cotentin Peninsula beyond. R. leucostachys Sm. A colony by the cliff path on the south side of Fermain Bay. [R. vestitus This was the species, very common in much of England, to which Rogers and his contemporaries generally misapplied the name R. leucostachys. In the supplementary list (Rogers 1899) he provided for Guernsey that appears with “generally distributed” and “very variable” against it. Presumably one or more of the island’s still-unnamed brambles passed as this - as was certainly the case with specimens of Druce’s in OXF and Rogers’ own (BM, 10081) from St. Andrew. ] R. coombensis Rilstone One bush in shade on the cliff path south of Le Vallon (WV330754). When discovered in 1990, this was believed endemic to south-west England, but it has since been recognised as a bramble local in quantity round Cherbourg. ae ete 428 D. E. ALLEN [R. moylei W. C. Barton & Ridd. Listed by Watson (1958) on unknown evidence. In his supplementary list Rogers (1899) included the name R. lejeunei var. ericetorum, by which he normally meant this species, but gave no locality. It is unlikely that Watson saw that paper, however. | R. norvicensis A. L. Bull & Edees Very local. Plentiful in plantations and shrubberies across the south-east corner, from above Moulin Huet Bay and Sausmarez Manor to the south edges of St. Peter Port. This species is otherwise known only in parts of England east of a line from Salisbury to The Wash and along one valley in south-west Wales. The compact nature of its Guernsey distribution and prevalence there in shady habitats comparatively recent in origin both suggest a colonist in the process of primary spread, perhaps following accidental introduction with nursery shrubs (as seemingly in some of its English stations). If so, that introduction must have been over a century ago, for this turns out to be the identity of one of the two brambles collected by Rogers’ son at Fermain Bay (BM) which, after some hesitation, his father concluded was R. rudis Weihe “type or near it” (Rogers & Rogers 1898; Rogers 1899) — an unlikely species to occur in the Channel Islands, however, on geographical grounds. R. percrispus_ This is the identity of the other Fermain “R. rudis” collected by F. A. Rogers, the specimen of which is in this case in K; it was initially determined by his father as the bramble now known as R. micans Godr. A depauperate example of probably this same species was collected on the cliff path above Petit Port in 1978. Recently described, like the previous species. R. percrispus is otherwise known outside the Channel Islands as yet only in parts of southern England and a single locality each in west Wales and Ulster. R. dentatifolius Very local. Shares old quarries in Vale with R. corbieri and still as abundant in one as when Rogers found it there. Most of the few other records are from the south coast. . leyanus Rogers Rue du Banquet, near Pleinmont, one patch in hedge of the northernmost bend. . bloxamii Locally common; much more plentiful than in Jersey. Mainly in the southern half of the island but conspicuously absent from the self-sown woodland at Fermain. A specimen in CGE collected by Babington at Fort George is labelled “R. koehleri var. lejeunei”’. [R. radula Recorded from Fermain Bay by Rogers (1899) in his supplementary list, but no specimen has been traced. A specimen in STP from Mount Durant determined as this by Miles (McClintock 1975) has turned out to be R. norvicensis. | [R. conjungens “Fairly frequent” according to Rogers, but there have been no later finds and his statement must be treated with reserve in the absence of a confirming specimen. ] R. transmarinus Rogers’ initial description of the bramble he knew as “R. dumetorum var. ferox” as generally distributed was modified in his supplementary list to “frequent”. It appears, however, to be thinly scattered at best, mainly in the vicinity of the south-east coast. poe) Innominati. Of numerous unnamed brambles, one is particularly widespread, a member of ser. Vestiti (Focke) Focke with deep pink flowers, red styles and aculeate sepals, this is common in woodland in the centre of the island, especially along the Fauxquets Valley. Seemingly the same plant occurs in Jersey, in a wood by Gargate Mill Farm (WV604518) in St. Peter’s Valley. SARK The smallest of the four main islands and unique among them in being entirely a lofty plateau, which gives it the appearance of an aircraft-carrier, Sark was probably a stretch of treeless heath before its first substantial colonisation in the sixteenth century. Like the very similar [le de Bréhat off the north coast of Brittany, it is essentially two islands, connected merely by a narrow isthmus; a detached portion of the larger one, the now strictly private Brecqhou, lies just 200 m off the west coast. High cliffs all round permit access to the sea in only a few, very restricted places. Protected from Atlantic gales to some extent by Guernsey, 13 km to the west, it presents a comparatively lush contrast to Alderney, as remarked upon by Druce (1907) after visiting the two for the first time in quick succession. Much less ground than in the other main islands has escaped cultivation and such of the little heath cover that remains is best represented by Eperquerie Common, at the north tip. R. altiarcuatus W. C. Barton & Ridd. In an isolated clump of tall gorse by La Valette (WV474764) and several further bushes along the cliff-top to the north-west. The sole Channel Islands locality for this widespread species of south-west Britain, also in the far west of both Ireland and Brittany. BRAMBLES OF THE CHANNEL ISLANDS 429 R. cordatifolius Scattered among gorse south-east of the Pilcher Monument (WV455754) — which is by no means “near the Dixcart Hotel”, where Rogers recorded this in quantity (specimen in BM). A search of the hotel’s vicinity has proved fruitless, but the area is no doubt much changed since 1897. dumnoniensis Several patches in a rough field west of D’Icart (WV466756) and one bush near the top of Eperquerie Common. nemoralis P. .J. Mull. In four widely separate places in Great Sark and one in Little Sark, mostly as single bushes and nowhere more than a patch. polyanthemus A heath relic. One or two bushes at Eperquerie Landing; also collected on a cliff on Brecghou by Mcclintock in 1968 (STP). prolongatus Several patches in the south-west of Little Sark and a bush in a gorse hedge just below D’Icart. rubritinctus Several patches at different levels on Greve de La Ville cliffs. ulmifolius Very common. corbieri Very common. A conspicuous feature of virtually every hedgebank, this probably post-dates their construction, having spread across from the north-west of the Cotentin Peninsula, where it occurs in comparable profusion locally. Rogers at first referred it to the group of brambles to which he applied the name R. schlechtendalii (Derrick 1898), but by the time he published his records (Rogers & Rogers 1898) had decided it best regarded as a strong and highly glandular form of R. macrophylloides Genev., a species described from Angers. Subsequent collectors were to have specimens referred also to R. adscitus or R. boraeanus before the plant’s true identity was finally established c. 1990 (Allen 1992). Patches of apparent hybrids with R. ulmifolius have been noted in two places. R. leucostachys One bush on cliff-top north-west of La Valette and a clump by the cliff path above Dixcart Bay. The latter may be much the same spot as the “furze brake near Dixcart Lane” from which Rogers recorded in a supplementary list (Derrick 1898) a bramble under this name. By this he normally understood R. vestitus, which is absent from the Channel Islands, or, in Jersey, as R. caesarius. R. dentatifolius Frequent in rough ground. Probably general in the pre-resettlement era. R. conjungens An unlocalised Sark specimen in CGE collected by Babington in 1838 and successively determined by him as “R. rhamnifolius” (as published in Babington 1839) and “R. thyrsoideus” is in fact this, as first recognised by Miles. R. nemorosus Frequent; much more plentiful than in the other islands. In places in open heath (in which it was doubtless prevalent formerly). R. transmarinus Occasional, mainly around the south-east coast. Perhaps in part the bramble determined by E. F. Linton under the aggregate name of “R. corylifolius” which Hurst (1903) claimed to be abundant on Brecghou. Bo re Se ae Re aU ae Innominati. Sark has two relatively widespread brambles that are also known outside the island but so far unnamed. One, a pink-flowered member of ser. Sylvatici with strongly hairy anthers and reminiscent of small forms of R. gratus Focke, occurs in some quantity on the cliff-top north-west of La Valette and in two places on Eperquerie Common and also as single clumps on Harbour Hill (WV476758) and by D’Icart. Since its original discovery (“Sa36” in Allen 1981) it has been found in the third-largest of the Isles of Scilly (Allen 1997) and on a wood margin in the south-east corner of Hampshire, v.c. 11. This distribution suggests it may occur in Brittany too. The other is a member of ser. Radula (Focke) Focke recalling R. longithyrsiger Lees ex Focke but with pink flowers, pale styles and hairy anthers. In addition to a colony at Eperquerie Landing this occurs on the east cliff-tops between La Valette and Gréve de la Ville as well as by D’Icart (WV466755); it is also in Jersey, dotted around Dannemarche Reservoir in Waterworks Valley. Also worthy of mention is a very distinct member of ser. Discolores (P. J. Mull.) Focke with hairy anthers, pink filaments, pink-based styles and mucronate terminal leaflets with a lengthy mucro. Patches of this occur in Sark on the cliff-top at Gréve de la Yule and in the gorse heath near the Pilcher Monument, in Alderney (where A. Newton has also seen it) in the old quarry on Val Longis (WA581076) and on the coast west of Cherbourg near Landemer. 430 DU EALEEN HERM Covering 2-5 hectares, the largest of Guernsey’s satellites, privately-owned but publicly accessible, Herm is an object lesson in the sharply disproportionate reduction in species diversity that can occur in the case of islands that are below a certain size — even in the case of plants that are bird- dispersed. In startling contrast to its neighbour some 5 km to the east, hardly any of Guernsey’s numerous Rubus species are to be found, not even most of the commonest. Apart from the predictably abundant R. ulmifolius, indeed, and a patch by the landing of a member of the “R. dumetorum” aggregate, there appears to be only one other — and that not a Guernsey one, so far as is known at present. Resembling R. orbus W. C. R. Watson of south-west England and belonging like that to ser. Vestiti, this is common in the south half in ground with a modicum of shelter. Herm’s own even smaller satellite, the more strictly private Jethou, had several bramble specimens collected from it by Babington. Borrer, to whom he sent one, referred it doubtfully to R. leucostachys (Babington 1839); but Miles, who discovered it in CGE in 1965, thought it was indeed that species, in the original sense of Smith; though that has since been established as occurring in both Guernsey and Sark. Babington’s material is unfortunately too scanty and immature to be given a confident determination. ALDERNEY The northernmost of the main islands, 13-5 km west of Cap de la Hague and thus almost twice as close to the mainland as Jersey, Alderney differs from the other main three in resembling an inverted saucer, the one town standing on a hill in the centre. As in Guernsey, the northern third is low-lying and only the more southerly part fringed by cliffs. Though its cultivation dates from a much earlier period than Sark’s, it still has extensive tracts of rough grassland and heath, of which those of La Giffoine are the most rewarding for the batologist. Introduced trees form dense thickets in places, but rarely shelter brambles of interest. The most exposed of the islands to the winds of the English Channel, Alderney’s Rubus flora mainly owes such distinctiveness as it has to its consequently more oceanic climate. R. viridescens (Rogers) T. A. W. Davis One patch among roadside gorse above Fort Tourgis just south of Roc a |’Epine. R. dumnoniensis Local; patches on the coast here and there in the western one-third. Relatively more plentiful than in the other islands (though it seems to be missing from the coastline round Cherbourg). R. iricus Rogers In plenty with R. viridescens. Locally abundant on La Giffoine southwards from east of Fort Clonque; also one bush in the Valley Gardens in the middle of St. Anne. R. polyanthemus A patch in gorse heath on top of Essex Hill. R. prolongatus Widespread and locally abundant on the south-west heaths; also abundant round the television tower on Les Rochers and in some quantity on Essex Hill. Markedly more plentiful than in the other islands. R. armeniacus “Himalayan Giant” Escaping from a garden in St. Anne. R. ulmifolius Very common. [R. winteri P. J. Mull. ex Focke Listed by Watson 1958) following his determination of one of Druce’s specimens in OXF as this, in place of Rogers’ “apparently R. leucostachys’. It is neither of those, however, but may be the same as an unnamed Guernsey bramble of hedgebanks in St. Andrew. | [R. adscitus Another of Druce’s specimens in OXF was considered by Watson at one time to be this (Druce 1932), but it appears to be merely a form of R. ulmifolius.] R. boraeanus Common. - R. corbieri Very thinly scattered, achieving quantity only in three particularly sheltered locations: Barrack Master’s Lane (“Essex Glen”), Vau du Saou and the top of Le Val Reuters. R. dentatifolius Common and locally abundant on the southern heaths and cliffs southwards from Fort Tourgis in the west and Le Longue Pierre in the south; elsewhere rare or absent. R. tuberculatus Bab. Common, taking the place of R. transmarinus. Innominati. Given its greater proximity to the mainland, Alderney has a proportionately higher share of brambles that cannot be matched with any known in Britain. One of these is widespread on the neighbouring Cotentin Peninsula, a member of ser. Discolores resembling R. rossensis Newton, this occurs at the north tip near the lighthouse and in two places on the south coast. A large patch of another member of ser. Discolores on the upper margin of Barrack Master’s Lane at WA591077 appears to correspond to a poorly-understood bramble of northern France that has BRAMBLES OF THE CHANNEL ISLANDS 43] passed under the illegitimate name of R. propinquus P. J. Mull. A very distinct, white flowered member of ser. Sylvatici with many tiny rachis prickles and long, acuminate leaflets is also widespread, with clumps at intervals along the south-east coast and several patches on La Giffoine. DISCUSSION Altogether, 39 entities currently accorded taxonomic recognition as species, including two naturalised horticultural ones, are accepted as reliably on record from one or more of the islands. Of these, 26 are on record from Jersey, 23 from Guernsey, 14 from Sark and 11 from Alderney. Though these totals roughly correlate with island size, the diminution is by no means a straight reflection of increasing failure to colonise on the part of certain species. Instead, as the Table below demonstrates, there are marked discrepancies between the four main islands in how the totals are made up:- Jersey Guernsey Sark Alderney R. briggsianus 0 0 R. couchii 3 R. laciniatus 1 py) R. plymensis 1 R. questieri 3 1 R. viridescens l R. altiarcuatus 1 R. cardiophyllus ] R. cordatifolius 1 l R. daveyi ] R. davisii ] R. dumnoniensis 2 1 2 R. iricus 3 R. nemoralis 1 R. polyanthemus 2 2 1 1 R. prolongatus l 2 1 3 R. rubritinctus l ] R. sprengelii | 0 R. armeniacus 2 2 1 R. lamburnensis 1 R. pydarensis 1 R. ulmifolius 3 3 3 3 R. adscitus 1 R. boraeanus 3 2 3 R. caesarius 3 R. corbieri 1 pp yy R. leucostachys ] 1 R. coombensis 1 R. leightonii 3 R. norvicensis 1 R. percrispus 0 0 R. dentatifolius D 2) 2 3 R. leyanus 1 R. bloxamii 1 3 R. peninsulae 3 R. conjungens 0 0 R. nemorosus 1 2 R. transmarinus 2 2 ye R. tuberculatus 3 O - extinct ?, 1 - rare to very local, 2 - widely scattered, 3 - locally to very common 432 D. E. ALLEN All of these species occur in Great Britain also - though in the case of R. iricus only in an island group far offshore - with the single exception of R. cordatifolius, which is apparently endemic to Guernsey and Sark. It is possible, however, that the seeming identity of R. caesarius with a bramble locally plentiful in one small district of S. Hants., v.c. 11, between Southampton and Portsmouth (Allen 1994), is the result of a fortuitous overlap in characters, in which case that would be a second apparent endemic (for despite the abundance of that species in Jersey no sign of it has so far been found in the nearer parts of the mainland, very surprisingly). Though no Continental taxa unknown in Britain have yet been detected in the islands, at least one or two of the many brambles to which no name can be put at present will almost certainly prove to come into that category once the taxonomy of the Rubus flora of north-west France has been more fully worked out. Until only two or three centuries ago, however, R. corbieri may have belonged in that category, for in the south-east corner of Dorset, v.c. 9, the sole area in which this conspicuous species is known on the other side of the English Channel, it has all the appearance of a comparatively recent colonist, though probably present well before the first specimen was collected, in 1892 (Allen 1992). Of the 37 species apart from R. caesarius that the islands share with Britain, eight are not yet known to occur on the European mainland. These are R. briggsianus, R. daveyi, R. davisii, R. iricus, R. lamburnensis, R. pydarensis, R. norvicensis and R. percrispus. R. norvicensis, as already intimated, appears to owe its presence to accidental introduction with nursery plants at some comparatively recent date, while R. percrispus, despite a wide scatter of localities across southern England, has had too brief a taxonomic history for its presence or absence on the European mainland to have had time to be adequately tested as yet. The rest of the eight, though, are all strongly western species whose ranges could well extend at least to Brittany. Nevertheless, extensive sampling of coastal areas of both Brittany and western Normandy in recent years has not only failed to reveal those, but has turned up instead other species with a similar range on the other side of the English Channel: R. hastiformis W. C. R. Watson, for example, and R. aequalidens Newton along the Céte de Granit Rose, the former in abundance, and R. rilstonei W. C. Barton & Ridd. in no less abundance on Cap Fréhel, west of Dinard - each of them seemingly no less likely candidates for the Channel Islands flora. More perplexingly, there are several brambles abundant in the adjacent parts of France that appear to be absent from the Channel Islands. The most conspicuous of these absentees is R. venetorum D. E. Allen, which is plentiful, commonly in profusion, almost throughout Brittany, including on the Ile de Bréhat, to Jerseys south-west. Almost as odd is the failure of R. thyrsigeriformis (Sudre) D. E. Allen to reach at least the northern islands. It has found its way from the north of the Cotentin Peninsula to a wide part of S. Hants., v.c. 11 (in which it gives the appearance, like R. corbieri, of being a comparatively recent invader still in the course of primary spread). Similarly, there is no sign of an unnamed member of ser. Sy/vatici that grows in profusion on the heaths around Cherbourg. Why has R. questieri, not noticeably more plentiful on the mainland than R. venetorum, been able to colonise Jersey and Guernsey and yet none of those? It may be that the transmarine movements of the relevant species of frugivorous birds have persistently been from other directions than from the mainland for the most part, in particular from the north-west. It certainly looks suggestive that two species otherwise confined to Britain and Ireland and especially plentiful in south-west England, R. leyanus and R. plymensis, are both known from just a single locality in the Channel Islands (and in the westernmost of those, Guernsey) and from just a single mainland locality too, in the Cotentin Peninsula to the east — as if they were legacies of one and the same fly-line. R. couchii has (or rather had, for at Cherbourg it is probably extinct now) a similar distribution pattern, though in its case the population in Jersey is a very large one and in Britain the species appears confined to a solitary district in south-east Cornwall, v.c. 2. Even more suggestively, R. daveyi, a common species of Cornwall which extends thinly into Devon, has as its only Channel Islands site a bullace thicket on top of a Guernsey cliff, precisely the kind of landfall likely to be made by a migrant coming in from the open sea to the west. A species of warbler, most probably a Blackcap, Sylvia atricapilla, is likely to have been responsible in such a case, for that is known to feed on blackberries (Snow & Snow 1988: 61) and there have been recoveries in Guernsey of individuals ringed in Dorset. Moreover, unlike members of the thrush family, the birds generally thought to be the principal disseminators of Rubus species. Blackcaps and other warblers are commonly seen on the Guernsey cliffs (N. Jee, M. Austin, pers. comm., 1990). BRAMBLES OF THE CHANNEL ISLANDS 433 Of all the British species that have turned out to be in the Channel Islands R. iricus was the least expected. At the time of its discovery in Alderney it was known elsewhere only along the Atlantic seaboard of Ireland, where it had long been presumed endemic and in parts of which it is very common. Subsequently it has proved to occur in abundance in the Isles of Scilly as well, which may thus have served as a half-way stepping-stone without which such a great distance seems much less likely to have been bridged by a bird with viable Rubus seed in its intestine (Allen 1997). The oceanic character of both Alderney and Scilly strongly suggests that the bird responsible for the original transmission between the two was a sea-going species, perhaps a member of the gull family. The possibility that gulls may disseminate Rubus species does not appear to have been entertained previously, but the recent find of a bush of R. daveyi in the middle of a gullery on Gugh in the Isles of Scilly (Allen 1997) lends that idea some credence. Certainly, there is evidence that gulls feed on the fruits of other plants: Solanum nigrum L. has been noted in quantity in gulleries on the Calf of Man, v.c. 71 (Allen 1984) and S. dulcamara L. in the crevices of cliffs alongside nests on the off-shore islands of Pembroke, v.c. 45 (Gillham 1956: 446), while Snow & Snow (1988: 167) cite records from Central Europe of different gull species eating the fruits of Vaccinium and Empetrum spp., Hippophae rhamnoides L. and even the blackberry-sized fruits of cherries (Prunus spp.), which flocks of Black-headed Gulls pluck from the trees in flight - an apparently recently-developed habit. The presence so far to the east of species characteristic of , or otherwise confined to, the extreme south-west of England is explained by the similar climate of the two regions. Although the Cotentin Peninsula of Normandy lies on the same longitude as Bournemouth and the Isle of Wight, its western half is so exposed to maritime influence in general and to the Gulf Stream in particular that its floristic affinity is rather with Cornwall and Devon (Allen 1996). That climate is shared with the Cotentin’s offshore islands, which can be expected to experience an added measure of oceanicity by virtue of their insular character. That essential homogeneity of the western Cotentin and at least the more northern of the Channel Islands is well shown by the distribution of R. corbieri, a robust, shaggy species clearly exceptionally well adapted to existence in such a wet and windswept region. Apparently relatively recent in origin, with R. bloxamii patently as one ancestor, it would seem to have spread aggressively at one period in the not-too-distant past, spilling across from the Cotentin on to each of the four main islands and colonising the northern three in proportion to the amount of shelter they offer. Though the most characteristic of its species, R. corbieri is not alone in being seemingly well- suited to the region’s relatively high oceanicity. R. iricus is a similarly robust and very hairy bramble, with presumably many centuries, if not millennia of adaptation to the climate of the west coast of Ireland. It can be no accident that Alderney is the only one of the Channel Islands in which it has become established - and in considerable quantity. Besides R. corbieri the archipelago is noticeably rich in other members of ser. Vestiti, a chief character of which is a dense indumentum: they include R. boraeanus, one of the commonest species of Jersey and Alderney (though absent from Sark), R. caesarius, similarly prevalent in much of Jersey, the unnamed bramble plentiful in woodland in the centre of Guernsey and the unnamed one that dominates the Rubus flora of Herm jointly with R. ulmifolius. Members of this series are likewise prominent in the Cotentin Peninsula. It is presumably for the same reason, that it can better cope with oceanic conditions, that R. prolongatus is distinctly more widespread in Sark and especially in Alderney than in Guernsey and Jersey, while the less hairy R. polyanthemus occurs in exactly the opposite island order of frequency. In just the same way R. prolongatus occurs on four of the isles of Scilly and in local abundance on the largest, whereas R. polyanthemus is known from only two and is very rare in both (Allen 1997). The difference between this pair in apparent oceanicity tolerance, though, would seem to obtain only at these latitudes, where R. polyanthemus, altogether a more northern species extending to Scandinavia, is at the southern limit of its range and thus presumably at a competitive disadvantage: on the west coast of Ireland it is R. polyanthemus that is locally common and R. prolongatus that is rare. That a preference for the relatively moister and milder climate of those parts of western Europe more especially subject to the influence of the Gulf Stream is not necessarily synonymous with a tolerance of oceanic conditions is particularly well exemplified by the distribution pattern of R. rubritinctus. Rubus specialists accustomed to the prevalence of that species in Devon and 434 D. E. ALLEN Cornwall (except the westernmost part) have difficulty in believing that in the Isles of Scilly it is represented by no more than a bush or two. No less puzzling at first sight is its wide scatter around Southampton Water yet total absence from the Isle of Wight, at most a mere 8 km offshore to the south (Allen 1990) - and despite the fact that this is a bramble especially dependent on birds for its dispersal, to judge from its marked tendency to occur as solitary bushes. The explanation for the second of these anomalies is that the Isle of Wight shields Southampton Water from the open sea. A slender plant with long ascending inflorescence branches and for the most part only sparsely hairy, R. rubritinctus is seemingly ill-adapted to oceanic rigours, and an overriding requirement for shelter from those largely excludes it from both the Channel Islands and the Cotentin Peninsula, just as it does from Scilly. Significantly, of the only two sites known for it in the Channel Islands, Jersey’s is on the inland margin of a coastal common and Sark’s is the only sheltered cove on the side that faces the mainland. The only area in that region where it thrives in quantity is the southernmost stretch of the Gulf of St-Malo coastline east of St-Brieuc, an area comparable to the Lancashire side of Morecambe Bay in England in the extent to which it constitutes an isolated “oasis” for this protection-demanding species. Another distribution that merits some final comment is that of R. leightonii, which Jersey alone of the islands possesses and that in considerable plenty. Together with R. bloxamii this has a relatively narrow corridor of abundance down the centre of England from Staffordshire to the Bournemouth district, to either side of which the two species diminish steeply. Though both ranges continue southwards across the sea into Normandy , they lose their tightly corseted character there and the species lose their closely parallel occurrence as well. Once across the English Channel R. leightonii becomes noticeably the more warmth-demanding and intolerant of oceanic conditions, extending apparently further south (to beyond the Loire) and much further east than R. bloxamii and, unlike that, missing Guernsey and all but the south end of the Cotentin Peninsula. Its prevalence in Jersey, however, is repeated along the southern coast of the Gulf of St-Malo, from which R. bloxamii is seemingly debarred climatically - as both species are from the main mass of Brittany to the west. These examples illustrate how differences between the islands in the Rubus species they possess or lack can in some cases be explained by scrutinising the wider ranges of those species for illuminating anomalies. They also serve to underline the substantial size of the addition to the phytogeographer’s armoury that the study of a critical group such as this can bring, by reason both of its considerable species diversity even in an archipelago of relatively small isiands and of the sensitivity of those individual species to subtle gradations in atmospheric moisture. ACKNOWLEDGMENTS A grant from the Appleyard Fund of the Linnean Society of London made it possible to extend to Alderney the fieldwork on which this paper is based. A grant from the B.S.B.I Welch Bequest also made it possible to replace the greater part of one year’s collection after its loss in transit. Iam most grateful for both. I would also like to thank David McClintock for introducing me to the Channel Islands and both him and Cherry Lihou for their assistance in the field on my initial visit, the late Frances Le Sueur, Patience Ryan and Marcia Marsden for hospitality and assistance in Jersey, Guernsey and Sark respectively, La Société Jersiaise and La Société Guernesiaise for access to their herbaria (and the latter also for the subsequent loan of specimens), Nigel Jee and Margaret Austin for information on Guernsey bird movements and ringing recoveries, Chris Preston for guidance on map grids, and Ray Gould, Len Margetts, Herman Vannerom, Giinter Matzke-Hajek and above all Alan Newton for assistance with determinations. McClintock and Gould were additionally kind enough to comment on this paper in draft. REFERENCES ALLEN, D. E. (1981). Brambles of Guernsey and Sark. Société Guernesiaise report & transactions 1980 20: 609-615. : ALLEN, D. E. (1984). Flora of the Isle of Man. Manx Museum and National Trust, Douglas. BRAMBLES OF THE CHANNEL ISLANDS 435 ALLEN, D. E. (1989). A neglected bramble of Guernsey. Watsonia 17: 436-437. ALLEN, D. E. (1990). The Rubus flora of the Isle of Wight. Watsonia 18: 21-31. ALLEN, D. E. (1992). Rubus corbieri Boulay (Rosaceae) in the British Isles. Watsonia 19: 131-133. ALLEN, D. E. (1994). Four new Anglo-Sarnian species of Rubus L. (Rosaceae). Watsonia 20: 143-151. ALLEN, D. E. (1996). British species of Rubus L. (Rosaceae) in the Cotentin Peninsula of Normandy. Watsonia 21: 129-133. ALLEN, D. E. (1997). Rubus L. (Rosaceae) in the Isles of Scilly: a revised list. Watsonia 21: 355-358. BABINGTON, C. C. (1839). Primitiae florae Sarnicae. Privately published, London. DERRICK, G. T. (1898). Additions to the Sark flora. Guernsey Society of Natural Science & Local Research report & transactions 1897 3: 166-169. DRUCE, G. C. (1907). Notes on the flora of the Channel Islands. Journal of botany 42: 395-402, 419-428. DRUCE, G. C. (1931). New county and other records. Botanical Society & Exchange Club of the British Isles report for 1930 9: 331-379. DRUCE, G. C. (1932). New county and other records. Botanical Society & Exchange Club of the British Isles report for 1931 9: 632-679. EDEES, E .S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. GILLHAM, M. E. (1956). Ecology of the Pembrokeshire islands, 3. Manuring by the colonial seabirds and mammals, with a note on seed distribution by gulls. Journal of ecology 44: 429-454. Hurst, C. P. (1903). The flora of the island of Brechou. Guernsey Society of Natural Science & Local Research report & transactions 1902 4: 1-14. KENT, D. H. (1992). List of vascular plants of the British Isles. B.S.B.1., London. LE SUEUR, F. (1982). The Jersey herbarium of Frére Louis-Arséne. Watsonia 14: 167-176. LE SUEUR, F. (1984). Flora of Jersey. Société Jersiaise, Jersey. MCCLINTOCK, D. (1975). The wild flowers of Guernsey. Collins, London. MARQUAND, E. D. (1892). The flora of Guernsey. Guernsey Society of Natural Science & Local Research report & transactions 189] 2: 129-151. MARSDEN, M. H. (1995). A new check list of flowering plants and ferns wild on Sark and its islets. Société Guernesiaise report & transactions 1994 23: 754-783. ROGERS, F. A. & ROGERS, W. M. (1898). On the Rubi and Rosae of the Channel Islands. Journal of botany 63: 13-15. ROGERS, W. M. (1899). The Rubi of Guernsey and Sark. Guernsey Society of Natural Science & Local Research report & transactions 1898 3: 296-297. ROGERS, W. M. (1900). Handbook of the British Rubi. Duckworth, London. SINEL, J. (1912). The geology of Jersey, with special reference to its stratigraphy and relation to the continental coast. J. T. Bigwood, Jersey. SNOW, B. & SNow, D. (1988). Birds and berries. A study of an ecological interaction. Poyser, Calton. VANNEROM, H. (1998). Wijzigingen aan de Rubus-inventaris (Sect. Corylifolii Lindl. en Sect. Rubus) in de derde editie van de ‘Flora van Belgie’. Dumortiera 70-71: 45-49. WATSON, W. C. R. (1958). Handbook of the Rubi of Great Britain and Ireland. Cambridge University Press, Cambridge. (Accepted June 2000) w 1 ' Watsonia 23: 437-442 (2001) 437 Conservation of Britain’s biodiversity: Rubus dasycoccus (Rosaceae), Thick-berried Bramble R. D. RANDALL and T. C. G. RICH Dept. of Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP ABSTRACT Rubus dasycoccus W. C. R. Watson is a rare British endemic species. Field and herbarium studies indicate at least 14 sites in three 10-km squares in v.cc. 34, 35 and 41, and many populations have fewer than ten plants. It is a perennial which reproduces by seed and appears to colonise new habitats readily. It occurs in a broad range of habitats and soils, but probably grows best in marginal habitats such as woodland edge and unkempt hedgerows on deeper brown earths. The main potential threats appear to be from inappropriate forestry and regular mowing of hedgerows or road verges, but its scattered distribution with broad ecological requirements indicate few significant threats to its long-term survival. KEYWORDS: Endemic, rare species, England, Wales. INTRODUCTION Rubus dasycoccus W. C. R. Watson, Thick-berried Bramble (from the Latin name dasy - thick, coccus - berry), is a rare British endemic species with a very localised distribution centred on Tintern and Trelleck in v.c. 35 Monmouth, with out-lying colonies at St Briavel’s and Bigsweir in v.c. 34 West Gloucester and Rudry, Rhydygwern and Llanfedw in v.c. 41 Glamorgan. Along with a number of other rare brambles, it has been included in the 3rd edition of the Vascular Plant Red Data Book (Wigginton 1999), but as yet little is known about its detailed distribution or ecology. In 1998 the information available about it was collated and combined with a field survey to establish its current status and determine its needs for conservation. A summary of the work is given below; full details are given in Randall & Rich (1999). In many respects R. dasycoccus is similar to R. trelleckensis Edees & Newton, Trelleck Bramble which grows in the same area (Randall & Rich 2000), but it has broader ecological tolerances, a wider distribution and would appear to be less threatened. TAXONOMY Rubus dasycoccus would probably not have received a name if it had not been distributed in the Set of British Rubi as supposedly representative of R. amplificatus Lees, a more widespread and superficially similar species. It was first collected near Tintern by A. Ley in 1891 and distributed through the Botanical Exchange Club as “R. montanus Wirtgen?”, where W. O. Focke suggested it might be a “var. of R. amplificatus Lees?” (Rogers 1892). Material collected in 1892 was then distributed in 1893 as No. 33 of the Set of British Rubi, labelled “R. macrophyllus Weihe & Nees var. amplificatus Lees”. Watson (1931) pointed out the characters which distinguished the Trelleck plant from R. amplificatus Lees and described it as R. lasiocarpus. Unfortunately this name was already in use and he was forced to allocate it a new name, R. dasycoccus (Watson 1933). R. dasycoccus 1s a distinct species and fits fairly well into its current position in Series Sylvatici, but there are a number of other species in Series Sylvatici and Series Rhamnifolii which resemble it to some degree. A guide to superficially similar species in South Wales and adjacent counties (v.cc. 34-36, 41-43) is given by Randall & Rich (1999). Distinctive characters of R. dasycoccus are the obovate leaflets with long cuspidate, often curved tip, cylindrical panicle, long patent sepals, rather pale pink petals, white filaments, green styles and the densely hairy carpels. No hybrids of R. dasycoccus are currently known, but in shaded habitats the fruits of R. dasycoccus are often poorly formed so plants could be easily confused with hybrids. 438 R. D. RANDALL AND T. C. G. RICH DISTRIBUTION Locality and habitat information was abstracted from the literature, and from herbarium sheets at the Natural History Museum (BM) and the National Museum of Wales (NMW) which between them include much Botanical Exchange Club material and the herbaria of W. M. Rogers, H. J. Riddelsdell and E. S. Edees (cf. Sewell et al. 2000). Because R. dasycoccus has been confused with R. amplificatus Lees, R. albionis W. C. R. Watson, R. questieri Lef. & P. J. Mull. and probably other species, earlier records cannot be accepted without voucher specimens. HISTORIC LOCALITIES V.C. 34, WEST GLOUCESTER Brockweir, near Tintern (BM). Bigsweir, 1894, Herb. W. A. Shoolbred, det. B. A. Miles (Newton 1986; specimen not located at BM or in herb. Shoolbred at NWW). V.C. 35, MONMOUTH Trelleck (BM). Beacon Hill, Trelleck (NMW). Trelleck to Tintern (BM, NMW),. Trelleck to Tintern (BM). Hedgerow between Tintern and Trelleck (NMW). Trelleck Hill, near Tintern (“Set No. 33”, BM, NMW). Bottom of Trelleck Hill, Tintern (Shade-grown specimen, “Set No. 33”, BM). Hedgerow, Catbrook Lane, near Tintern (NMW). Hedgerow, lower end of Catbrook Lane, near Tintern (BM). Catbrook road, near Tintern (BM). Open space in Wye Wood, Mon. (NMW). Wooded hill, Barbadoes hill, Tintern (NMW). Near Tintern (BM, NMW). V.C. 41, GLAMORGAN Forestry track, Coed Cefn-pwll-du (NMW). FIELD SURVEY 1998 The historic records indicate that R. dasycoccus is widely distributed in the Tintern and Trelleck area, and these and other suitable sites were consequently checked during the field survey. The survey was carried out between mid-July and early-September 1998, the best time being mid-July to early-August when plants were flowering. When estimating population sizes of Rubus dasycoccus it was not possible to give exact counts as plants spread vegetatively and were often densely entangled. For linear populations it was assumed that if the plant was found in two consecutive lengths of 2 m that one plant was found in each; in some cases this may result in an underestimate of the population size but it is unlikely that a serious overestimate has been made. Where the plant was found to colonize larger areas, as at Upper Hales Wood, one plant per 2 m x 2 m was assumed; this is also likely to be an underestimate but when a blanket cover occurs it is impossible to be certain where one plant ends and another begins. Populations were estimated using the following log scale where exact counts could not be made: A = 1-9 plants, B = 10-99 plants, C = 100-999 plants and D = 1000 or more plants. It should be stressed that all counts are crude estimates and should only be used to indicate the relative sizes of the populations. V.C. 34, WEST GLOUCESTER The Nedge, St Briavel’s (SO554039), one non-flowering plant was found on the path through woodland (A). V.C. 35, MONMOUTH Hale Wood; plants occurred in four areas of this wood which is conifer or mixed plantation at different stages of maturity. At Upper Hale Wood (SO512012), a population in excess of 2500 plants was found dominating the ground flora of the Picea plantation (D). In another section of the Plantation (SO517017) on a gentle slope near “The Oaks”, fewer than ten plants were seen scattered along the forestry tracks (A). At Bardadoes Hill (SO522007) more than 100 plants were present along track sides (C). On the minor road between Yew Tree House, Whitelye (SO513015) and Botany Bay (SO525019) several patches occurred scattered in hedges and along verges and walls (B). Botany Bay to Coed Beddick (SO525021 to SO526019), about 30 CONSERVATION OF RUBUS DASYCOCCUS 439 Vt tpl a 3 a FIGURE |. Current distribution of Rubus dasycoccus (¢). scattered plants were counted along a 300 m section of track through the plantation (B). Catbrook Lane (SO518025 to SO519024), scattered along 200 m of the public bridleway through the plantation off Catbrook Lane (B). Bargain Wood (SO520033), three clumps were found at the bottom of the slope on level ground in typical heathland community regenerating following tree felling and replanting (A). Roadside, Cicelyford to Cleddon (SO504038 to $0513038), scattered colonies were found along the roadside and in an adjacent forestry track (B). Parkhurst Rocks (SO501031), one small colony of about five patches was seen by the footpath through the ancient Quercus petraea woodland (Park House Wood S.S.S.I.) (A). Beacon Hill, Trelleck (SO512054), scattered patches occurred along the tracks and in recently cleared plantations (B). Beacon Hill - North Slope (SO513057), abundant on the lower forestry tracks but rare or absent from the upper heathy areas (B). Vicar’s Allotment, Trelleck (SO512060), thriving populations occurred adjacent to the footpath on the open heathland which until recently has been forestry plantation (B). Trelleck Hill near New House (SO504072), one young, non-flowering plant was found along c. 200 m of track through a dense plantation (A). V.C. 41, GLAMORGAN Llyn Hir/Coed Cefn-pwll-du (ST2086, ST2087, ST2187), up to 100 plants were present along rides and track sides through the plantation (B). Morphologically the plants differed from those in the Wye Valley with the leaflets here being mostly quite long and narrow and the stems and petioles generally more hairy, and with variable development of felting under the leaves. Some plants showed the typical broad-based, slightly curved reddish prickles but others had narrower straighter prickles and a darker stem. Cefn Mably Woods (ST225842), a vigorous colony was found along about 100 m of the drive to Cefn Mably Hospital (A), exhibiting the same variation as noted for Llyn Hir above. Rudry Common (ST192868), four bushes were found towards the top of the common where the bracken was not dense (A). 440 R. D. RANDALL AND T. C. G. RICH OTHER LOCATIONS SEARCHED A number of other locations were searched without success including v.c. 34 Plantation opposite Bearse Farm, St Briavel’s (SO573052), Hudnalls, St Briavel’s (SO5403) and St Briavel’s Common (SO5402), v.c. 35 Church Hill Common, Penallt, (SO5210) and New Wood, Fedw (ST4998 NE, ST5098 NW), and v.c. 41 Michaelston-y-Fedw (ST2484), Pensylvania (ST2585) and Pen-y-lan (ST2584). Rubus dasycoccus has thus been found in at least 14 sites in three 10-km squares in v.cc. 34, 35 and 41, and it may be more widespread. It was refound in all the historic localities with the exception of the West Gloucester sites whose exact location is not now known (and where it may still be present), and was found in several new localities. Many of the populations are small with fewer than ten plants, but some are also quite large. The current distribution map is given in Fig. 1. It appears that R. dasycoccus is a significant, if uncommon, element of the “Archenfield” regional flora (cf. Newton 1980). ERRONEOUS AND UNCONFIRMED RECORDS Watson (1931) suggested that Ley had collected R. dasycoccus at Aconbury Wood, v.c. 36 Hereford, in 1888 but this has not been confirmed and is not repeated in his subsequent handbook (Watson 1958). Watson (1958) listed R. dasycoccus for v.c. 46 Cardigan but the source has not been traced and an error is suspected. Watson (1958) stated that it was recorded from Vierset-Barse in Belgium under the name R. dumnoniensis var. amplificatus (Lees) Sudre, presumably in Sudre’s account of Belgian brambles (Sudre 1928-1929). There is a record for R. pyramidalis Kaltenb. subsp. amplificatus Lees in Legrain (1958), where he mentions it recorded for Vierset-Barse. This, no doubt, refers to the same plant, since in this work R. dumnoniensis is also relegated to a subspecies of R. pyramidalis. An error has been assumed. ECOLOGY LIFE CYCLE Rubus dasycoccus is, like all blackberries, a more or less evergreen perennial with some leaves often remaining until the new year. Panicles arise from buds in the axils of the previous year’s leaves as in other species and vegetative branches sometimes grow from the axils of the stronger stems in the first season. The plants are fairly bushy and so it seems likely that individual stems sometimes live longer than the usual two years. The flowering time is from mid-July to late August (exceptionally into September). Fruits were beginning to ripen at the end of August 1998 while the last flowers were still opening, but on many plants the seed set was rather uneven; whether this was due to environmental or other factors is not known. Dispersal of seed by birds and mammals probably occurs, the normal mode of dispersal for brambles. Vegetative spread by adventitious roots from the ends of the arching stems was noted but only in a few places did this result in extensive colonies developing, as at Upper Hale Wood where it has become the dominant ground cover. The occurrence of single vegetative plants in new sites suggests it has the potential to colonise further areas where the environment is suitable. HABITATS, VEGETATION AND SOILS Rubus dasycoccus was found in a broad range of habitats, including woodland, conifer plantations, woodland rides and tracks, hedges, heathland, roadside banks and verges. An attempt was made to relate the vegetation types of some of the sites to the National Vegetation Classification (Rodwell et al.1991 et seq.), but its occurrence in recently disturbed plantations, edge habitats and road sides has made this quite difficult (Table 1). The original vegetation types of the sites it now inhabits may have been W16 Quercus - Betula - Deschampsia woodland and W10 Quercus - Pteridium - Rubus community), and the H8e Calluna - Ulex gallii heath. Marginal habitats like woodland edge and unkempt hedgerows probably had the best-grown plants. The least: healthy plants occurred in the dry, exposed areas on the more acidic soils, and plants at Bargain Wood were suffering from Felt disease, which may indicate that conditions were CONSERVATION OF RUBUS DASYCOCCUS 44] TABLE 1. PROBABLE NATIONAL VEGETATION CLASSIFICATION TYPES AT SITES OF RUBUS DASYCOCCUS Site National Vegetation Classification types Botany Bay W25 Rubus - Pteridium scrub in hedge over-topped by Salix caprea and S. cinerea, woodland ride edges in W14 Fagus - Rubus woodland, W8 Fraxinus-Acer- Mercurialis woodland hedge. Bargain Wood H8e Calluna - Ulex gallii heath, Vaccinium sub-community developing in 5-year old cleared and replanted conifer plantation. Cicelyford to Cleddon, W23 Ulex-Rubus scrub grading into W16 Quercus - Betula - Deschampsia woodland roadside along roadside wall. Beacon Hill Open H8e Calluna - Ulex gallii heath, Vaccinium sub-community in recently cleared conifer plantation, and abundant in W16 Quercus - Betula - Deschampsia woodland on disturbed edge of possibly original W10 Quercus - Pteridium - Rubus woodland. Vicars Allotment H8e Calluna - Ulex gallii heath, Vaccinium sub-community developing in 5-year old cleared and replanted conifer plantation. Llyn Hir W24 Rubus - Holcus underscrub and W25 Rubus - Pteridium underscrub. Cefn Mably Woods W24 Rubus - Holcus underscrub. unfavourable. Plants growing under sunny but more sheltered and moister conditions were generally vigorous but a reasonable amount of shade was tolerated at Upper Hale Wood and Parkhurst Rocks. It is distributed on a range of soils, tolerating poor, dry heath soils but growing best on deeper brown earths (Table 2). With the exception of Llyn Hir where plants grow on Dolomitic limestone, all colonies grew in situations where the underlying geology involved sandstones. CONSERVATION REQUIREMENTS Rubus dasycoccus is certainly a rare plant. It is not protected under Schedule 8 of the Wildlife and Countryside Act 1981, but is listed in the Vascular Plant Red Data Book (Wigginton 1999) which should draw attention to its rarity, and this should be sufficient to ensure its survival. It occurs on two S.S.S.I.s (at Parkhurst Rocks and Cleddon Bog), though populations at both sites are small. If S.S.S.I.s need to be designated for this species, the largest site at Hale Wood and the type locality at Trelleck Hill are the best candidates. Given its widely scattered distribution with broad ecological requirements in a range of habitats, there appear to be relatively few significant immediate threats to its survival overall. The main potential threats to individual colonies appear to be change of land-use from forestry to farmland or recreation, weed killing during forestry operations, long periods of deep shade (e.g. colonisation TABLE 2. SOILS COLLECTED FROM THE RUBUS DASYCOCCUS ROOTING ZONES Site Soils Upper Hale Wood, Plantation: a gritty brown organic soil, pH 3-1. Recently disturbed ground: a gritty Whitelye dark-brown soil some organic matter and pebbles, pH 7-5. Botany Bay A fine, light-brown sandy soil, pH 7-0. Bargain Wood A fine grey-brown heathy soil with lots of organic matter, pH 4-0. Beacon Hill Top: A fine brown heathy soil with organic matter, pH 5-0. North side: A gritty light brown soil, pH 5-0. Track edge: A red-brown, gritty stony soil, pH 7-9. Vicars Allotment A blackish organic soil with large quartz sand grains, pH 3-4. Llyn Hir A reddish brown earth with many limestone fragments on track side, pHs 7:5 and 7-7. Cefn Mably Woods A red brown earth, pH 6-8. pH was measured in a 50:50 deionised water: soil slurry. 442 R. D. RANDALL AND T. C. G. RICH by Rhododendron, lack of clearance of conifers), and introduction of regular close cropping of paths, hedgerows and roadside scrub, or destruction of hedgerows due to road-widening or other schemes. The most favourable forestry management would be a regime similar to that at Beacon Hill, where trees have been cropped in rotation with adjacent areas being felled several years apart allowing colonisation into recently felled areas. For woodland that is being maintained as an amenity or a nature reserve, natural regeneration after cropping followed by selective thinning to produce small clearings and glades would be beneficial. For hedgerows and other marginal situations annual trimming should be avoided as it is detrimental as two or three years are required for stems to reach the flowering and fruiting stage. Where populations are significant, footpaths could be cleared on alternate sides in alternate years, or perhaps on both sides once every 3-4 years. A small number of fruits from four plants at Beacon Hill (2 populations) and Whitelye have been deposited in the Millennium Seed Bank at the Royal Botanic Gardens, Wakehurst Place for long-term storage; larger samples from a wider range of plants would be worthwhile collecting. No plants are known to be held in cultivation. ACKNOWLEDGMENTS This work was jointly funded by Countryside Council for Wales and National Museum & Gallery, Cardiff as part of a joint project into conservation and ecology of critical species. The Royal Botanic Gardens, Kew, contributed towards collecting blackberries for the Millennium Seed Bank. We would like to thank Andy Jones and Lindi Rich of CCW for their help, and Alan Orange and Jessica Carey for help with field work. We would like to thank the Keeper of BM for access to the herbarium. REFERENCES EDEES, E. S. & NEWTON, A. (1988). Brambles of the British Isles. Ray Society, London. LEGRAIN, J. (1958). Rubus L., in ROBYNS, W., ed. Flore Générale de Belgique 3: 69. Bruxelles. NEWTON, A. (1980). Progress in British Rubus studies. Watsonia 13: 35-40. NEWTON, A. (1986). Rubus L., in HOLLAND, S. C. ed. Supplement to the Flora of Gloucestershire. Grenfell Publications, Bristol. RANDALL, R. D. & RICH, T. C. G. (1999). Distribution and conservation of Rubus dasycoccus, Thick-berried Bramble. Countryside Council for Wales and National Museum of Wales, Cardiff. RANDALL, R. D. & RICH, T. C. G. (2000). Conservation of Britain’s biodiversity: Rubus trelleckensis (Rosaceae), Trelleck Bramble. Watsonia 23: 317-322. ROGERS, W. M. (1892). R. montanus Wirtg.? Rep. Botl. Exch. Club Br. Isles 1: 329. ROGERS, W. M. (1900). Handbook of British Rubi. London. SEWELL, R. M., EVANS, M., RANDALL, R. D. & RICH, T. C. G. (2000). Catalogue of Rubus specimens in the Welsh National Herbarium (NMW), National Museum & Gallery, Cardiff. National Museums & Galleries of Wales, Cardiff. WATSON, W. C. R. (1931). Bramble Notes, 1930. Rep. Botl. Exch. Club Br. Isles 9: 434-435. WATSON, W.C. R. (1933). R. dasycoccus nom. nov. Rep. Botl. Exch. Club Br. Isles 10: 21. WATSON, W. C. R. (1958). Handbook of the Rubi of Great Britain and Ireland. Cambridge University Press, Cambridge. WIGGINTON, M. J. (1999). British Red Data Books: Vascular Plants, ed. 3. J.N.C.C., Peterborough. (Accepted June 2000) Watsonia 23: 443-453 (2001) 443 Notes NITELLA GRACILIS (SMITH) AGARDH, AN ELUSIVE CHAROPHYTE NEW TO CARDIGANSHIRE (V.C. 46) The Slender Stonewort (Nitella gracilis) is, globally, a very widespread species, recorded from every continent except Antarctica. At least in Europe, however, it is very sparsely scattered, with only a few localities recorded in each country. This is also true for Britain and Ireland where it has been recorded from just a few, widely scattered sites in Sussex, Essex, Cornwall, Gwynedd, Cumbria, Ayrshire, Sutherland and Co. Wicklow, and not in some of these for over 50 years (Stewart in litt.) It is consequently listed in the Stoneworts British and Irish Red Data Book (Stewart & Church 1992), and as a Priority Species in the UK Biodiversity Action Plan (UKBAP; Anon 1995). The picture has, however, been complicated by a number of misidentifications, particularly for the similar — but more robust — N. mucronata (A. Braun) Miq. var. gracillima J. Groves & Bull.- Webst.; records from Hampshire, Greater London, Norfolk, Perth, and Co. Dublin are now thought to be doubtful or in error. This has confused the view of these species’ ecology, since Nitella mucronata var. gracillima grows in mesotrophic to eutrophic waters whilst N. gracilis appears to be restricted to nutrient-poor water bodies, particularly in the uplands (cf. Stewart & Church £9): Most records of Nitella gracilis have been from chance captures by grapnels or rare jetsam specimens after stormy weather and many of its sites have only yielded one or two specimens — although this could be due to collecting difficulties, seasonal growth or transient colonisation. Even so, there are good indications that the species has also disappeared from some sites (Stewart in litt.) For all these reasons there is little published information about the ecology of N. gracilis in Britain and Ireland and this short note aims to add to the current understanding, and to encourage further surveys of likely habitats. LLYN GYNON (V.C. 46) At an altitude of 425 m in the Elenydd uplands of Ceredigion and straddling two hectads (SN76— 86), Llyn Gynon is probably the richest lake in Cards (v.c. 46). Floating Water-plantain (Luronium natans) 1s locally abundant (Monteith, 1995), as has been Spring Quillwort (/soetes echinospora) in the past, and there is a small colony of Six-stamened Waterwort (Elatine hexandra) towards the north-western inflow. In 1964 Brian Seddon also found Pillwort (Pilularia globulifera) here at a depth of about 45-60 cm, by grapnelling from a boat, but it has not been refound since, despite diligent grapnel hauls from shore and boat, searches by wading, and searches of driftline debris (A. Chater, pers. comm.). Each of these species is nationally scarce; Pilularia and Luronium are priorities in the UKBAP and placed in the IUCN category “Vulnerable” in Europe and Luronium has statutory protection in national and international law (Schedule 8 of the Wildlife and Countryside Act; EC Habitats and Species Directive, Annexes IIb & IIv). As a consequence, the Countryside Council for Wales undertook a sample survey of the western side of the lake, on 25 September 1998, to map the distribution of important plant species and to search for deep-water colonies of Pilularia. The methodology followed from earlier, largely unpublished surveys for scarce aquatic macrophytes (Lomas et al. 1998; Scott 1996 & 1998), and included drysuits, snorkels and sub-aqua equipment. The survey gave a detailed description of the distribution of Luronium and Elatine and of their associated species, but did not find any Pilularia or Isoetes echinospora. Between 1964 and 1989 I. echinospora was found abundantly in Llyn Gynon and it seems from these and other observations (A. Chater, pers. comm.) that populations of the species may be subject to sudden and rather mysterious fluctuations. Where Isoetes lacustris dominance declined (at about 2-5 m depth) and with occasional plants of Myriophyllum alternifolium and Callitriche hamulata, vegetative L. natans (forma submersum) became the most frequent deep-water species (Fig. 1). Individual 444 NOTES Watsonia 23 (2001) KEY TO SPECIES Nitella gracilis ly \Z Juncus bulbosus Isoetes lacustris a Fontinalis antipyretica 5 va Ae Littorella uniflora Luronium natans ue oy Lobelia dortmanna ole ees Subularia aquatica Batrachospermum sp. L. natans (shallow water) Myriophyllum alternifolium f Porifera sp. (indet.) “Ke, Ranunculus flammula @ ss Stone Callitriche hamulata FIGURE 1. Llyn Gynon submerged vegetation. Diagrammatic section (western shore). NOTES Watsonia 23 (2001) 445 rosettes of L. natans with leaves 20-30 cm in length were scattered in deep, soft silt at a frequency of 4-10 plants per square metre. It was this association which yielded occasional plants of a Nitella sp., subsequently determined by Arthur Chater & N.F.S as N. gracilis. Five plants were noted by R.A.J. in a sample of approximately 30 m of “fringe” vegetation along the western shore of Llyn Gynon, at c. 3-5 m depth. Individual plants were c. 30-40 cm across, domed in profile and, compared to the associated species, notably free of silt deposition. Of the branches from two plants collected for identification, both turned out to have well-formed, orange-coloured oogonia, and both broke up substantially within 24 hours of collection, although the tougher, peat-stained, lower branches on one plant seemed more resilient. The location of N. gracilis in Llyn Gynon at 3-5 m depth on the western shore is distinctly sheltered, but it is not possible to generalise from this without further survey. Similarly, the fertility of plants so late in the year, their stained lower branches and absence of silt cover could all represent seasonal growth patterns (even, perhaps, perennation) but confirmation of this would require at least another visit. Other sources (e.g. John et al. 1982) have noted the value of charophytes as invertebrate habitat, and N. gracilis might be locally significant in some oligotrophic, acidic lakes for this reason. What can be concluded with confidence, however, is that surface-based methods of sampling (such as grapnel, grabs and bathyscopes) stand a low chance of detecting delicate species growing at such low density and at this depth and distance from the shore (see also Wade & Bowles 1981). This might also explain why attempts to relocate the plant by N.F.S. at its two other Welsh sites in Snowdonia (Llyn Dwythwch and Llynau Mymbyr) in October 1998 by shore grapnel surveys were unsuccessful (Stewart 1999). Nitella gracilis is likely to be an overlooked species of nutrient-poor, acidic lakes in Britain and Ireland and it will be located with certainty only by the use of trained divers. ACKNOWLEDGMENTS We should like to thank Mike Bailey, Alan Hale, Duncan Smith and Gabrielle Wyn for their help with the fieldwork, and Arthur Chater for his overall contribution and his comments on an earlier draft of this note. REFERENCES ANON (1995). Biodiversity: the UK Steering Group Report. Volume 2: Action plans. H.M.S.O., London. JOHN, D. M. , CHAMP, W. S. T. & Moore, J. A. (1982). The changing status of Characeae in four marl lakes in the Irish Midlands. Journal of life sciences of the Royal Dublin Society 4: 47-71. LOMAS, E., TEEARU, T. & ROWLANDS, A. (1998). Survey to determine the status of Luronium natans (L.) Raf. in a sample of six mid-Wales lakes. Unpublished report to the Countryside Council for Wales, Bangor. MONTEITH, D., ed. (1995). Integrated classification and assessment of lakes: Phase II — Final report. Unpublished report to the Countryside Council for Wales, Bangor. SCOTT, S. (1996). Action for Pillwort, Pilularia globulifera. Unpublished report to Scottish Natural Heritage / Plantlife, London. ScoTT, S. (1998). Partnership project for Action Plan work on Pillwort, Pilularia globulifera, 1997-2000. Annual Report 1997-8. Unpublished report to Scottish Natural Heritage / Plantlife, London. STEWART, N. F. & CHURCH, J. M. (1992). Red data books of Britain & Ireland: Stoneworts. Joint Nature Conservation Committee, Peterborough. STEWART, N. F. (1999). Welsh Stoneworts Survey 1998. Unpublished report to the Countryside Council for Wales, Aberystwyth. WADE, P. M. & BOWLES, F. (1981). A comparison of the efficiency of freshwater macrophyte surveys carried out from underwater with those from the shore or a boat. Progress in underwater science 6: 7-11. R. A. JONES Countryside Council for Wales, Plas Gogerddan, Aberystwyth, Ceredigion SY23 3EE N. F. STEWART Kingfishers, Cholwell, Posbury, Crediton, Devon, EX17 3QF 446 NOTES Watsonia 23 (2001) COLONISATION BY COCHLEARIA DANICA L. ALONG TRUNK ROADS IN CENTRAL SCOTLAND FROM 1996 TO 2000 The spread of halophytes along motorways and trunk roads in England and Wales in the 1980s and early 1990s was widely observed (Scott & Davison 1985; Scott 1985; Leach & Rich 1989; Leach 1994). The factors causing and influencing this spread have been much debated (Leach 1990: Scott 1990; Roper 1994), but it is generally held that coastal roads were colonised first and the halophytes were then dispersed by vehicular traffic along roadside habitats made suitable by road- salting (Scott & Davison 1985). For Cochlearia danica (Danish scurvygrass), rates of advance of 10-20 km yr" were deduced by Leach (1994) from observations on first colonisation. In Scotland C. danica became established on trunk roads much later than in England. The first record according to Leach (1994) was made in 1993 when plants were observed on the A74 in Dumfriesshire (v.c. 72). By this year much of the English motorway system had been colonised, and the species was absent inland in only six English vice-counties (Wight, E. Kent, E. Norfolk, Salop, Derbys, S.E. Yorks) (Leach 1994). As Cochlearia danica was frequent along the M6 through Cumberland (v.c. 70) in the early 1990s, it is very likely that the Dumfriesshire colony resulted from northward spread, rather than being a fresh colonisation from the nearby Solway coast. Similarly, the colonies of C. danica observed on the Al in Berwickshire (v.c. 81) in 1993 (Braithwaite 1997) probably arose from dispersal northwards from colonies along the Al in Northumberland (v.c.c. 67 and 68). From the mid 1990s onwards, I have made 4—6 journeys each spring between N.W. England and N.E. Scotland using the M74/A74/M73/A80/M80/M9/A9/A90 roads up to Stonehaven (Fig. 1). In 1995 and 1996 C. danica became increasingly obvious on the M74 and A74, and I reported what appeared to be its northern limit in 1997 near Hamilton (Welch & Welch 1998), there being a Stonehaven Laurencekirk @." ¢ 50 km oe Dunblane ve t Stirling je e?® ef e e FIGURE 1. Map of Central Scotland, showing the trunk roads monitored (dashed) and other nearby trunk roads (dotted). NOTES Watsonia 23 (2001) 447 sizeable patch on the M74 central reservation by the Atlas Works close to Junction 9 (Table 1). From 1997 onwards I have carefully looked for Cochlearia along the whole route north of Junction 9 of the M74, but driving at an average speed of 60-70 mph it is inevitable that small plants and non-flowering colonies will have been missed. In 1997, 1999 and 2000 I also drove along the M90 between Perth and the Forth Bridge (Fig. 1), and in 2000 I made a single traverse of the M876/M$9 triangle east of Stirling Services and the M8 from Newbridge to Newhouse. TABLE 1. LOCATIONS AND SIZE OF COLONIES OF COCHLEARIA DANICA ON CENTRAL SCOTTISH TRUNK ROADS Vice Road _ Place Grid square county 1997 1998 [O99 2000 A90__Laurencekirk NO77 91 4 ptchs” - ¢. 350 plis” 51 plts in60m* in57m* A90 Dundee Ring Road NO33 90 X X X 1 patch A90 1kmELongforgan NO33 89 Xx Ks X 1 patch A990 1kmWLongforgan NO22 89 x 1 patch 1 patch 1 patch (20m long) (35m long) A90 _ nr Inchture Jn NO22 89 X x x 2 ptchs A90 _— nr Inchmichael Jn NO22 89 X X 1 patch 3 ptchs A90 nr Leetown Jn NO22 89 % x X 1 patch A90__ nr Inchyra Jn NO12 89 X few ptchs x X M90 around J9 NO11 88 x . sev. ptchs _c. 250 m M90 1kmNJn4 NT19 85 x - x few ptchs” M90 lkmS Jn3 NT18 85 x x 1 patch A90 ‘just N of Forth Bridge NT18 85 X - sev. ptchs _ sev. ptchs A9 nr Findo Gask Jn NOO1 88 X X few ptchs few ptchs A9 nr B9141 Jn NN91 88 Xx X few ptchs few ptchs A9 Blackford bypass NN90 87 X x X 2 ptchs A9 3 km W Blackford NN80 87 x X 1 patch 1 patch (20 m long) A9 5 km W Blackford NN80 87 x x X 2 ptchs M9 =- Jn 11-Jn10 NS79 87 m8 X few ptchs occas. M9 ~—=- Jn10—Jn9 NS78,79 86 x x frequent frequent M9 = Jn9-Jn7 NS88 86 - - - occas. M876 Jn8—Jn5 NS88 86 . - - occas. M876 Jn7—W of Jn3 NS88 86 - - - occas. M876 _ near Jn3 NS98 86 - . . few ptchs M80 = _-Jn9-Jn5 NS88 86 x X frequent frequent M/A80 Jn5—Jn3 NS77 86 X few ptchs _—_ frequent frequent M73 = Jn3-Jn1 NS76,77 ie, x frequent occas. occas. M8 near Jn2 NN17 83 . - - frequent M8 ~~ W of Jn2-Jn3A NN06,07 84 . . - frequent M8 = Jn3A—Harthill Services NS96 84 - - occas. M74 Jn4-Jn5 NS66;75,76 77 X Xx few ptchs _—_ few ptchs M74 Jn5-Jn8 NS74,75 Woh, x few ptchs occas. frequent M74 Jn8—Jn9 NS74 Tal Xx few ptchs _ occas. occas. M74 — around Jn9 NS74 77 few ptchs occas. frequent frequent Y= on verge or hard shoulder; other colonies are on the central reservation; - = not searched; x = not seen; Jn = Junction; ptchs = patches; frequent = c. 10% of central reservation occupied; occas. = c. 1% of central reservation occupied; sev. patches = >5 patches observed; few patches = 2-5 patches observed. 448 NOTES Watsonia 23 (2001) The direct route between Junction 9 of the M74 and Laurencekirk on the A90 is 225 km long, and C. danica colonised most of this between 1996 and 2000 (Table 1). In 1998 and 1999 Cochlearia was much more frequent along the southernmost 80 km of this route up to Dunblane than on the next 80 km from Dunblane to Dundee, and its abundance on the M73 in 1998 suggested it had been overlooked there in 1997. However, Macpherson & Macpherson (1999), recording C. danica along Lanarkshire (v.c. 77) roads on a 1-km square basis, also made many new records in 1998, three of their total of 17 records being from the M73 and ten from the M74. Important factors affecting the presence and visibility of C. danica along roads are the laying of gravel, the spraying of herbicides, and competition from dense grass swards. Fresh gravel laid along the M73 central reservation in 1996 or 1997 probably contributed to the abundance of Cochlearia there in 1998, whereas its apparent depletion in 1999 (Table 1) may have been related to grass and coarse plants spreading. Herbicide treatment along most of the route across central Scotland up to Dundee in spring 1999, which became obvious in mid May, much reduced grasses and weeds, but had only short-term impact on C. danica judging from its distribution and frequency in spring 2000. It has been suggested that C. danica is actually introduced to inland roads in the gravel and hard core used in their construction (Coombe 1994; Lansdown & Pankhurst 1995), and this is a possible explanation for the phenomenal increase of the species in central Scotland. A pinky-red gravel has been placed in the central reservation on many sections of the M73, the A80, the M80, the M876, the M9 and the M90, and possibly this gravel contained Cochlearia seeds. However, I doubt this explanation because there are long sections of red gravel with no Cochlearia plants visible especially between Stirling and Dundee (Fig. 1). Also the colony on the M90 near the Inchyra junction (Table 1) was lost when a dressing of red gravel was laid in summer 1998 after the erection of a crash barrier. The Laurencekirk colony of C. danica on the A90 at the far north of the monitored roads perhaps did not arise from colonisation produced by vehicles sweeping seeds along the carriageways. The four small patches observed in 1997 actually occurred on the nearside verges of the two carriageways about 3 km apart, and could have resulted from introduction in road-making materials besides transport on vehicles. Similarly, the colony of C. danica observed on a roadside at Flemington (v.c. 77) in 1989 (Macpherson & Macpherson 1999) would seem not to be part of the trunk-road invasion. The potential rate of increase of C. danica was shown at the Laurencekirk colony between 1997 and 1999 (Table 1). Two patches of c. 400 cm’ on the southbound verge in 1997 had become c. 350 scattered plants in 1999 over a 60 m section of verge extending 0-3 m from the road kerb. Some disturbance appears to have occurred, as some stiff red clay was exposed in spring 1999, perhaps from skimming during grass-cutting in the previous summers. However, drain digging during 1999 c. 4 m from the kerb, and intermittent parking, compacted the verge soil and damaged the plants, and by spring 2000 the colony was reduced to 51 plants along a 57 m length 0-2 m from the road kerb. Also, the two patches observed on the northbound verge could not be refound in 1999 or 2000, nor any plants of Cochlearia in the 200 m of verge around their position. The rate of spread of C. danica across central Scotland, calculated on the basis of strong highly visual patches at the north ends of fairly continuous colonisation (Atlas Works on M74 in 1997, J10 on M9 in 1999), is 40 km yr'!. This is considerably greater than Leach*s estimate of 10-20 km yr'', and perhaps the 1997 start point was actually further north than the Atlas Works. This would in turn mean an advance north up the A74/M74 from 1993 to 1997 at more than 20 km yr’. Why C. danica was so slow to colonise Scottish motorways and trunk roads is puzzling. The suggestion that less salt was being used in Scotland than northern England in the 1980s (Scott 1985; Scott & Davison 1985) is unlikely, given the climate; the success of C. danica in the last few years indicates that roadside conditions had become very suitable by the 1990s. I believe that for successful colonisation the quantity of seeds being dispersed has to be great, which needs both large fruiting colonies and much traffic moving seeds along, and also the receiving habitat has to be extensive and fairly continuous. Probably other local establishments, like the Laurencekirk colony, have occurred unnoticed in Scotland, and have not produced enough seed for nearby suitable habitat to be colonised before becoming extinct due to grass competition, road works, herbicide treatment, etc. The great build-up of C. danica on the trunk roads leading into Scotland by the early 1990s, the increased traffic density compared to the 1980s, and the greater extent of NOTES Watsonia 23 (2001) 449 suitable habitat in the central reservation (bare gravel now being preferred to grass because maintenance is easier), have combined to produce sufficient seed dispersal to create the present spectacular carpets of pale pink flowers across central Scotland in spring. ACKNOWLEDGMENTS My wife Muriel has helped in the spotting and recording of roadside colonies. Peter Macpherson has kindly given me details of his Lanarkshire observations. REFERENCES BRAITHWAITE, M. E. (1997). Maritime plants on roads in the Scottish Borders. B.S.B.J. Scottish newsletter 19: 15. COOMBE, D. E. (1994). “Maritime” plants of roads in Cambridgeshire (v.c. 29). Nature in Cambridgeshire 36: 37-60. LANSDOWN, R. & PANKHURST, T. (1995). Coastal gravel as a source of Cochlearia danica on inland roadsides. BSBI news 69: 17-18. LEACH, S, J. (1990). Cochlearia danica on inland roadsides. BSBI news 56: 20-21. LEACH, S. J. (1994). Cochlearia danica on inland roadsides — an update. BSBI news 65: 12-13. LEACH, S. J. & RICH, T. (1989). Scurvy-grasses take to the road. BSBI news 52: 15-16. MACPHERSON, P. & MACPHERSON, A. C. (1999). Danish scurvygrass — another seaside-roadside plant in Lanarkshire. B.S.B.I. Scottish newsletter 21: 11. ROPER, P. (1994). Motorway scurvygrass. BSBI news 65: 13. ScoTT, N. E. (1985). The updated distribution of maritime species on British roadsides. Watsonia 15: 381-— 386. ScoTT, N. E. (1990). Cochlearia danica on inland roadsides — source of seed? BSBI news 56: 11-12. ScoTT, N. E. & DAVISON, A. W. (1985). The distribution and ecology of coastal species on roadsides. Vegetatio 62: 433-440. WELCH, D. & WELCH, M. J. (1998). Colonisation by Cochlearia officinalis L. and other halophytes on the Aberdeen — Montrose main road in North-east Scotland. Watsonia 22: 190-193. D. WELCH East Fernbank, Woodside Road, Banchory, Kincardineshire AB31 5XL ALCHEMILLA GLAUCESCENS WALLR. IN V.C. 81 BERWICKSHIRE Alchemilla glaucescens has its headquarters in Britain on the limestone of the Ingleborough district of Yorkshire but has small disjunct populations in Northern Scotland and Ireland. It is a Red Data Book species. Two populations have now been found in Berwickshire. LAMBERTON On 18 September 1999 M.E.B. was studying Rosa in scrub on the steep sea braes at Hilton Bay NT9659 within the Lamberton Coast S.S.S.I. An unexpected open area was reached which carries a limestone flora with several species of note. An interesting Alchemilla was collected in fruit, suspected to be A. glaucescens. The site was revisited by M.E.B. with P. S. Lusby on 16 October 1999 and again by M.E.B. on 1 May 2000 when the Alchemilla was in full flower. P. S. Lusby and Dr S. M. Walters confirmed the determination from the autumn plants, but with mild reservations. The spring growth allowed fresh specimens to be presented to P. S. Lusby whose reservations were dispelled. The Alchemilla is closely associated with Sanguisorba minor subsp. minor, Primula veris and Leontodon saxatilis. The grassland is also notable for large populations of Carlina vulgaris and Catapodium rigidum. Other associates include Anthyllis vulneraria, Carex flacca, Leontodon hispidus, Linum catharticum, Ononis repens and Viola hirta. The Alchemilla certainly seems to be native, though, with the railway adjacent, introduction cannot be wholly discounted. The population is estimated at 500 plants. The site has gradual erosion that keeps the vegetation open to some degree. The substratum of Hilton Bay is Carboniferous limestone and sandstone but there are also igneous porphyritic dykes near the site (Greig 1988). 450 NOTES Watsonia 23 (2001) Although the bay is secluded and not overgrazed there is one major threat. The very erosion that the existence of the vegetation community depends on is a threat to the main North-South railway nearby. British Rail (now Railtrack) have fairly recently carried out works to halt the erosion as much as possible. Concrete structures have been erected at the edge of the beach and plastic mesh has been dug into the slopes above. The concrete has holes in it and these have colonised splendidly with such species as Carlina and Catapodium. However, the area with the mesh is now covered by quite coarse vegetation and if any of the rarities were formerly present there they have probably been lost. It should be noted that large areas of the braes are not strongly calciferous and are covered naturally by coarse vegetation - indeed that is partly why these open communities have not been discovered before - it is not obvious that this is the most interesting part of the braes. Adjacent coastline was searched for the plant on 27 September 1999, 16 October 1999 and 1 May 2000 but no further colonies have been found. CHIRNSIDE Alchemilla glaucescens had earlier been discovered by M.E.B. in 1982 in a disused railway cutting near Oldcastles, Chirnside, NT8558. It was described by him in ‘The Botanist in Berwickshire’, (Braithwaite & Long 1990), as growing “on ballast” and treated as an introduction. This statement is not accurate. The site was visited by M.E.B., P. F. Braithwaite and L. Gaskell on 6 June 1998. The Alchemilla grows on a bank that appears to be part of the original hillside and to have been preserved by being enclosed within the former railway fence. The cutting as such is below. The plants are not growing on ballast and there is no evidence of ballast being used in the drainage of that particular section of cutting. There is no reference to ballast in M.E.B.’s field notes in 1982. M.E.B. had found interesting communities on ballast-drained railway cuttings in Roxburghshire (Braithwaite 1976). The Alchemilla is associated with depauperate base-rich grassland and is closely associated with Trifolium medium. Other associates are Crepis capillaris, Fragaria vesca, Luzula multiflora, Primula vulgaris, Succisa pratensis and Vicia sativa subsp. nigra. Linum catharticum and Trisetum flavescens were recorded in 1982. This site must now be reconsidered. The Alchemilla is in all probability native here also. The Alchemilla population is healthy with about 100 plants, the determination of which has been checked by Dr S. M. Walters in 1982, and P. S. Lusby in 1999. The substratum is “unusually coarse-grained red sandstone” of the Upper Old Red Sandstone (Greig 1988). The vegetation is kept open by cattle grazing, which appears to be beneficial. Nevertheless the site is so small that the future of the Alchemilla is most doubtful. DISCUSSION The locally scarce limestone species with which the Alchemilla is associated at Hilton Bay can all be presumed to have reached the Scottish Borders by a fairly clear-cut path from the limestone areas of Northumberland where they are more frequent (Swan 1993). Some of them have other stations in the Scottish Borders on the Silurian and on the Old Red Sandstone where there are lime-rich pockets. There is no reason why the Alchemilla should not have colonised by a similar route but, if so, no ready explanation can be offered as to why it is not found between Berwickshire and the Ingleborough area today. It seems therefore as likely that it is a chance introduction in the distant past, perhaps by bird-borne seed from Scandinavia, as the site is very much on a bird migration route. However, Tofieldia pusilla was found near Berwick by John Ray in 1671 (Ray 1677), probably within two km of the Hilton Bay site, and it is possible that the former moorland of this area held, in limestone pockets, refugia for a specialised post-glacial community that included the Alchemilla and Tofieldia. REFERENCES BRAITHWAITE, M. E. (1976) A Railway Flora of Teviotdale. Hawick BRAITHWAITE, M. E. & LONG, D. G. (1990) The Botanist in Berwickshire. The Berwickshire Naturalists Club, Buccleuch Printers, Hawick. NOTES Watsonia 23 (2001) 451 GREIG, D. C. (1988) Geology of the Eyemouth District. British Geological Survey, H.M.S.O. London. Ray, J. (1677) Catalogus Plantarum Angliae (2nd ed.). London. SWAN, G. A. (1993) Flora of Northumberland. The Natural History Society of Northumberland, Newcastle upon Tyne. M. E. BRAITHWAITE Clarilaw, Hawick, Roxburghshire, TD9 SPT GERANIUM ROBERTIANUM L. VAR. MARITIMUM (GERANIACEAE): AN EARLIER AUTHOR For a century and a half the authorship of this name for the prostrate and often glabrous coastal ecotype of G. robertianum has been attributed to Babington, who first published it in the third edition (1851) of his Manual of British botany. It has, however, been overlooked that the same epithet was validly published for the very same plant by Cooper (1834), whose authorship has priority. Cooper cited only the one locality: Cockbush Common, Shoreham (W. Sussex, v.c. 13). There is a specimen of the plant in CGE collected at Cockbush Common the previous year by G. E. Smith, by whom and, even more, Borrer (the two leading Sussex botanists of that period) the inexperienced Cooper was provided with most of the localised records that featured in his list — a fact which Cooper concealed from the public, to the lasting indignation of Borrer (1851; Allen 1979). It is thus probable that the name was a manuscript one already in use by one or other of those two. Babington may well have been apprised of it later by Borrer, who may by then have forgotten (or wanted to forget) that it had appeared in print already. Baker (1956) considered that the rather distinctive plant at Shoreham (which was also the type locality of Babington’s taxon) represented just one extreme in a cline of hairiness, allowing all the prostrate plants of shingle beaches to be grouped together under a single name. Because these have a wide geographical distribution and he had detected in cultivation a physiological difference in addition to a loose cluster of external characters, he felt justified in promoting the variety to subspecific rank. In this he has been followed by subsequent authors, though it might perhaps be questioned whether an ecotype with, as pointed out by Stace (1991), only one constant external character really merits such an elevated rank. REFERENCES ALLEN, D. E. (1979). The plagiarisms of Thomas Henry Cooper. Journal of the Society for the Bibliography of Natural History 9: 275-279. BAKER, H. G. (1956). Geranium purpureum Vill. and G. robertianum L. in the British flora. Il. Geranium robertianum. Watsonia 3: 270-279. BORRER, W. (1851). Botanical memoranda. Botanical gazette 3: 98. CoopER, T. H. (1834). The botany of the county of Sussex. Sussex Press, Lewes. (Preprint of Cooper in HORSFIELD, T. W. (1835). The history, antiquities and topography of the county of Sussex, 2: 5-22). STACE, C. A. (1991). New Flora of the British Isles. Cambridge University Press, Cambridge. DD. E: ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ 452 NOTES Watsonia 23 (2001) EPILOBIUM x KITCHENERI McKEAN (ONAGRACEAE) IN EAST CORNWALL The hybrid between Epilobium pedunculare A. Cunn. and E. montanum L. has hitherto been reported only as a single plant found in West Perth (v.c. 87) in 1996, from which it was named as E. x kitcheneri by McKean (1999). This is the sole report from the British Isles of a hybrid involving the alien Rockery Willowherb E. pedunculare from New Zealand, which is only locally established in the wild in the British Isles, whereas the Broad-leaved Willowherb E. montanum is a widespread native piant that forms hybrids with numerous congeners (Stace 1997; Kitchener & McKean 1998). In 1996 Mary and Tony Atkinson discovered that E. pedunculare was naturalised in abundance alongside tracks through Hancock’s Wood, a coniferous plantation on the northern side of the Glynn Valley, East Cornwall (v.c. 2) (Murphy 1998). During 1999 we searched independently for possible hybrid Epilobium in the vicinity of the naturalised E. pedunculare in the Glynn Valley. These searches resulted in the discovery of two plants of E. x kitcheneri, providing the second and third records for this hybrid and the first reports of it outside Scotland. The first find of the hybrid in East Cornwall was made on 9 July 1999 by D.T.H., in woodland north-east of Trago Mills, grid reference SX182648; specimen in herb. D.T.H. The single plant was growing on the rocky edge of a track in a coniferous plantation, close to E. ciliatum Raf., E. montanum, E. obscurum Schreb. and E. pedunculare, the last of these species being naturalised along c. 150 m of the track edges. The hybrid had two main stems, the longest of them 28 cm tall and stiffly erect above its procumbent base, the other stem 27 cm tall with two shorter branches, both stem and branches being erect throughout. The specimen accords well with the Scottish type material of E. x kitcheneri in having leaf shape midway between the orbicular leaves of E. pedunculare and the ovate leaves of E. montanum, the influence of E. pedunculare apparent in strongly toothed leaf margins and the leaves bronzed on the underside, and the influence of E. montanum apparent in the presence of a shallowly four-lobed stigma. The second find of the hybrid in East Cornwall was made a month later, at Hancock’s Wood just over | km away (SX171649). It was found by Ian Bennallick in company with Mary and Tony Atkinson and G.D.K. The hybrid plant was also found at the edge of a woodland track with the same Epilobium associates as the first find. It consisted of three closely associated, separately rooted stems, 41 cm, 35 cm and 20 cm long. These may have been separate plants or the results of layering of one original plant which may have overwintered. Subsequent cultivation by G.D.K. has demonstrated that overwintering can take place, with rooting from the nodes. The north side of the Glynn Valley in East Cornwall apparently has the most extensive naturalised colonies of E. pedunculare in the British Isles. The species extends in quantity along more than a kilometre of forestry tracks in Hancock’s Wood and Cross Plantation. The means by which it was originally introduced there is uncertain, although the adjoining woodlands have some exotic plantings, including a bamboo. There is evidence that forestry operations have contributed to its subsequent spread along tracks. In the Glynn Valley E. pedunculare mostly grows in damp, shaded situations alongside woodland tracks, including ditches and moist banks shaded by coniferous trees. In August 1999 it was noticed that it was often dried up and perished where it was growing in the more open parts of tracks in Hancock’s Wood and Cross Plantation. This preference for moist shaded sites shown by E. pedunculare in Cornwall corresponds to its predilection for moist shaded banks in Nothofagus forests in New Zealand (Raven & Raven 1976). Although E. pedunculare is extensively naturalised in the Glynn Valley and it often grows alongside native Epilobium species there, a thorough search revealed only one other putative, but unidentifiable, hybrid with E. pedunculare parentage. The species grow intermixed on track edges with partial shade, but the damp and shade favoured by E. pedunculare are circumstances which have been found to delay opening of the flowers in Epilobium (Brockie 1959). This increases the likelihood of self-fertilisation, and therefore diminishes the opportunities for hybridisation. ACKNOWLEDGMENTS We are grateful to Mary and Tony Atkinson for information on the whereabouts of Epilobium NOTES Watsonia 23 (2001) 453 pedunculare in East Cornwall. They also participated, along with Ian Bennallick, in a search for Epilobium hybrids that resulted in the second find of E. x kitcheneri in Cornwall. REFERENCES BROCKIE, W.B. (1959). Breeding systems in New Zealand species of Epilobium L. Transactions of the Royal Society of New Zealand 87: 189-194. KITCHENER, G. D. & MCKEAN, D. R. (1998). Hybrids of Epilobium brunnescens (Cockayne) Raven & Engelhorn (Onagraceae) and their occurrence in the British Isles. Watsonia 22: 49-60. MCKEAN, D. R. (1999). A new Epilobium hybrid from Scotland, FE. pedunculare A. Cunn. x E. montanum L. Watsonia 22: 417-419. MurPHY, R. J. (1998). Plant notes and records. Botanical Cornwall 8: 47-86. RAVEN, P. H. & RAVEN, T. E. (1976). The Genus Epilobium (Onagraceae) in Australasia: a systematic and evolutionary study. New Zealand Department of Scientific and Industrial Research Bulletin 216, Christchurch, New Zealand. STACE, C. A. (1997). New Flora of the British Isles. 2nd ed. Cambridge University Press, Cambridge. D. T. HOLYOAK & Edward Street, Tuckingmill, Camborne, Cornwall, TR14 SPA G. D. KITCHENER Crown Villa, Otford Lane, Halstead, Sevenoaks, Kent, TN14 7EA i > . - ‘ a Watsonia 23: 455461 (2001) 455 Book Reviews Ireland. A Natural History. David Cabot. Pp. 512, with over 230 illustrations. Harper Collins, London, 1999. Hbk £34-99. ISBN 000-—220079-1. Pbk £17.99. ISBN 000—220080-5. This book has been much needed. Ireland, always in the news, is a source of intense fascination to outsiders, not least naturalists and botanists. This splendid addition to the New Naturalist Library, not so much introduction as comprehensive synthesis, is a fact-packed guide to the natural wonders of our smaller, enigmatic western partner. A friend, quizzing me about Ireland, observed dryly “They tell me it is very green”. Yes, for here is an Atlantic land of water — soft rainy days, lakes, rivers, streams, boglands, beaches and lush pastures. Conversely some of Ireland’s richest habitats are the (relatively) dry islands of glacial eskers, sand-dunes, cliffs and, above all, limestone pavements. Natural history in Ireland is all about the ecology of animals and plants at the furthest north- western fringe of Europe, made the more exciting by millennia of isolation. For the visitor, or a naturalised denizen like David Cabot, Ireland is irresistible, not least for plentiful opportunities to compare and contrast with Britain. The author’s Preface, which tells how he was instantly grabbed by the sight of unfamiliar hooded crows, and much else beside, has a familiar ring (in my case it was Brent geese feeding near my tatty Dublin lodgings). Cabot is primarily a zoologist but his interests range far and wide. The book starts from a historical perspective, taking us from the Christian saints and scholars of the Dark Ages through bemused 17th century English visitors and invaders, the confident and learned world of 18th century Georgian Dublin and the 19th—early 20th century heyday of Irish natural history, to more recently departed polymath giants like Robert Lloyd Praeger (1865-1953) and the B.S.B.I.’s own David Allardice Webb (1912-95). The author notes the remarkable upsurge in Irish plant and animal recording during the last 25 years, although he could perhaps have taken a littke more account of recent floristic literature and so many exciting discoveries such as Trifolium occidentale. He then talks us skilfully through the biological history and chapters on the vegetation, flora and fauna: mountains and uplands, peatlands, lakes, rivers, the limestone pavements and turloughs, the (sadly depleted) woodlands, the farm landscape,and the coastline and Surrounding seas. The book covers many well-known botanical areas, notably the legendary Burren landscape and the Killarney oakwoods, and equally fascinating but less publicised treasures such as the Scragh Bog fenland in Westmeath, the species-rich mineral flush in Bellacorrick Bog in West Mayo, Uragh Wood in Co. Kerry (with its Kerry spotted slugs), the precipitous arctic-alpine plant communities of Benbulbin, Co. Sligo, and the extensive sand-dunes and machair of the north- western coasts. The text also covers several of those individual Irish plants such as Strawberry Tree, Arbutus unedo; Cottonweed, Otanthus maritimus (on the Co. Wexford coast) or the special heaths of Connemara, that are absent — or largely so — from Britain, and so fascinate natives and visitors. The chapter on mountains looks at the conundrum whereby Ireland has a relatively poor mountain flora, but one of huge botanical interest, with arctic-alpines frequently descending to sea- level. The last chapter is devoted to conservation, for a long time haphazard in Ireland but now given strength by EU law. Appendices at the end of the book list, with brief notes on interest and location, the nature reserves in the Republic of Ireland and Northern Ireland. Although the two political regions of the island have by necessity had separate statutory bodies for conservation, it is to the credit of the natural history community that it has never established any cross-border or sectarian division. Here is a fine example of the high quality scientific natural history writing that is at the heart of the best New Naturalist volumes. For the British botanist keen to know Ireland, I can think of few better prescriptions than to read and digest Cabot’s book, browse through the classic works of Praeger and, while you are at it, take out a subscription for Irish Naturalists’ Journal to keep the story updated. J. R. AKEROYD 456 BOOK REVIEWS Flora Nordica. Volume I. Edited by B. Jonsell. Pp. xxii + 344. The Bergius Foundation, the Royal Swedish Academy of Sciences, Stockholm. 2000. €60. ISBN 97 7190 033 0. The Flora Nordica Project was initiated in 1987 to produce “‘a comprehensive, diagnostic Flora of high scientific standard covering the vascular plants”, some 4,600 altogether, of Denmark, Finland, Iceland, Norway and Sweden, including the Faeroes, Jan Mayen, Bear Island and Spitsbergen. The chairman and editor-in-chief is Bengt Jonsell, well known in the B.S.B.I. and one of our Honorary Members. The first of the expected eight or so volumes is now published, covering Lycopodiaceae to Polygonaceae in Flora Europaea order. All but one of the 31 authors are from the Nordic countries. A good flora, and this is clearly a very good one, can only be properly assessed after years of constant use, so rather than writing a review I will attempt to indicate how useful this book is likely to be to botanists in Britain and Ireland. Flora Nordica is written in English. It gives practical and concise dichotomous keys, plenty of synonyms which are in most cases adequate for our purposes, vernacular names where these are already available in the Nordic languages, and indication of types for all names based on Nordic material. The good and clear descriptions are about as long as those in Sell & Murrell’s Flora of Great Britain and Ireland. Distribution and habitat are given in considerable detail, and most species are mapped by provinces. There are paragraphs on Biology, Variation, Hybridization and Similar Taxa, and there are numerous line drawings, mostly of diagnostic features. Chromosome numbers are included, with the origin of those based on Nordic material being given. Subspecies are given full treatment, and varieties are included though these do not get much of a showing in the present volume. An outstanding feature is that all hybrids are described or at least diagnosed. Casuals are treated in reduced typeface and format depending on their frequency, and this leads to a rather fussy appearance and, for British readers, a sometimes annoying inconsistency of descriptive information. Flora Nordica thus provides a great deal more information, both in kind and in depth, than any modern British Flora. Volume | turns out to be very applicable to the British flora. In the families covered, 200 out of the 270 species in Stace are also in Flora Nordica, 1.e. 74%. (Conversely 200 out of the 288 species in Flora Nordica are also in Stace, i.e. 70%.) The accounts of the two largest families serve us particularly well, 90% of the Stace species in the Salicaceae being covered, and 81% in the Polygonaceae. In both ferns and conifers 69% are covered. In Salix, Flora Nordica describes no fewer than 111 hybrids, while Stace has 67 and describes only 15. Probably no other foreign Flora is for us so relevant and worth studying. For its descriptive treatment of hybrids alone it is unique, and in the paragraphs on biology, habitat and variation in particular there is a vast amount of information otherwise available, if at all, only widely dispersed in the literature. Although they are working from opposite ends of the classification and direct comparison is impossible, on the evidence of the available volumes Flora Nordica and Sell & Murrell provide similar amounts of description and infraspecific taxonomy, while the former provides a great deal more other information. The taxonomy is generally familiar, and where unfamiliar it is usually enlightening. For example, Polygonum aviculare has six subspecies, including our P. boreale, P. rurivagum and P. arenastrum, surely a more reasonable solution than trying to separate them absolutely as species. On the other hand, the subspecies of Rumex crispus are not recognised, and it is suggested that part of the variation may be due to introgression with other species. In the very thorough account of Salix, S. fragilis is very narrowly defined, with the material with hairs on the buds and young leaves being ascribed to S. x rubens, the hybrid with S. alba. (In one of the few conspicuous errors in the book, S x rubens, though included in the vegetative key, is omitted from the key to material with female flowers.) Quite extensive essays are devoted to some of the major taxonomic problems, such as in Betula pubescens and B. tortuosa, Rumex acetosella and Salix repens. There are numerous instances where one is stimulated to apply characters and ideas to our own plants. Is it really true that anther-length provides an absolute distinction between Rumex acetosella subsp. acetosella (1-2—1-5 mm) and subsp. angiocarpus (0-8-1 mm)? If so, it would be very useful in this dioecious species, but the occurrence of mixed populations makes it difficult to check with certainty. As Flora Europaea neared completion in the mid 1970s, a group of British botanists under David Valentine began planning a “Critical Flora of Great Britain and Ireland”. This envisaged a ee BOOK REVIEWS 457 work very much along the lines of Flora Nordica, but it never got beyond the planning stage, partly for lack of funding. Much of the momentum went instead into the Floras by Stace on the one hand and Sell & Murrell on the other, with Clapham, Tutin and Moore occupying the middle ground. We must congratulate the Royal Swedish Academy of Sciences for supporting such an uncompromisingly thorough international project as Flora Nordica through the Bergius Foundation, and our Nordic colleagues for bringing this first volume to completion. Anyone seriously interested in the British flora will find it both stimulating and very useful, and will be consumed by envy. A. O. CHATER William Turner. Libellus de re herbaria novus 1538. Edited with a translation into English by Mats Rydén, Hans Helander & Kerstin Olsson. Acta Societatis Litterarum Humaniorum Regiae Upsaliensis 50 (ISSN 0280-0918). Pp. 146. Swedish Science Press, P.O. Box 118, SE-751 04 Uppsala, Sweden. 1999. Paperback with dustcover, SEK150 (+ postage and handling). ISBN 91- 630—8620-4. “The chief purpose of the present volume,” begins the preface to this simple but scholarly and elegantly produced book, “is to make the whole of William Turner’s Libellus de re herbaria novus (1538), the first scientific botanical treatise published in England, available in English, with historical, textual, linguistic and botanical commentaries.” For this, the team who produced it deserves the gratitude of anyone interested in the history of British botany, and particularly of our plant names. William Stearn’s lavish edition of Libellus and The names of herbes (1548), published by the Ray Society in 1965, provided facsimiles of the two texts and revised versions of indexes identifying Turner’s plants, originally published by B. Daydon Jackson and James Britten over a hundred years ago, but no full translation of Turner’s Latin text. William Turner, described by Charles Raven as “the true pioneer of natural history in England”, was born at Morpeth in Northumberland around 1510, became a student and then a Fellow of Pembroke Hall (now College), Cambridge, and died in London in 1568. He is best known for his great Herball (1551-1568), written after several years of self-imposed exile on the Continent for religious reasons from 1540 had widened his experience; but his pioneering work, Libellus, published before he left Cambridge, reveals him as first and foremost a botanist (with a primary interest in the naming and description of plants) rather than a herbalist. This new edition of Libellus has a 13-page introduction, which succinctly discusses Turner as botanist and plant-name scholar; the significance, sources and contents of his first botanical book; the English, scientific and officinal plant names in it; Turner’s Latin; previous editions of the book; and the editorial principles of the present one. We learn that the quarto volume of 20 pages is organised, in roughly alphabetical order, in 148 sections in the sequence of the scientific names, describes or mentions some 175 plant taxa, the great majority flowering plants (native and non- native), and includes some 255 English names (or forms); of these “no fewer than 44% are used as standard or Flora names today (form and spelling disregarded), with same or different botanical reference” and some 15% are, the editors believe, “first attested there”. There follows a facsimile of the only known extant copy of Libellus, in the British Library. Though the facsimile in the Ray Society edition is cleaner, this one, despite some quite serious show-through from the reverse sides of the pages, creases and other blemishes often removed in the former, is actually clearer where the printing is faint or smudged. It also provides a more authentic representation of the original. Next come a transcript and translation, set out on facing pages. The latter, derived from a “basic translation ... carried out as a term paper in Latin” by Ms Kerstin Olsson of the Department of Classical Philology, Uppsala University, is unstilted but faithful to Turner’s Latin. [Indeed, having compared the translations of the preface and the entry under “NARCISSVS” with those of Charles Raven reproduced in the Ray Society edition, I tend to favour the new ones.] This section is followed by a helpful commentary on selected phrases, set out under headings identical to those in the transcript and translation. Together, these two sections occupy 69 pages. One of my few criticisms is that, when one is looking something up, it is difficult S see at a glance which of the two one has opened: running headings would have been helpful efe. 458 BOOK REVIEWS Much has been made of the unwieldy length of pre-Linnaean plant names, but, as the introduction to this book points out, there are “no polynomial phrase-names in Turner’s works”: indeed “one-term terminology” (e.g. “ACORVM” for Iris pseudacorus) is the rule, occasionally with a synonym, while 9% of the headings are binomials and more of these occur within the text; a few (e.g. “Plantago maior” and “rumex acetosus’’) are “still with us as standard names, unaltered or only slightly modified”. The book ends with three tables (the first with six pages of notes), a brief index of persons and places, and five pages of bibliography. The first table, in four columns, gives Turner’s English and scientific names with their modern equivalents, the second Turner’s scientific names with their modern equivalents, and the third the modern scientific names and the corresponding English names in Libellus. These are approximately equivalent to Tables C, A and B respectively on pp. 61-73 of the Ray Society volume, based on those of Jackson (1877 and 1878), and reveal few changes in the identification of Turner’s taxa. I strongly recommend this excellent little book (or libellus!). P. H. OSWALD Flora of North America north of Mexico. Volume 22 Magnoliophyta: Alismatidae, Arecidae, Commelinidae (in part), and Zingiberaceae. Flora of North America Editorial Committee. Pp. Xxiil + 352. Oxford University Press, New York & Oxford. 2000. £65-00. ISBN 0—19-513729-9 (y..22), This is the fourth volume of the Flora of North America to be published. It provides accounts of the monocot families from Butomaceae to Marantaceae in the Cronquist system, excluding the families Cyperaceae and Poaceae which are reserved for later volumes. Many of the 30 families included are widespread in the North Temperate zone, and have numerous representatives in the British Isles. The largest of the Temperate families are Juncaceae (118 species), Potamogetonaceae (37) and Alismataceae (34). However, a substantial minority of the families covered are primarily Tropical, with many of their North American species restricted to the southern fringes of the continent. These include Commelinaceae (51 species), Arecaceae, perhaps more familiarly as Palmae, (29), and the slightly more widespread Xyridaceae (21). The format of this Flora has been outlined in the reviews of two earlier volumes in Watsonia 21: 141 (1996) and 22: 195-6 (1998). In addition to the basic features common to most Floras, it contains distribution maps, some biological, ecological and ethnobotanical information and very useful literature references. (Perhaps mindful of the ever-litigious nature of American society, the editors warn readers that they do not encourage or recommend the reported folk remedies or culinary practices!). Botanists accustomed to Stace’s New Flora will notice relatively few differences in the treatment of species which are common to the British Isles and North America. Generic concepts are usually similar, although Potamogeton Subgenus Coleogeton is separated as the genus Stuckenia. At lower taxonomic ranks there are occasional divergences in nomenclature (e.g. Potamogeton oblongus rather than P. polygonifolius) and in taxonomy (e.g. Juncus balticus is treated as J. arcticus var. balticus, part of a “wide-ranging and obviously polymorphic complex”). As European authors usually restrict their attention to European plants and North American authors to the American flora, there are often unresolved questions about the relationships between the two. Are the American Juncus arcticus var. alaskanus, Potamogeton strictifolius and Sparganium eurycarpum synonymous with the European J. arcticus sens. str., P. rutilus and S. erectum? These issues are sometimes discussed, sometimes ignored. In the account of the Potamogetonacaeae by R. R. Haynes & C. Barre Hellquist, I was less struck by divergences in taxonomy than by the differences in approach to the genus compared with European authors. They “strongly recommend that no one collect specimens of Potamogetonaceae that are lacking reproductive structures” and that “all specimens should be collected when in fruit”. A herbarium built up on these lines would be a useful collection of voucher specimens, but would provide no information on the over-wintering habit of species, or their first leaves or phyllodes: even flower and pollen characters would be past their best and turions probably under-represented. Species such as P. robbinsii (which “rarely flowers”) would rarely be collected, and sterile hybrids would be overlooked. In fact, although the authors say that “hybridization is common among BOOK REVIEWS 459 members of the genus” they provide very little information about hybrids, simply listing those accepted by Hagstr6m in his 1916 monograph and by later authors. This comes as a severe shock to anyone familiar with the history of Potamogeton studies, as Fernald in his 1932 revision of the North American linear-leaved species famously criticised Hagstr6m for proposing “preposterous hybrids (of hypothetical parents thousands of miles apart)”. It is a pity that the opportunity was not taken to produce a revised treatment of the North American hybrids in this volume. The account of Typha provides a contrast to that of Potamogeton, as C. Galen Smith draws on the results of 40 years’ study of the genus in North America to provide a treatment in which hybrids are keyed out and T. x glauca (a “noxious weed’’) is illustrated alongside its parents T. angustifolia and T. latifolia. Exchange of plant material between Europe and North America has been taking place ever since the European discovery (or rediscovery) of the continent in 1492. This Flora provides details of North American aliens which are to varying degrees naturalised in Britain, including Elodea canadensis, E. nuttallii, Juncus tenuis, Lysichiton americanus, Sagittaria latifolia and S. rigida. In the account of Elodea Haynes notes that in contrast to their behaviour in Europe, he knows “of no instance in North America where Elodea nuttallii or E. canadensis is weedy”. Surprisingly, there is nO mention in this account of the occurrence of hybrids between these two species in North America, although the monograph by Cook & Urmi-KG6nig in which this possibility is suggested is cited. There are many more European plants naturalised in North America than vice versa, and some of the species included in this volume are Butomus umbellatus, Hydrocharis morsus-ranae, Juncus capitatus(!), J. compressus, J. inflexus, Luzula pallidula and Potamogeton crispus. Hydrilla verticillata has spread widely since it was first recorded in the U.S.A. in 1959; it is described as “widely distributed in the Eastern hemisphere but it is uncertain as to where it is truly native’. Could this species, like Aponogeton distachyos (S. Africa) and Egeria densa (S. America), have been introduced to both Europe and North America? Botanists who need to identify plant material from North America will clearly find this Flora indispensable. I hope that I have written enough to show that those whose interest lies in the taxonomy and distribution of the British and Irish flora will also find much of interest in it. C. D. PRESTON A flora of Tiree, Gunna and Coll. D. A. Pearman and C. D. Preston. Pp. 168. Published privately by the authors. 2000. £9.00. ISBN 0—9538111-1-5. After several recent floras which have left a visible bend in my bookshelves, it is a pleasure to have a book which can be taken along on a trip without a second rucksack. And to use the book will involve a trip, as Coll and Tiree, along with Colonsay, are the least accessible of the Argyll islands and taking a car is an expensive luxury. In terms of pages per unit area, it is arguable that this is actually a very large flora, as the islands are small, covering bits (sometimes very small bits) of only eight hectads and this, at my reckoning, runs out at about 20 pages per hectad. The relative value is even better if you work it out in tetrads. Whatever currency you use, this excellent account of the flora of these islands is a good buy. From a decision to produce a flora in 1996 to its publication has taken a mere four years and only one dedicated field trip, and this during a period in which both authors were much involved in other projects. Much of the usual process of balancing coverage of different areas and habitats and the checking of records has clearly been cut short. As the authors acknowledge, it would have been a difficult process from hundreds of miles away. I am glad that this was not regarded as an obstacle, as I would much rather have this somewhat incomplete account than no flora at all. As with all floras, this account has the sense of a “work in progress”, heightened by the remoteness of the islands and the history of recording here. Throughout most of the Highlands and Islands, resident botanists are (and have always been) very sparse and most recording is done by a limited number of expeditions, usually in high summer with all the bias that this induces. Fortunately, Coll and Tiree have proved an attractive target, both for itinerant botanists, particularly Joan Clarke and friends, and institutional study, notably Heslop Harrison in the past and Scottish Natural Heritage more recently, as the impressive bibliography shows. As a result, Coll and Tiree are probably as well-recorded as much of mainland Argyll and, certainly, the plant 460 BOOK REVIEWS communities on the islands have attracted more learned discourse. This book is a synthesis of the results of these botanical visits and one of its strengths is that a lot of disparate data are drawn together into a concise whole. With such diverse sources of information there will always be a number of erroneous records and these have been dealt with in a common-sense manner, as has the particular problem posed by the stigma attached to the accounts of the flora by Heslop Harrison. The detective work this has involved makes interesting reading. In fact, the whole of the introduction is excellent. The descriptions of the main habitats are succinct and it is good to see NVC community categories being used as a matter of course. The section on phytogeography may seem rather long for these small islands but given their position on the edge of things, the association with Heslop Harrison and his theory of per-glacial survival and Chris Preston’s recent work on geographical relationships in plant distribution, this seems justified. The comparison with the nearby island of Mull and the somewhat more distant Outer Isles is fascinating. The layout is clear and straightforward with the comparatively long introductory section followed by the species accounts and finally by a short account of the stoneworts by Nick Stewart and ecological notes, with tetrad maps, on the Cyperaceae of Tiree by David Pearman. Other tetrad maps are provided only for a selection of aquatic plants, probably reflecting the lack of precision of older records, the main thrust of visits by the authors and a desire to save space. For most species of interest, the frequent citing of localities in the text with map references more than makes up for this. The species accounts are commendably concise but fuller accounts are given where there is a story to be told or a question to be asked, be it about rarities like Eriocaulon aquaticum or the more mundane Eriophorum vaginatum. Some accounts are rather enigmatic with Juncus squarrosus “common on heathy ground” but “very poorly recorded”. To someone used to the mainland Argyll flora, the rarity of plants like Digitalis purpurea, Eriophorum vaginatum and Conopodium majus, abundant plants in main Argyll, comes as a surprise. Do go and visit Coll and Tiree, they are wonderful islands, and do take this flora with you. Almost every page of the species accounts has at least an implied query on which the curious botanist could shed some light. G. P. ROTHERO Lichens. Oliver Gilbert. Pp. 214. The 86th volume in the New Naturalist Series. Harper Collins. 2000. Hbk £34-99. ISBN 0 00220081 8. Pbk £19.99. ISBN 0 00 2200821. The high standard we have come to expect of this invaluable series is maintained or even excelled in this latest volume. The student of the smaller fungi, like the bus passenger, has had a patient wait (55 years since the first New Naturalist Volume) and now two books have turned up at once with the publication of Ingram and Roberston’s “Plant Disease” as number 85 in 1999. In an age when “biodiversity” is on everyone’s lips and in theory the otter or eagle now have equal status with the rarest lichen or plant pathogen the publishers and editors of this series are to be congratulated in bringing to the press two books which must further the appreciation and understanding of this huge slice of our wildlife. In his opening chapter Gilbert shows that lichens are of more than just scientific interest with a review of the way they have impinged on our life. If you enjoy the look, feel and scent of Harris tweed, relish the long lingering qualities of the finest perfumes or just want to test the pH of your soil with litmus paper he describes the essential part lichens play. You are then led through clear and succinct chapters on the nature of lichens, creatures that need lichens, and air pollution and lichens, to the main part of the book. Here the important role that lichens play in most of our major ecosystems is described, with chapters on woodland, acid rocks, heaths and moors, chalk and limestone, the built environment of church and village, our mountain areas, the rivers, lakes and coastal habitats. At a glance these pages seem liberally splattered with Latin names (since few lichens are blessed with English names). The non-lichenologist should not be put off. Read any paragraph and they trip easily by with Gilbert’s lively writing style almost invariably linking them into some thought-provoking ecological process or ecological niche you never thought of. These range from the effects of wind blasted ice and grit in subarctic heaths of the Cairngorms to the “canine zone” of tree bases in suburbia and from crashed aircraft on St Kilda, through old shoes at Girvan, to the drip zone under galvanised motorway crash barriers. BOOK REVIEWS 461 Three appendices provide information on the use of lichens for dating objects, the biological estimation of air pollution and the use of lichens as indicators of ecological continuity. There is also an adequate glossary and extensive bibliography. Misprints and errors are few and mostly inconsequential. Occasionally a now superseded lichen name creeps back in and the reference to “Soya” sheep on page 237 should probably not be proffered as evidence of a previously overlooked transgenic industry in western Scotland. The high standard of paper, print and plates we have seen in some of the recent “New Naturalists” is maintained here. This book conveys the excitement of the recent rapid accumulation of knowledge of the part lichens play in British ecosystems and identifies areas where the amateur can still make a significant contribution. It handsomely realises the aspirations of the editors in “seeking to interest the general reader by recapturing the enquiring spirit of the old naturalists”’. If this spirit moves you to discover the lichen genus Rocella in Ireland the author offers a more tangible spirit prize in the shape of a bottle of whiskey to the finder. Even without this inducement I am certain this book will stimulate you to get out and look at lichens, towns, cities and countryside in a new and exciting way. R. G. WOODS Alien Plants of Yorkshire. G. T. D. Wilmore. Pp. 316. Yorkshire Naturalists’ Union. 2000. Price £15.00 (reduced price to members of the Yorkshire Naturalists’ Union £12.00) ISBN 0—9504093- 3-2. It is perhaps surprising that in a small country with a flora that has been extensively botanised for over 400 years, so little time and effort has been devoted to the study of alien plants. For most of that period they have formed an increasingly substantial minority of the flora of the British Isles, (over 40% of taxa at the last count) but have usually been treated with suspicion or disdain. Often regarded as “second class citizens” (try telling that to people in areas of the country suffering from the ravages of Fallopia japonica, Crassula helmsii or Rhododendron ponticum) they were at best relegated to appendices in Floras or at worst completely ignored. Clive Stace’s New Flora of the British Isles changed the position overnight, not only were botanists able to see what alien plants were present they also had help in identifying them. However, some areas of the Country did have a better record in studying the alien flora and this excellent and fascinating book sets out the history of such studies in Yorkshire and catalogues the present state of our knowledge. Some 3,046 taxa in 131 families are included and details given include: Scientific and English name; a brief statement on its status in Yorkshire, with Country or region of origin and list of Yorkshire vice-counties in which it has been recorded. For most taxa there then follows a more detailed list of records for each vice-county, but for about 90 common taxa there is a “thumbnail sketch” of habitats and history. The comprehensiveness of the book is well illustrated by the reviewer having penned a criticism on the omission of taxa alien to Yorkshire but native to Britain, only to find that appendix 3 covers this subject very well. Other appendices deal with less well documented taxa and taxa not included in Stace or the B.S.B.I.’s two “Alien” books, and a bibliography and comprehensive index complete the book. Misprints and “typos” are commendably rare; over half a dozen taxa are wrongly attributed with illustrations in the index and page numbers are missing from the contents page. But this reviewer does have an intense dislike of single lines of a paragraph at the top or bottom of a page., and there are too many of these “orphans”. However, these are no more than minor irritations that in no way detract from the excellence of the book itself. Geoffrey Wilmore and the Yorkshire Naturalists’ Union are to be congratulated on producing a book which deserves to be on the shelf of every botanist with an interest in the flora of these islands and hopefully will serve as an inspiration to others to produce similar volumes for other regions. R&GoEREIS Watsonia 23: 463-464 (2001) 463 Obituary RAYMOND PAUL BOWMAN (1931-1999) It was with much sadness and some shock that naturalists (particularly, but not only, in Hampshire) learnt of the sudden death of Paul (as he was known to us), on Calshot Spit on the Hampshire coast in July 1999. He had gone there to count the estuary birds and to record the local flora, and suffered a heart attack on this very hot day, apparently after a swim. Paul was born in Hythe, Hampshire, on the east side of the New Forest, in 1931. His remarkable mother Dorothy was born near Exbury on the south-east edge of the New Forest. When he was three years old, the family (which came to include his four younger sisters, June, Dawn, Wendy and Heather) moved to the Maybush district of north-west Southampton, but summer holidays were nearly always spent at the caravan at Longdown, on the edge of the Forest, south east of Ashurst. His mother describes how his schoolboy passion for wildlife blossomed at an early age. Tragically, at the age of 15, an infection of meningitis destroyed his hearing. This disaster had a profound influence on his lifestyle, and focused his interests even more on wildlife. Though best known later as a botanist, he patiently and methodically studied the bird life with binoculars, as of course he could not hear their calls, and became an outstanding ornithologist. He was a member of the Southampton NHS and Hants Ornithological Society for over 40 years, and the Recorder of the NHS for over 20. His skills as a field worker were outstanding - indeed quite exceptional - in the effort he expended in verifying records. Sadly, he never learnt to lip-read effectively (except within his own family) so although he was by nature a most outgoing and sociable person, communication with other naturalists was not easy. He could tell others all he wanted to say, but they had to write things down in reply. He became a quite outstanding botanist with an encyclopaedic knowledge of the flora of South Hampshire. As a member of the Botanical Society of the British Isles, his knowledge and meticulous work led to him being appointed botanical Vice-County Recorder for South Hants (v.c. 11), a post he held for 20 years until his death. When the work on the Flora of Hampshire was revived after various misfortunes, he became one of the three co-authors of this book (published eventually in 1996) with Lady Anne Brewis and myself; and the most casual inspection of the pages of this book reveals how much work he did towards its completion; not least in the detailed distribution maps, but in his patient and meticulous care in making records. I personally think he did more work for the Flora of Hants than most of the rest of us combined! And all of an incredibly high standard. In this period he did not neglect his ornithological studies, and every October he travelled, usually to the Scilly Isles, to take part in annual bird migration counts - incredibly without being able to hear a single bird song. Professionally he worked as a draughtsman and cartographer with the Ordnance Survey at Southampton. It is clear that his professional work in this field was of the same outstanding quality as his natural history work, and his work was recognised by the award of the British Empire Medal. His cartographic skills were of great help in his botanical work, and he helped much with the geographical and mapping aspects of the Flora of Hampshire. Unlike many botanists I know (some quite eminent!), if Paul gave you a grid reference you knew you could rely on it implicitly. He was a very good and patient photographer - some of his photos are included in the Hants Flora, and are superb. His ready smile and amiable, enthusiastic personality managed to shine through his severe disability, particularly during the protracted (and sometimes heated!) arguments that developed in committee during the gestation of the Flora. He took all these problems in his stride, with his excellent sense of humour. As soon as he was given brief written notes of what was actually going on in committee, he ‘cottoned-on’ instantly to the problems and was always very helpful. He was a botanist of reputation well beyond Hampshire, and had extensive correspondence with naturalists all over Britain. 464 OBITUARY Apart from his work as a naturalist and, above all, as a botanist, he has permanent memorials not only in the published Hants Flora, but in the horsetail hybrid, Equisetum x bowmanii C. N. Page (E. sylvaticum L. x E. telmateia Ebrh.) named after him by the pteridologist Dr C. N. Page. He found this hybrid, new to science, in 1986 in the northern part of the New Forest. It is still there, but it has not yet been found outside the New Forest, (Page 1987 - Watsonia 17: 273-277). I found Paul to be a lovable personality and a great naturalist whom we miss very much in botanical circles in Hampshire. FRANCIS ROSE Watsonia 23: 465-467 (2001) 465 Report ANNUAL GENERAL MEETING, 8 MAY 2000 The Annual General Meeting of the Society was held at the Ulster Museum, Belfast, at 10.45 a.m. The President, Mrs M. Briggs took the chair 56 members were present. Apologies for absence were read and the minutes of the 1999 Annual General Meeting (published in Watsonia 23: 373-375) were circulated, approved as a correct record and signed by the President. REPORT OF COUNCIL The report had been circulated to members and the President went through it commenting on the main achievements of the Society during the past year, noting especially the good progress with the Threatened Plants Database and Atlas 2000 and she thanked all who had contributed — the Co-ordinators, the Atlas organiser, the inputters and all the members. She finally thanked every one who had, voluntarily, worked so hard for the Society. The Treasurer then presented the financial part of the Report. Summary accounts had been circulated to members; full sets of accounts had been available on request by post and were now available at the meeting. The Treasurer reported that the Society’s accounts were in good order and that we had worked within our means. He pointed out that as Gift Aid had replaced Covenants, the Society is able to recover tax paid on subscriptions on completion of a simple form which would be circulated to all members; in response to a question, however, it was pointed out that this could not be done if the subscriber was, as a professional, claiming the subscription against tax in his business accounts. The Membership Secretary had reported that the Society membership had declined, slightly, over the past year but that this did not appear to be as a result of the increase in subscription. The Treasurer proposed the adoption of these reports which was seconded by Mr E. F. Greenwood. The President thanked the Treasurer for all his hard work, particularly the extra effort involving the Charity Commissioners. ADOPTION OF REVISED RULES Copies of the proposed revised rules had been available by post from the Honorary General Secretary and were now available at the meeting. The Treasurer stated that the changes were purely administrative and were in response to advice from the Charity Commissioners:- They emphasise compliance with Charity Law. They emphasise best practice in financial control. They strengthen the quorum requirements at meetings of Council. They update details of permanent working committees. Under a requirement of Charity Law, they remove the vote from junior members under the age of 18. They clarify that family members have only one vote per family. The adoption of the Revised Rules was proposed by Mrs M. Lindop and seconded by Mrs J. Robertson and carried unanimously. At this point, Mrs M. Briggs stood down as President and, as she did so, thanked all members for their help and support during her presidency; in particular she thanked Mr Peter Thomson, Council Minuting Secretary and Mrs Margaret Perring who was retiring after handling B.S.B.L publications for 23 years; she welcomed Mr Jon and Mrs Sue Atkins of Summerfield Books who 466 REPORT were taking over. Finally, she welcomed her successor as president, Dr Geoffrey Halliday, well known to members for his work with Records Committee and for the acclaimed Flora of Cumbria. ELECTION OF PRESIDENT Dr G. Halliday was nominated by Council. He gave the Society’s thanks to Mrs M. Briggs M.B.E. for all her good work on behalf of the Society — 2 years as President, preceded by 23 years as Hon. General Secretary. He expressed his gratitude for his election and said that he was looking forward to making his own contribution to the office of President. He was elected unanimously. ELECTION OF VICE-PRESIDENT Professor C. A. Stace was proposed by Miss A. Burns and seconded by Mrs M. Briggs. His election was unanimous. ELECTION OF ACTING HONORARY GENERAL SECRETARY The President thanked Mr R. G. Ellis, the retiring Hon. Gen. Sec. for all his hard work in that office and noted with gratitude that he was continuing as Editor of ‘News’. Mr Ellis thanked Mrs Briggs and the Treasurer for their assistance and support and also recorded his thanks to Mr P. J. Fry for his work as Assistant Secretary at the B.S.B.I. desk at the Natural History Museum. The administration of the Society was under review, pending the outcome of the review, Miss A. Burns is nominated by Council to serve as Acting Honorary General Secretary. RE-ELECTION OF HONORARY TREASURER Mr M. E. Braithwaite, nominated by Council was re-elected, proposed by Mr R. G. Ellis, seconded by Mr A. O. Chater. The President thanked him very much for all his good work, particularly his continuing efforts with the Charity Commissioners. RE-APPOINTMENT OF EDITORS, INDEXERS AND REPRESENTATIVES OF COMMITTEES The President proposed the re-appointment of the Honorary Editors, Indexers and Representatives of Committees and thanked them very much for all their hard work, mentioning especially, Mr C. R. Boon for indexing Watsonia. At this point, the President also recorded his own thanks to Mr D. A. Pearman, Dr C. D. Preston, Mrs J. M. Croft and Dr T. D. Dines for their enormous amount of work with the Atlas. ELECTION OF COUNCIL MEMBERS In accordance with Rule 12, nominations had been received for Miss R. Davis, Dr J. Edmondson and Mr M. S. Porter. Profiles had been circulated and would be published. The election of these new Council members was proposed by Mrs M. Briggs and seconded by Mr E. F. Greenwood; their election was unanimous. ELECTION OF HONORARY MEMBERS The President stated that three members had been nominated for Honorary Membership of the Society this year and that all three had made outstanding contributions to the Society in their different ways. Sponsors for all three had produced appreciations of their respective candidates and these were either given, or, if not able to be present, read to the meeting and would be REPORT 467 published in ‘News’. Their election was proposed by Mr R. G. Ellis, seconded by Dr D. L. Kelly and was carried unanimously. The three new Honorary Members are, Dr F. E. Crackles M.B.E., MSc., sponsor Mrs M. Briggs; Dr P. Macpherson F.R.C.P., F.R.C.R., sponsor Mr M. E. Braithwaite; Mr R. D. Meikle, sponsor Dr E. F. Greenwood. RE-ELECTION OF HONORARY INDEPENDENT EXAMINER The President expressed the Society’s gratitude to Mr John Coats for his exemplary examination of the Society’s accounts. His re-election as Honorary Examiner was proposed by Mr A. O. Chater and seconded by Mr R. G. Ellis. There was no other business and the meeting closed at 12.40 p.m. AILSA BURNS BOTANICAL SOCIETY OF THE BRITISH ISLES (B.S.B.L.) The B.S.B.I. traces its origin to the Botanical Society of London founded in 1836 and has a membership of 2,850. It 1s the major source of information on the status and distribution of British and Irish flowering plants and ferns. This information, which is gathered through a network of vice-county recorders, is the basis for plant atlases and for publications on rare and scarce species and is vital to botanical conservation. The Society published Atlas of the British flora in 1962 and revised editions and is working towards a new atlas, to be completed in 2000. It organises plant distribution surveys, publishes handbooks on difficult groups of plants and has a panel of referees available to members to name problematic specimens. The B.S.B.I. arranges conferences and field meetings throughout the British Isles and, occasionally, abroad. It welcomes as members all botanists, professional and amateur alike. Details of membership and any other information about the Society may be obtained from: The Hon. General Secretary, Botanical Society of the British Isles, c/o Department of Botany, The Natural History Museum, Cromwell Road, London, 5W7 S5BD. BSBI Handbooks Each Handbook deals in depth with one or more difficult groups of British and Irish plants. No.1 Sedges of the British Isles A.C. Jermy, A.O. Chater & R.W. David. Revised edition, 1982. 272 pp., with descriptions, drawings and distribution maps for all 73 species of Carex. Paperback. ISBN 0 901158 05 4. No.2 Umbellifers of the British Isles T.G. Tutin. 1980. 200 pp., with descriptions and drawings of 73 species of Apiaceae (Umbelliferae). Paperback. ISBN 0 901158 02 X. No.3 Docks and knotweeds of the British Isles J.E. Lousley & D.H. Kent. 1981. 208 pp., with descriptions and drawings of about 80 native and | alien taxa of Polygonaceae. Paperback. ISBN 0 901158 04 6. Out of print. New edition with distribution maps in preparation; orders recorded. No.4 Willows and poplars of Great Britain and Ireland R.D. Meikle. 1984. 200 pp., with descriptions and drawings of 65 species, subspecies, varieties and hybrids of Salix and Populus. Paperback. ISBN 0 901158 07 0. No.5 Charophytes of Great Britain and Ireland J.A. Moore. 1986. 144 pp., with descriptions and drawings of 39 species and varieties of Characeae and 17 distribution maps. Paperback. ISBN 0 901158 16 X. No.6 Crucifers of Great Britain and Ireland T.C.G. Rich. 1991. 344 pp., with descriptions of 148 taxa of Brassicaceae (Cruciferae), 129 of them with drawings, and 60 distribution maps. Paperback. ISBN 0 901158 20 8. No.7 Roses of Great Britain and Ireland G.G. Graham & A.L. Primavesi. 1993. 208 pp., with descriptions and drawings of 13 native and nine introduced taxa of Rosa, descriptions of 76 hybrids, and 33 maps. Paperback. ISBN 0 901158 22 4. No.8 Pondweeds of Great Britain and Ireland C.D. Preston. 1995. 352 pp., with descriptions and drawings of all 50 species and hybrids of Potamogeton, Groenlandia and Ruppia, most of them with distribution maps: detailed introductory material and bibliography. Paperback. ISBN 0 901158 24 0. No.9 Dandelions of Great Britain and Ireland A.A. Dudman & A.J. Richards. 1997. 344 pp., with descriptions of 235 species of Taraxacum, most of them illustrated by silhouettes of herbarium specimens; drawings of bud involucres of 139 species and 178 distribution maps. Paperback. ISBN 0 901158 25 9. No.10 Sea beans and nickar nuts E. C. Nelson. 2000. 156 pp., with descriptions of nearly 60 exotic seeds and fruits found stranded on beaches in north-western Europe (many illustrated by Wendy Walsh) and of the mature plants (some with drawings by Alma Hathaway), accounts of their history and folklore, growing instructions etc. Paperback. ISBN 0 901158 29 1. Other publications List of vascular plants of the British Isles D.H. Kent. 1992. 400 pp. Nomenclature and sequence as in Clive Stace’s New Flora of the British Isles (1991, 1997), with selected synonyms. Paperback. ISBN 0 901158 21 6. Supplied with errata lists and Supplement 1. Alien plants of the British Isles E.J. Clement & M.C. Foster. 1994. 616 pp. Lists 3,586 recorded non-native species (of which 885 are established), with English names, frequency of occurrence, status, areas of origin, location of voucher specimens, references to published descriptions and illustrations, and selected synonyms. Paperback. ISBN 0 901158 23 2. Alien grasses of the British Isles T.B. Ryves, E.J. Clement & M.C. Foster. 1996. 234 pp. A companion volume to the last, listing over 700 non-native grasses; includes keys to bamboos and eight of the larger and more difficult genera and 29 pp. of illustrations. Paperback. ISBN 0 901158 27 5. Plant crib 1998 T.C.G. Rich & A.C. Jermy. 1998. 400 pp. An identification guide for some 325 ‘difficult’ taxonomic groups, with explanations, keys and illustrations of plant details. A4 paperback. ISBN 0 901158 28 3. British Red Data Books 1 Vascular plants M.J. Wigginton, ed. 1999 (3rd ed.). 468 pp. Up-to-date information on 408 taxa (including 118 microspecies) regarded as threatened in Great Britain, with details of their habitats and associated species, maps showing pre-1970, 1970-1987 and post-1987 records by 10-km squares, and 1l-km square frequency maps for 63 taxa. Hardback, published by JNCC, Peterborough. ISBN | 86107 451 4. Scarce plants in Britain A. Stewart, D.A. Pearman & C.D. Preston, comp. & ed. 1994. 518 pp. Accounts of 254 nationally scarce taxa (occurring in 16-100 10-km squares in Great Britain) and of 71 taxa formerly thought to be so, with details of their habitats and associated species, reproductive biology, changing British distribution and world range; updated distribution maps. Hardback, published by JNCC, Peterborough. ISBN 1 873701 667. Aquatic plants in Britain and Ireland C.D. Preston & J.M. Croft. 1997. 365 pp. Summarises the distribution, habitat and reproductive biology of 200 aquatic plants in 72 genera, with 200 distribution maps and 72 drawings. Hardback, published by Harley Books, Colchester. ISBN 0 946589 55 0. Available from the official agents for BSBI publications, Summerfield Books (Jon & Sue Atkins), Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX (Tel: 017683 41577. Fax: 017683 41687. E-mail: bsbipubs @beeb.net). — Ct ee ee ae — ee Se - fh AW FWNnNr NH OHOMAMDMAAOYIYNI YY ANA ABRWNEFDOMIDARWDHND Hil. He, HS: H4. HS. H6. H7. H8. He, H10. PL, Hi. H13. H14. NOR RR RR Re eS eR SomranansPwnw West Cornwall Scilly East Cornwall South Devon North Devon South Somerset North Somerset North Wiltshire South Wiltshire Dorset Isle of Wight . South Hampshire . North Hampshire West Sussex . East Sussex . East Kent West Kent . Surrey . South Essex . North Essex . Hertfordshire . Monmouthshire . Glamorgan . Breconshire . Radnorshire Carmarthenshire . Dumfriesshire . Kircudbrightshire Wigtownshire . Ayrshire Renfrewshire . Lanarkshire Peeblesshire Selkirkshire Roxburghshire . Berwickshire East Lothian . Midlothian West Lothian . Fife & Kinross South Kerry North Kerry West Cork Mid Cork East Cork Co. Waterford South Tipperary Co. Limerick Co. Clare North Tipperary Co. Kilkenny Co. Wexford Co. Carlow Laois NAMES OF VICE-COUNTIES IN WATSONIA Zl. PIPE pis 24. 20% 26: rua 28. 2. 30, oF 32. 33: 34. 36. cre 38. a”, 40. 53% 45. 46. 47. 48. 49. 86. 87. 88. 89. 90. ol De. 02. 94. 05: 96. 96b. OT: 93. H15. H1@. H17. H18. H19. H20. lgvae 22, H23. H24. H25. H26. H27. H28. S (113) Channel Islands ENGLAND Middlesex 54. Berkshire Oo) Oxfordshire 5S: Buckinghamshire 56. East Suffolk D1 West Suffolk 58. East Norfolk 59. West Norfolk 60. Cambridgeshire OF Bedfordshire 62. Huntingdonshire 63. Northamptonshire 64. East Gloucestershire 65. West Gloucestershire 66. Herefordshire 67. Worcestershire 68. Warwickshire 69. Staffordshire 69b. Shropshire 70. South Lincolnshire Tele WALES Pembrokeshire 50. Cardiganshire ai Montgomeryshire a2. Merionethshire Caernarvonshire SCOTLAND Stirlingshire 99. West Perthshire 100. Mid Perthshire 101. East Perthshire 102. Angus LOS: Kincardineshire 104. South Aberdeenshire 105. North Aberdeenshire 106. Banffshire 107. Moray 108. Easterness 109. Nairnshire 110. Westerness IRS Main Argyll 112. IRELAND South-east Galway H29. West Galway H30. North-east Galway H31. Offaly H32. Co. Kildare H33. Co. Wicklow H34. Co. Dublin H35. Meath H36. Westmeath H37. Co. Longford H38. Co. Roscommon H39. East Mayo H40. West Mayo Co. Sligo North Lincolnshire Leicestershire Rutland Nottinghamshire Derbyshire Cheshire South Lancashire West Lancashire South-east Yorkshire North-east Yorkshire South-west Yorkshire Mid-west Yorkshire North-west Yorkshire Co. Durham South Northumberland North Northumberland Westmorland Furness Cumberland Isle of Man Denbighshire Flintshire Anglesey Dunbarton Clyde Isles Kintyre South Ebudes Mid Ebudes North Ebudes West Ross East Ross East Sutherland West Sutherland Caithness Outer Hebrides Orkney Shetland Co. Leitrim Co. Cavan Co. Louth Co. Monaghan Fermanagh East Donegal West Donegal Tyrone Co. Armagh Co. Down Co. Antrim Co. Londonderry . , Fy 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. 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.rbge.org.uk/data/BSBI/Publications/watsonia.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: Mr D. A. Pearman, The Old Rectory, Frome St Quintin, Dorchester, Dorset DT2 OHF Plant Records: the appropriate vice-county recorder, who should then send them to Dr C. D Preston, Biological Records Centre, Monks Wood, Abbots Ripton, Huntingdon, PEI7 21S. Obituaries: Mrs M. Briggs, 9 Arun Prospect, Pulborough, West Sussex, RH20 IAL. Back issues of Watsonia are available from the official agents for BSBI Publications, Summerfield Books, (John & Sue Atkins) Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BX. Tel: 017683 41577. Fax: 017683 41687. E-mail: bsbipubs @ beeb.net to whom orders for all issues prior to Volume 22 part 1 should be sent. Recent issues (Vol 22 part 1 onwards) are available from Mr M. Walpole, B.S.B.L., 68 Outwoods Road, Loughborough, Leicestershire, LE11 3LY. Watsonia February 2001 Volume twenty three Part three Contents DIXON, J. M. On the status of the genus Koeleria Pers. (Poaceae) in Britain MASKELL, L. C. & RAYBOULD, A. F. The decline of eee Coe (Poaceae) in the British Isles RICH, T. C. G. & BIRKINSHAW, C. R. ee of Britain’ S bioversy cues depauperata With. (Cyperaceae), Starved Wood-Sedge FOLEY, M. J. Y. A morphological comparison between some British Orieue species (Orobanchaceae) and their closely-related, non-British aaa from continental Europe: Orobanche rapum-genistae Thuill. s.1. ALLEN, D. E. Brambles (Rubus L. sect. Rubus, and sect. ic: Lindley Rosaceae) in the Channel Islands RANDALL, R. D. & RICH, T. C. G. Conservation of Britain’s biodiversity ihe dasycoccus (Rosaceae), Thick-berried Bramble ea Nee ae hee NOTES Jones, R. A. & Stewart, N. F. Nitella gracilis (Smith) ean an elusive charophyte new to Cardiganshire (v.c. 46) a Welch, D. Colonisation by Cochlearia danica L. song sink a in eed Scotland from 1996 to 2000 Braithwaite, M. E. Alchemilla glaucescens Wallr. in v.c. 81 Benda Allen, D. E. Geranium robertianum L. var. marinum: an earlier author ; Holyoak, D. T. & Kitchener, G. D. planet x kitcheneri McKean (Onagracene) in East Cornwall BOOK REVIEWS OBITUARY REPORT Published by the Botanical Society of the British Isles ISSN 0043-1532 Typeset by D. K. & M. N. SANFORD Printed in Great Britain by PALLADIAN PRESS, UNIT E, CHANDLERS ROW, PORT LANE, COLCHESTER, ESSEX CO1 2HG 377-390 391-400 401-411 413-419 421-435 437-442 443445 446-449 449-451 451 452-453 455-461 463-464 465-467 AK i w 44 2 A Botanical Society of the British Isles Volume 23 Part 4 August 2001 Editors: M. Briggs, R. G. Ellis, D. L. Kelly, D. R. McKean, D. A. Pearman, M. N. Sanford, D. A. Simpson Botanical Society of the British Isles Patron: Her Majesty Queen Elizabeth the Queen Mother 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 2001—2002 President, Dr G. Halliday President-elect, Mr R. D. Pryce Vice-Presidents, Mr D. J. McCosh, Prof. C. A. Stace, Dr A. J. Richards Honorary General Secretary, Miss A. Burns Honorary Treasurer, Mr M. E. Braithwaite Editors of Watsonia Papers and Notes, D. L. Kelly, D. R. McKean, M. N. Sanford*, D. A. Simpson Book Reviews, D. A. Pearman Plant Records, R. G. Ellis Obituaries, M. Briggs *Receiving editor, to whom all MSS should be sent (see inside back cover). © 2001 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. Watsonia 23: 469-480 (2001) 469 What is Anthyllis vulneraria L. subsp. corbierei (Salmon & Travis) Cullen (Fabaceae)? te C.G. RICH Dept. Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP ABSTRACT Anthyllis vulneraria L. subsp. corbierei (Salmon & Travis) Cullen, originally distinguished by having spreading hairs on the stems and petioles, is reputed to be a British endemic confined to a few localities in Anglesey, Caithness, Cornwall and the Channel Islands. Examination of herbarium material indicates that plants with spreading hairs occur widely around the British coast and occasionally inland, encompassing a range of forms. Similar material with spreading hairs also occurs on the coast of northern France. Field work at previously reported sites (including the type locality) showed that most populations were polymorphic with respect to hair, corolla and calyx characters. Different hair types occurred in random combinations with different corolla and calyx characters. The occurrence of spreading or appressed hairs is probably simply polymorphic variation. I do not recognise subsp. corbierei either as an endemic or as a subspecies. KEYWORDS: Britain, France, Red Data Book, polymorphism, infraspecific variation. INTRODUCTION Anthyllis vulneraria L. is taxonomically very complex and consists of numerous inter-grading variants which are to some extent ecologically and geographically separated (Cullen 1976). Cullen (1968, 1976, 1986) provided accounts of the infraspecific variation, in which three native and two introduced subspecies were recognised in the British Isles. One of these, subsp. corbierei (Salmon & Travis) Cullen, is reputed to be a rare plant endemic to Britain, and was included in the third edition of the vascular plant Red Data Book (Wigginton 1999). However, Wigginton acknowledged that the RDB account was based on little recent information, and he suggested further investigation was required. This prompted R. A. Jones and T.C.G.R. to review the information available about subsp. corbierei and to carry out field work to assess its current status. The taxon was first described by Salmon & Travis (1917) as A. maritima Schweigg. var. corbierei (coastal forms of Anthyllis were at the time ascribed by some botanists to A. maritima Schweigg., now often referred to A. vulneraria subsp. vulneraria var. langei Jalas; Cullen 1986). It was distinguished as having an abundance of spreading hairs on the stems and petioles (‘Caules folia petiolique valde patento-villosae’) in contrast to typical A. vulneraria which has appressed hairs. After studying herbarium material (including type specimens), cultivation experiments and field work at South Stack, Cullen (1967) raised the variety to subspecific rank for Flora Europaea (Cullen 1968). In his résumé of the A. vulneraria complex, Cullen reported subsp. corbierei from only four localities, in Anglesey (stated as forming large, distinct, homogenous populations), Cornwall and Sark, and suggested it could occur elsewhere (Cullen 1976). Cullen (1986) provided a description, clearly based on his holotype in LIV: ‘Upright perennial. Stems hirsute over their whole length. Leaves somewhat fleshy, concentrated in the lower part of the stem. Calyx usually not red-tipped. Corolla clear yellow’. This description has formed the basis of descriptions in subsequent floras (e.g. Stace 1997; Akeroyd 1998) except for the poor description in Clapham et al. (1987) which was derived indirectly from Cullen’s Flora Europaea account. The taxonomic complexity of A. vulneraria forced me to a wider investigation of subsp. corbierei from the literature, herbarium specimens, field work and limited cultivation experiments to try to clarify which populations were that taxon and therefore which might require conservation. The work has led to a different conclusion from Cullen about the most appropriate rank for the taxon and its distribution, and consequently its priority for conservation. The work applies only to A. vulneraria from the British Isles and northern France, and does not apply to other European taxa with spreading hairs (see Cullen 1968, 1976). 470 TC, G-RICH ANTHYLLIS VULNERARIA FORMS WITH SPREADING HAIRS IN THE BRITISH ISLES As the taxon was originally distinguished on the basis of spreading hairs on the stems and petioles, a review of such herbarium material was carried out. All material of A. vulneraria from the British Isles was examined at BM, BRISTM, CGE (mostly seen by J. Cullen or J. R. Akeroyd), DBN, K (mostly seen by J. Cullen, also Marsden-Jones & Turrill’s 1933a material), LIV (which now includes Cullen’s research collection transferred from LIVU), NUW, OXF, RNG (mostly seen by J. R. Akeroyd), TCD (mostly seen by J. R. Akeroyd) and material named as subsp. corbierei was borrowed from E and LTR. The herbarium specimens with spreading hairs showed a large range of variation, especially with regard to corolla and calyx coloration. Notes on the specimens with most of the stem with spreading hairs seen, and their differences from Cullen’s description of subsp. corbierei are given in Appendix 1 with other records. The records are mapped in Figure 1 with selected silhouettes: these show a large range of variation in general growth form, distribution of leaves on the stem and in leaf shape. The herbarium material showed significant variation in the density, coverage and coloration of spreading hairs. Plants with spreading hairs right the way up the stem were uncommon, and many specimens had appressed hairs on the top and/or lateral peduncles (including Cullen’s holotype). Inland plants tended to be less densely hairy than coastal plants. In plants with spreading hairs, the occurrence of red-tipped calyces was more frequent than green-tipped calyces; these character states are simple polymorphic variation which is widespread in many populations in Britain, with red-tipped calyces being dominant in most (but not all) populations. Variation in corolla colour was also frequent, especially in Cornish material. Spreading hairs occurred in material which had been determined by various botanists as subsp. vulneraria or subsp. lapponica (Hyl.) Jalas, which are more or less recognisable subspecies, or as intermediates with subsp. corbierei. Plants with red corollas and spreading hairs were often ascribed to subsp. vulneraria var. coccinea L. but many could equally have been treated as subsp. corbierei. ANTHYLLIS VULNERARIA FORMS WITH SPREADING HAIRS IN NORTHERN FRANCE It has been suggested that subsp. corbierei might occur on sea cliffs in the western part of France (e.g. Cullen 1976, noting he had seen little material from this area). Corbiére (1894) had noted variation in pubescence in Normandy similar to that described by Salmon & Travis (1917), but the name he had applied (var. sericea Bréb.) could not be clearly related to Salmon and Travis’s taxon from Brébison’s description. Couderce (1975) revised the French infraspecific taxa of Anthyllis vulneraria, and cited subsp. corbierei as synonymous with his subsp. vulgaris (Koch) Willk. & Lange var. maritima (Koch) Willk. His description included plants with a mixture of hair types, including those with spreading hairs throughout the stem: ‘a pilosité généralement apprimé, mais parfois hirsute dans la moitié inférieure ou méme sur toute leur longueur’ . Specimens from the Musée National d’ Histoire Naturelle, Paris (P) and the Herbier de la Société des Sciences Naturelles et Mathématiques de Cherbourg (CHE) were examined for material which might be regarded as subsp. corbierei (these herbaria contain Corbiére’s collections, the former having been transferred from Caen CN). Eight collections from the coast had dense spreading hairs below and at least sparse spreading hairs above (Appendix 2). These are regarded as falling within the variation noted in the British Isles (cf. above and Appendix 1). Corbiére’s annotations on the sheets indicate that he regarded such plants with spreading hairs as var. sericea Bréb., as surmised by Salmon & Travis (1917). Similarly, there are two more recent inland collections from the Pas-de-Calais in the Centre Régionale de Phytosociologie/ Conservatoire Botanique National de Bailleul herbarium. ANTHYLLIS VULNERARIA SUBSP. CORBIERE]I 47] FIGURE 1. Map showing distribution of forms with spreading hairs in Britain with silhouettes drawn from selected specimens. 472 T.C.G_RICH FIELD INVESTIGATION OF KNOWN POPULATIONS The variation in herbarium material prompted a more detailed investigation of variation in the field. Field work was carried out in four areas, concentrating on those from which Cullen had cited material: Anglesey (including the type locality “grassy banks, cliffs near South Stack’), Cornwall and the Channel Islands (the latter by M. Marsden, B. Ozanne, R. Rabey and R. Veall in 1999), and also Dorset. At each site the phenotypes present were noted. The variation is summarised in Table 1 with phenotypes coded as follows: Corolla: R = red, Y = yellow, W = red standard with white keel, O = orangey-red. This includes the most frequent phenotypes, though there is variation within them (cf. Marsden-Jones & Turrill 1933a). Stem and petiole hairs: A = appressed, S = spreading to ascending. The hairs on stems of some plants can be spreading to ascending below but appressed above (cf above). Calyx: G = whitish-green-tipped, T = red-tipped. These are the two variants, though the development of the redness in the tips is variable, and may sometimes be almost purple. I have not seen any plants with red or orangey corollas with green-tipped calyces. Thus plants coded YAT have yellow corollas, appressed hairs and red-tipped calyces (typical ‘yulneraria’), those coded YSG have yellow corollas, spreading hairs and whitish-green-tipped calyces (subsp. corbierei sensu Cullen), etc. Representative specimens have been deposited in NMW. ANGLESEY Most populations investigated were polymorphic, containing two or more phenotypes (Table 1). The composition of the populations differed between sites; for instance, populations from Aberffraw to Porth Trecastell were dominated by plants with appressed hairs and red standards with white keels (other forms being rare), whilst those at South Stack all had yellow corollas and a mixture of hair types. Plants equating to subsp. corbierei were found only at South Stack, Addoloy-Penrhosfeilw chapel and Porth Trecastell. No plants were refound at Porth Nobla, Rhosneigr or Aberffraw, though the exact original sites are not known and vegetative plants with spreading hairs were found in the latter in 1993 (R. A. Jones, pers. obs.). At the type locality at South Stack there was significant variation in the growth form. Some plants in deep turf sheltered by rocks were erect, but others in more exposed positions were decumbent (in cultivation, some variation in growth form is retained; photograph in NMW). The leaves were somewhat fleshy (thick by comparison with inland plants), but the cauline leaves did not appear to be noticeably concentrated in the lower part of the stem (this is also true of the type specimens at LIV and BM). All plants seen had + uniformly yellow corollas. Four phenotypes were present, differing in the combinations of spreading or appressed hairs and red-tipped or green-tipped calyces, but otherwise identical. For plants with spreading hairs, there was some variation in the density and distribution of hairs. Most plants had dense spreading hairs along the length of the stem, though a few had sparse spreading hairs, and one was noted with spreading hairs below and appressed hairs on the peduncles. There was also a little variation in the redness of the calyces, with some plants with only the very tips red. CORNWALL Plants with spreading hairs equating to subsp. corbierei were found at five of the previously- reported localities (Hayle Towans, Tubby’s Head, Perranporth, Porth Joke and Pentire Point East), and at two new sites (Watergate Bay and Rocky Valley) (Table 1). No plants were refound at Bedruthan Steps or Tintagel, but they could still be present. Plants with spreading hairs may be widespread around the Cornish coast, but they are not present in every population. The north Cornish Anthyllis populations are highly polymorphic, and probably the most variable in Britain. In all localities, spreading hairs occurred with practically all other combinations of corolla and calyx colour, and such plants could not otherwise be separated from the variation in plants with appressed hairs. They were quite variable in growth form, probably related to exposure. ANTHYLLIS VULNERARIA SUBSP. CORBIEREI 473 TABLE 1. PHENOTYPES OF ANTHYLLIS PRESENT IN SELECTED SITES IN ANGLESEY, CORNWALL AND DORSET SITE PHENOTYPES Anglesey (June 1999) Tywyn Aberffraw (SH356688) WAT Ynys Meibion, cliffs opposite (SH327686) RAT, WAT Porth Trecastell, cliffs and roadside (SH3370 area) OAT, RAT, WAT, YAG, YAT, YST Rhosneigr, roadside (SH319726) VAT Rhosneigr, roadside (SH324739) YAT Black Creek, Bodior, Rhoscolyn, rocks (SH294757) YAG, YAT Gromlech, Rhoscolyn, cliffs and tracks (SH262764 area) OAT, RAT, WAT, YAT Porth Diana, Trearddur (SH253782) WAT, YAT Porth-y-post, Trearddur, roadside (SH243796) WAT, YAT Addoloy-Penrhosfeilw chapel, roadside (SH226806) YAT, YST Pen-y-bonc, Penrhosfeilw, roadside (SH217812) YAT, YAG South Stack, cliffs (SH2082 area) YAT, YAG, YSG, YST Porth Tywyn-mawr, rocks (SH285850) YAG, YAT Ynys y Fydlyn, Llanfairynghornwy, cliffs (SH292917) RAT, WAT, YAT Cornwall (June 2000) Hayle Towans, fixed dune grassland (SW554385) OAT, RAT, WAT, YAG, YAT, YSG, YST Porthtowan, cliffs (SW69248 1) OAT, RAT, WAT, YAG, YAT Chapel Porth, cliffs (SW697495) OAT, RAT, WAT, YAG, YAT Tubby’s Head, cliffs (SW698505) OAT, WAT, YAG, YAT, YSG, YST Droskyn Point, Perranporth, cliffs (SW755545) OAT, WAT, YAG, YAT, YSG Porth Joke, cliffs (SW772604) OAT, OST, WAT, YAG, YAT, YSG, YST Pentire Point East, grassland (SW782615) OAT, OST, RAT, RST, WAT, WST, YAG, YAT, YSG, YST Watergate Bay, cliffs (SW841648) OAT, OST, RAT, WST, YAG, YAT, YSG, YST Bedruthan Steps, cliffs (SW8469) OAT, RAT, WAT, YAG, YAT Tintagel, cliffs (SX0589) OAT, RAT, WAT, YAG, YAT Rocky Valley, cliffs (SX073896) OAT, YAG, YAT, YSG, YST Dorset (July 2000) South of Fortuneswell, Portland, old quarry (SY693731) YAG, YSG, YST Easton, Portland, car park (SY696723) YAG, YAT, YSG, YST Handfast Point, Ballard Down, grassland (SZ054825) YAG, YAT, YSG, YST See text for coding of phenotypes DORSET Three sites were investigated, all of which had a mixture of phenotypes (Table 1). Plants equating to subsp. corbierei were found in all three, though they had less dense spreading hairs than plants in Anglesey and Cornwall, and all had appressed hairs above. Again, plants with spreading hairs covered the same range of variation in corolla and calyx colour as those with appressed hairs. Growth form also varied markedly between sites. CHANNEL ISLANDS Plants on Sark had spreading hairs, yellow corollas and red-tipped calyces (M. Marsden, R. Veall, pers. comms., 1999). Plants with appressed hairs have also been recorded on the island (K; see above). Similarly, on Guernsey the plants with spreading hairs had yellow corollas and red-tipped calyces, and occurred in mixed populations with plants with appressed hairs (B. Ozanne, R. Rabey, pers. comms., 1999). The field work showed that most of the larger populations are polymorphic, and that plants with spreading hairs were otherwise identical to those with appressed hairs and always grew with them. 474 TCG, RICH VARIATION IN FREQUENCY OF PHENOTYPES WITHIN POPULATIONS The variation of the corolla, calyx and hair characters prompted further analysis to see if plants with spreading hairs were most commonly associated with yellow corollas and green-tipped calyces (cf. Cullen’s 1986 description), or occurred in random combinations with all corolla and calyx colours. The frequency of characters was scored for 100 plants along transects at four sites. The numbers of plants with each combination of characters recorded in three transects at the type locality in Anglesey are shown in Table 2 (all plants had yellow corollas). 27% of the plants had spreading hairs (range on transects 24-32%), and 69% of the plants had green-tipped calyces (range 66-74%). All phenotypes were scattered + randomly along the transects, though often patchily. A y? analysis showed that the calyx and hair characters occur together in random combinations independent of each other (y? = 0-42, p>0-1), and there is no particular association of green-tipped calyces with spreading hairs (typical ‘corbierei’ ) or red-tipped calyces with appressed hairs (typical ‘vulneraria’ ). Single transects were recorded at Hayle Towans, Pentire Point East and Watergate Bay in Cornwall (Table 3). The composition of the populations differed between sites, and the population was especially variable at Pentire Point East. The percentages of plants with spreading hairs at these sites were 14%, 28% and 15% respectively. Again, a y? analysis showed that the corolla, calyx and hair characters occurred together in random combinations (y? = <0-001, p>0-9; x? = 0:58, p>0-:5; 7? = 0-03, p>0-9 respectively). TABLE 2. NUMBER OF PLANTS OF EACH PHENOTYPE RECORDED IN EACH OF THREE TRANSECTS OF 100 PLANTS AT TYPE LOCALITY, SOUTH STACK, JUNE 1999 Phenotype NW facing SW facing SW facing grassy slope grassy slope rocks Mean YAG 49 3D 4 49-33 YAT (‘vulneraria ) De 19 24 23°33 YSG (‘corbiere?’) 7 19 23 19-66 YST 7 7 9 7:70 See text for coding of phenotypes TABLE 3. NUMBER OF PLANTS OF EACH PHENOTYPE RECORDED IN TRANSECTS OF 100 PLANTS AT CORNISH SITES, JUNE 2000 Phenotype Hayle Towans Pentire Point East Watergate Bay OAT l 18 9 OST 0 8 1 RAT 0 11 3 RST 0 Z 0 WAT 0 3 0 WST 0 4 2 YAG DS 6 58 YAT (‘vulneraria ) 60 34 16 YSG (‘corbierei’ ) 9 2 8 YST 5 12 6) See text for coding of phenotypes ANTHYLLIS VULNERARIA SUBSP. CORBIEREI 475 DISCUSSION The herbarium material showed that plants with spreading hairs vary in morphology between sites in Britain (cf. Figure 1) and France. The field investigation also showed that much variation occurs within populations, in which plants with spreading hairs could not otherwise be separated from the variation in those with appressed hairs. Whilst there were always some plants with spreading hairs and yellow corollas which conformed to Cullen’s description of subsp. corbierei in the populations investigated (allowing for the variation in calyx colour which I do not regard as of taxonomic significance), there were other plants with spreading hairs which did not. The analysis of the frequency of phenotypes within four populations showed that spreading hairs occurred in apparently random combinations with corolla and calyx colour. Cullen (1976) described the type population of subsp. corbierei at South Stack as homogeneous, but the field work showed that this is not true for hair type, the essential character on which the original variety was based. Indeed, Salmon & Travis (1917) had already noted that most of the plants were typical with regard to hair-clothing (i.e. had appressed hairs). Genetic experiments are required to establish the basis of inheritance of the spreading and appressed hair characters, though the complex variation in pubescence seen suggests it is unlikely to be a simple Mendelian inheritance controlled by one gene. Marsden-Jones & Turrill (1933a) cultivated plants to investigate the genetic basis of variation (primarily corolla colour), but the only observation reported on hair type was that three out of 20 plants cultivated from wild seed collected from one plant at Par Harbour (R.21) had spreading hairs on the stems, the others having appressed hairs. Three seedlings grown by T.C.G.R. from seed collected from a plant with spreading hairs at South Stack also had spreading hairs showing that it can breed true. I am thus unable to delimit a homogeneous taxon which can be equated with Cullen’s subsp. corbierei, and which can be consistently distinguished from variation in other Anthyllis vulneraria with which it always grows. My current interpretation of the hair characters in Britain is that they are polymorphic, and that plants with spreading hairs do not merit recognition at subspecific rank. Indeed, plants with spreading hairs occurred in collections which would otherwise be treated as subsp. vulneraria or subsp. lapponica. What is interesting is that spreading hairs occur most frequently in populations of the western Atlantic coast, indicating some ecological and distributional integrity. There is little value in giving names to each random combination of characters (ten names would be required to name the different combinations of hair, corolla and calyx characters observed at Pentire Point East alone, cf. Table 3), and I follow Marsden-Jones & Turrill’s (1933b) advice in refraining from making new combinations until the genetic basis has been resolved. Meanwhile, I regard subsp. corbierei as synonymous with subsp. vulneraria and suggest it should be removed from the Red Data Book (Wigginton 1999). ACKNOWLEDGMENTS This work was stimulated by Andy Jones who kept asking questions that were difficult to answer, at least initially, and I am grateful for his prompting and assistance. I would like to thank the Keepers of the herbaria for loan of or access to material, and Arthur Chater, John Edmondson, Sally Ellis, Trevor Evans, Vera Gordon, Richard Gornall, Gwyll Lewis, Alex Lockton, Marcia Marsden, David McClintock, Andy McVeigh, Bridget Ozanne, Alison Paul, Chris Preston, Rachel Rabey, Benoit Toussaint, Roger Veall, Rosie White and Goronwy Wynne for assistance with field work or specimens, Andrew Pang for drawing the silhouettes, and Jane Croft for details of records at the Biological Records Centre. The field work in England was funded by the Threatened Plants Database. The map in Figure | was plotted using DMAPW by Alan Morton. REFERENCES AKEROYD, J. R. (1998). Anthyllis vulneraria, in RICH, T. C. G. & JERMY, A. C. (1998). Plant Crib 1998. pp. 181-182. Botanical Society of the British Isles, London. ALLEN, A. & HILTON, B. (1994). Flowers of Sark. Privately published, Barnstaple. CORBIERE, L. (1894). Nouvelle Flore de Normandie. Caen. 476 TC-GaRICH COUDERC, H. (1975). Contribution a la révision des taxons infraspécifiques de l’Anthyllis vulneraria L. en France. Revue Générale de Botanique (Paris) 82: 93-118. CULLEN, J. (1967). Anthyllis vulneraria L. Watsonia 6: 295. CULLEN, J. (1968). Anthyllis L. in TUTIN, T. G. et al. eds., Flora Europaea 2: 177-182. Cambridge University Press, Cambridge. CULLEN, J. (1976). The Anthyllis vulneraria complex: a résumé. Notes from the Royal Botanic Garden Edinburgh 35: 1-38. CULLEN, J. (1986). Anthyllis in the British Isles. Notes from the Royal Botanic Garden Edinburgh 43: 277— 281. JERMY, A. C. & CRABBE, J. A., eds., (1978). The island of Mull. British Museum (Natural History), London. MARGETTS, L. J. & DAVID, R. W. (1981). A review of the Cornish flora. Institute of Cornish Studies, Redruth. MARSDEN, M. (1994). A new check list of flowering plants and ferns wild on Sark. Report and Transactions Société Guernesiaise 1994: 754-783. MARSDEN-JONES, E. M. & TURRILL, W. B. (1933a). Studies in variation of Anthyllis vulneraria. Journal of genetics 27: 261-285. MARSDEN-JONES, E. M. & TURRILL, W. B. (1933b). Notes on the taxonomy of British material of Anthyllis vulneraria. Journal of botany 71: 207-213. MCCLINTOCK, D. (1975). The wild flowers of Guernsey. Collins, London. ROBERTS, R. H. (1982). The flowering plants and ferns of Anglesey. National Museum of Wales, Cardiff. SALMON, C. E. & TRAVIS, W. G. (1917). Notes from Carnarvon and Anglesey. Journal of botany 55: 320. STACE, C. A. (1997). New Flora of the British Isles. 2nd ed. Cambridge University Press, Cambridge. WIGGINTON, M. J. (1999). Anthyllis vulneraria ssp. corbierei (Salmon & Travis) Cullen (Fabaceae), in WIGGINTON, M. J., ed., Red Data Books of Britain and Ireland 1: Vascular plants. 3rd ed. p. 40. Joint Nature Conservation Committee, Peterborough. (Accepted February 2001) ANTHYLLIS VULNERARIA SUBSP. CORBIEREI AT7 APPENDIX 1. ANTHYLLIS VULNERARIA V ARIANTS IN THE BRITISH ISLES WITH MOST OF THE STEM WITH SPREADING HAIRS WEST CORNWALL, V.C. 1 Lizard, 4 June 1926, A. H. Wolley-Dod; plants with red calyces, some of which have spreading hairs but the sheet is named by C. E. Salmon as var. coccinea and not var. corbierei (BM). Housel Bay, Lizard, 6 June 1926, A. H. Wolley-Dod (BM). Caethillean Bay, Lizard, 2 June 1900, no collector (BRISTM). Cadgwith, cliffs, 26 May 1900, no collector (BRISTM). St Agnes cliffs, 27 and 28 May 1913, F. Rilstone; red corollas, appressed hairs on peduncles (NMW). St Agnes cliffs, 19 June 1913, F. Rilstone (NMW). St Agnes Head, 28 May 1989, C. D. Preston/BSBI meeting; very variable population, some plants of which det. as subsp. corbierei by J. R. Akeroyd, 1992 (CGE). Mullion, cliff top on serpentine, 11 June 1969, P. D. Sell, D. E. Coombe & S. Wegmiiller; dense spreading hairs except top peduncle (CGE). Perranporth cliffs, 22 May 1952, T. J. Wallace; mixture of phenotypes present in small area (LIV). Headland by Droskyn Castle Hotel, Perranporth, 26 May 1952, D. A. Reid (K). Hayle Towans, 8 June 1933, A. A. Bullock, red-tipped calyces, and both red and yellow corollas (K; Cullen 1976, cited incorrectly as Hazel Towans). It was also reported from Bedruthan Steps, 1958, J. H. Chandler (Margetts & David 1981; no material traced) and at Porth Joke and Pentire Point East in 1962 (Biological Records Centre; no further details known). EAST CORNWALL, V.C. 2 Tintagel, 23 May 1961, I. W. Evans (BRISTM). SOUTH DEVON, V.C. 3 Salcombe Regis beach, June 1845, R. Cupnell; mixture of plants with spreading and appressed hairs (TCD). NORTH SOMERSET, V.C. 6 Railway embankment near Winscombe, 10 June 1933, H. S. Thompson; spreading hairs throughout, though sparse above (LIV). Congresbury, May 1882, D. Fry; sparsely hairy (BRISTM). DORSET, V.C. 9 Portland Bill, 21 May 1938, herb. Dunston (BM). Old quarry south of Fortuneswell, Isle of Portland, 16 May 1987, S. L. Jury & M. F. Watson no. 8195; four plants with red-tipped calyces, det. J. A. Akeroyd, 1988 (RNG). Car park bank, Easton, Portland, 1999, R. White (specimens shown to T. Rich, not retained). Ballard Down, Studland, 17 May 1959, W. S. Cattling; one plant with red-tipped calyces ‘varying towards subsp. vulneraria...,’ det. J. R. Akeroyd, 1988 (RNG). ISLE OF WIGHT, V.C. 10 Woody Point, 3 July 1925, R. Melville (NMW). Ventnor, June 1888, J. H. A. Stueart; red-tipped calyx, spreading hairs all way up (LIV). Carisbrook Castle, 13 June 1827, no collector (CGE) and 1865, F. Stratton (OXF); material in other herbaria from this site is more variable. KENT, V.C. 15 AND 16 Material with sparse spreading hairs, sometimes along the whole length of the stem, has usually been determined as A. vulneraria subsp. vulneraria var. vulneraria forma polyphylloides Sagorski in a number of herbaria (e.g. Durham Downs, 29 May 1948, J. E. Lousley; det. J. Cullen, CGE). SURREY, V.C. 17 Field on chalk near Merstham, 28 April 1945, N. Y. Sandwith (BRISTM). MONMOUTH, V.C. 35 Coastal bank, Sudbrook, June 2000, T. G. Evans (NMW). WORCESTER, V.C. 37 Broadway, 8 June 1898, L. Rea; sparse spreading hairs throughout (BM). 478 T.CsG-RICH GLAMORGAN, V.C. 41 Baldwin’s Tip, Crymlyn Burrows, 19 June 1939, J. A. Webb; corollas purplish, one of five specimens with spreading hairs (NMW). Taffs Well, railway embankment, 21 June 1927, A. E. Wade (NMW). CARMARTHEN, V.C. 44 Machynys, urban waste ground, 23 June 1992, I. K. Morgan (NMW). PEMBROKE, V.C. 45 Trevine, grassy cliff, 16 June 1963, T. A. W. Davis (BSBI field meeting) ‘calyx concolorous, one plant with calyx-tipped reddish. Plants rather small, those with hairy stems a small proportion of a large population..’; “shaggy form of subsp. vulneraria var. langei... J. Cullen 1963’; some plants also with spreading hairs below only (LIV; NMW). St Anne’s Head, Dale, 5 June 1963, T. A. W. Davis; peduncle with appressed hairs (LIV). CARDIGAN, V.C. 46 Shaley cliffs above the sea west end of Cribach Bay, R. A. E. site, Aber-porth 5 June 1994, A. O. Chater; calyces purple-tipped, plants with appressed hairs also present (NMW). Dry, shaley bank by track above Cribach Bay, R.A.E. site, Aber-porth, 25 May 1997, A. O. Chater; calyx purple- tipped (NMW). Dry shaley track side bank near west end of R. A. E. site, Aber-porth, 3 May 1997, A. O. Chater; calyx purple-tipped (NMW). DENBIGH, V.C. 50 Bryn Euryn, Colwyn Bay, 1966, W. E. Hughes; appressed hairs on peduncles (NMW) FLINT, V.C. 51 Prestatyn, 1923, Miss B. Allen; with plant with appressed hairs (herb. G. Wynne). ANGLESEY, V.C. 52 Holyhead Island, Anglesey, June 1850, J. Ball; red corollas (E). Holyhead, Anglesey, 4 June 1859, J. Ball; det. J. R. Akeroyd 1986 (E). South Stack, June 1916, W. G. Travis (syntype of var. corbierei and holotype of subsp. corbierei LIV; syntype also at BM; with the type at LIV is another specimens with the same details but with appressed hairs, cf. Salmon & Travis 1917). Trearddur Bay, July 1970, A. Jones; mixed forms with spreading and appressed hairs (NMW). South Stack, 28 July 1998, R. A. Jones; with plants with appressed hairs (NMW). Cliffs near Holyhead, 3 June 1960, V. Gordon (exact site not noted, possibly c. ¥2 mile south of South Stack; pers. comm. 1999); mixed gathering, one plant with red corollas and spreading hairs (NMW). Coast near Aberffraw, 9 July 1927, A. Wilson; one specimen only with red corollas and spreading hairs, sparsely so on peduncle (NMW). According to Cullen (1976) there are specimens he collected from South Stack, 29 March 1958 and Rhosneigr, 22 March 1961 in Liverpool University herbarium (LIVU); this collection was transferred to Liverpool Museum (LIV) but the specimens could not be traced in 1999. A collection from Rhosneigr, 9 May 1961, J. M. Scott has all plants with appressed hairs only and was det. by J. Cullen as intermediate between subsp. corbierei and subsp. vulneraria (LIV). MID PERTH, V.C. 88 There are several collections of plants with spreading hairs either all the way up or not including the last peduncle from the Breadalbane Mountains which would otherwise be referable to subsp. lapponica. (e.g. rock ledges, Creag na Caillich, 6 August 1933, J. E Lousley; RNG, and other forms with spreading hairs all way up in OXF and NMW). MID EBUDES, V.C. 103 Staffa, 2 July 1969, British Museum Mull Survey, no. 3706; ‘basically subsp. /apponica .. but also reminiscent of subsp. corbierei, J. Cullen, 1971’ (BM; Jermy & Crabbe 1978). Ardalanish, Mull, 22 May 1971, A. G. Kenneth & A. McG. Stirling; noted both by J. Cullen and J. R. Akeroyd to be subsp. lapponica but varying towards subsp. corbierei (BM). WEST SUTHERLAND, V.C 108 Bettyhill, July 1951, F. Rose; variably hairy, spreading hairs sparse or absent on peduncle (NMW). ANTHYLLIS VULNERARIA SUBSP. CORBIEREI 479 CAITHNESS, V.C. 109 Duncansby Cliff, Caithness, May 1928, R. Meinertzhagen; a distinctive plant with lemon-coloured corollas, and close to corbierei but intermediate with subsp. lapponica (BM). Sea cliffs east of Port of Brims, 1 August 1988, T. C. G. Rich & G. M. Kay, det. J. R. Akeroyd, 1988 (collected as subsp. lapponica; specimen not traced). Holborn Head, | July 1936, J. E. Lousley, det. as subsp. vulneraria but varying towards corbierei, J. R. Akeroyd (RNG). Dunbeathe, riverbank near the sea, 11 July 1968, M. McC. Webster; sparse patent hairs (CGE). CHANNEL ISLANDS Guernsey. Le Jaonnet, 31 May 1970, D. McClintock, det. J. Cullen (herb. Société Guernesiaise; not seen T.C.G.R.). Le Jaonnet, base of steps, June 1999, B. Ozanne; all with spreading hairs, calyx red-tipped (NMW). Le Jaonnet summit, 27 June 1999, R. Rabey; mixed population with spreading and appressed hairs, all corollas yellow and calyx red-tipped (NMW). It was also reported from La Bette Bay adjoining Le Jaonnet Bay by M. English, 1957 (McClintock 1975), but was not refound in 1999 and no material has been traced. Sark. Cliff at Creux harbour, 11 June 1929, F. Ballard & E. A. Golby; red-tipped calyces, and some material with appressed hairs (K; Cullen 1976). Creux harbour, 14 May 1999, R. M. Veall; two plants with red-tipped calyces (NMW). La Coupée, 25 May 1965, A. A. Bullock (K). La Coupée, 4 June 1986, R. J. Murphy; one plant with red-tipped calyx (BM). La Coupée, 11 June 1999, Mrs M. H. Marsden; three plants with red-tipped calyces (NMW). According to some of the literature, only subsp. corbierei is present on the island (Allen & Hilton 1994; Marsden 1994). IRELAND No material has been found in all the herbaria examined, though a very few plants with sparse spreading hairs below have been seen (e.g. Killiney Hill, 1849, no collector, TCD). J. R. Akeroyd has also not seen any plants with spreading hairs in the field (pers. comm., 2001). 480 T. C. G. RICH APPENDIX 2. ANTHYLLIS VULNERARIA FORMS WITH SPREADING HAIRS IN NORTHERN FRANCE MANCHE Jobourg et Herqueville, 4 Juillet 1883, herb. L. Corbiére (as var. sericea Breb.); mixed collection, one with spreading hairs throughout and others with spreading hairs below (CHE). Herqueville, falaises, 20 Juin 1886, herb. L. Corbiere (as var. sericea Breb.); spreading hairs throughout though sparsely above (CHE). CALVADOS Cotes de Calvados, dans le falaises, 1838, A. Lenormand (P). Pentes des falaises calcaires du rivage de la Manche a Fontenailles, 18 Juin 1890, Bertot; spreading hairs throughout (CHE, P). Falaises maritime, Longues prés Bayeux, 8 Aout 1872, L. Corbiére (as var. sericea Breb.); spreading hairs throughout though very sparsely above (CHE). Falais d’ Arromancher, Calvados, Juillet 1846, E. Cosson, det. “forma inter subsp. vulgaris (Koch) Hayek et subsp. polyphylla (Kit.) Cab.’ by J. Cullen, 1961 (P). SIENE-MARITIME Falaises de Fécamp, Siene-Inferieur, 12 Juin 1892, I. Thériot & L. Corbiére (as var. sericea Breb.); sparse spreading hairs throughout (CHE, P) and I. Thériot; one plant with spreading hairs throughout and others with spreading hairs below (CHE). PAS-DE-CALAIS Pernes, 19 July 1995, B. Destines (Herbarium CRP/CBNBL). Setques, 8 June 1995, B. Toussaint (Herbarium CRP/CBNBL). Watsonia 23: 481-491 (2001) 48] The distinction between Cerastium tomentosum L. and C. biebersteinii DC. (Caryophyllaceae), and their occurrence in the wild in Britain M. K. KHALAF and C. A. STACE Department of Biology, University of Leicester, Leicester LEI 7RH ABSTRACT Differences between Cerastium tomentosum and C. biebersteinii are summarised. Due to the enormous variation of the former in most readily observable characters, certain identification can be effected only by using features of the morphology of the capsule teeth (not present in most material) and the anatomy of the non-glandular hairs (a microscopic character). No material of C. biebersteinii has been seen from the British Isles other than in a few botanic gardens, and there is no good evidence of its existence here either as a garden ornamental or in the wild, nor of its hybridization with C. tomentosum. Hybrids between C. tomentosum and C. arvense are frequent both in cultivation and in the wild. KEYWORDS: Cerastium, morphology, chromosome numbers, hair anatomy, hybridization. INTRODUCTION Cerastium tomentosum L. and C. biebersteinii DC. belong to a group of up to twenty species that can conveniently be termed the C. tomentosum group. This was recognised by Schischkin (1936) as series Grandiflora (Borza) Schischk. in the following classification: Subgenus Dichodon (Bartl.) Fenzl Subgenus Eucerastium (Boiss.) Pax (= Cerastium) Section Strephodon Ser. Section Schizodon (Fenzl) Schischk. Section Orthodon Ser. (= Cerastium) Subsection Fugacia (Fenzl) Pax & K. Hoffm. Subsection Perennia (Fenzl) Pax & K. Hoffm. (= Cerastium) Series Lasiostemona (Fenzl) Schischk. Series Leiostemona(Fenzl) Schischk. Series Alpina (Borza) Schischk. Series Arvensia (Hayek) Schischk. (= Cerastium) Series Grandiflora (Borza) Schischk. The names of four of the above taxa have to be corrected to Cerastium, as shown, since C. arvense L. is the type of the genus. None of these five series is well defined, however, and certain hybrids between species in the last three can be synthesised with ease (Khalaf & Stace 2000). The C. tomentosum group was defined by Buschmann (1938) by its perennial habit, presence of many non-flowering axillary shoots, relatively large flowers and presence of (usually very dense) long white hairs. Although these characters are not strictly diagnostic of the group, either individually or in combination, all apply to C. tomentosum and C. biebersteinii, which can be considered among the core species of this informal grouping. Subsection Perennia (= Cerastium), as redefined by Sdllner (1954) when he removed series Leiostemona (which includes C. fontanum Baumg.), seems to be a well defined taxon which is best not subdivided above the species level (Khalaf & Stace 2000). Cerastium tomentosum itself is an extremely variable taxon, and there have been considerable divergences of opinion about its delimitation. Moggi (1963) pointed out that at least six species 482 M. K. KHALAF AND C. A. STACE had been segregated from it in Italy and Sicily alone. On the other hand the name has been used for plants from Spain (e.g. Linnaeus 1753), the Alps (e.g. Nyman 1855), the Balkans (e.g. Stojanoff & Stefanoff 1924), the Crimea (fide Schischkin 1936) and the Caucasus (e.g. Fenzl 1842). Today it is recognised as endemic to Italy and Sicily, where no other taxa from the C. tomentosum group occur. Of all its variations, specimens from higher altitudes on Mt Etna, Sicily, are the most distinct in having only sparse hairs, narrower leaves, shorter petals and capsules and smaller seeds; these can be recognised as var. minus C. Presl (var. aetnaeum Janka). The only other Linnaean species possibly in the C. tomentosum group, Cerastium repens L. from France and Italy, is a name of uncertain application. No type specimen has been traced; the specimen in LINN is not a type and is in our opinion referable to C. arvense, a conclusion apparently behind De Candolle’s (1822) inclusion (with some doubt) of the name under the latter species. It has also been used for the Italian species (C. tomentosum) by Buschmann (1938) and for the Crimean species (C. biebersteinii, see below) by Marschall von Bieberstein (1808). The first species recognised in the group from outside the Mediterranean area were C. repens M. Bieb. (from Crimea) and C. argenteum M. Bieb. (from Tiflis, now Tbilisi in Georgia), described by Marschall von Bieberstein (1808). The Crimean species was realised by De Candolle (1822) to be different from C. repens L., so he renamed it C. biebersteinii. De Candolle’s material consisted of cultivated plants originating from material collected by Steven in Crimea. Buschmann (1938) noted “I have seen the plants which were collected by Steven in Tauria [Crimea] in 1821. He called it C. repens and certainly it belongs to C. biebersteinii’. Schischkin (1936) stated that “C. biebersteinii is an endemic Crimean plant, very widespread in the mountainous part of the Crimea. It displays the closest relationship to C. szowitsii Boiss. (Eastern Transcaucasia) and C. argenteum M. Bieb. (Central Transcaucasia) which are similarly white-tomentose but differ in some other characters. Early investigators identified the Crimean plant with C. tomentosum, although this species is confined to the central and southern parts of Italy. Cerastium biebersteinii does not occur in a wild state outside the Crimea. The report by Taliev and Voinovskii (1902) referring to the occurrence of this species in inundated meadows of Buguruslan County of Samara Province [north of the Caspian Sea in Ukraine] is certainly wrong”. However, Buschmann (1938) considered that C. biebersteinii does occur in Transcaucasia, since he favoured a wider taxonomic concept of it, including two of Schischkin’s Caucasian species in his C. biebersteinii subsp. transcaucasicum Buschm. According to Schischkin (1936) the only other member of the C. tomentosum group to occur in Crimea is C. stevenii Schischk., which is a very different-looking species and perhaps does not belong to this group at all. It is therefore likely that C. tomentosum and C. biebersteinii each occur in widely separated areas neither of which supports any other closely similar taxa. C. tomentosum var. tomentosum has long been introduced in central and northern Europe as a garden ornamental. From here it has escaped into the wild and in places has become extremely well naturalised not only in marginal and disturbed ground but sometimes in natural habitats, such as sandy grassland and maritime sand-dunes. Already by 1960 it was found scattered in the wild all over the British Isles (Perring & Walters 1962). It has caused taxonomic problems in two ways. Firstly, it has quite frequently hybridised (in eastern and south-eastern England and one part of central Scotland) with the native C. arvense (belonging to a different series from C. tomentosum, see Classification above) to produce plants that superficially resemble another member of the C. tomentosum group, the Balkan C. decalvans Schlosser & Vuk., and which have been wrongly so called. Hybrids between C. tomentosum and C. arvense have also been reported from Canada (Morton 1975), Sweden (Karlsson 1997) and Luxemburg and Finland (Kaleva 1966), as well as in several botanic gardens throughout Europe. Secondly, there are several reports of the introduction to gardens of the Crimean C. biebersteinii DC., and the variability of C. tomentosum has often been partly explained by determination of some naturalised plants as this or as various Balkan species of the group. Clapham (1987) stated that C. biebersteinii “is also much grown in gardens” and that “some cultivated material may well be of hybrid origin”. Walters (1989) claimed that both C. biebersteinii and Caucasian taxa were introduced in the 19th century and that “complex hybridization seems to have taken place in cultivated plants”. As part of a study of the breeding relationships of subsection Perennia sensu Schischkin, we attempted to describe precisely the differences between C. tomentosum and C. biebersteinii and to investigate the occurrence in the wild of these two taxa in Britain. CERASTIUM TOMENTOSUM & C. BIEBERSTEINII 483 Due to their wide geographical separation the two species are not often treated in the same Flora. Fenzl (1842) covered both species, but his C. tomentosum from “provinciis caucasicis. Iberia” must have been misapplied. Boissier (1867) stated that C. biebersteinii has a denser tomentum, longer narrower sepals and capsule teeth with flat (not revolute) margins. Buschmann (1938) separated C. biebersteinii from a group of species including C. tomentosum in his key (couplet 8): capsule straight, strong-walled, with patent teeth flat on the margins in the former; capsule straight or curved, soft-walled, with erect teeth revolute at margins in the latter. Jalas et al. (1964) agreed about the narrower sepals and flat capsule-teeth in C. biebersteinii, and added shorter stems, shorter and narrower leaves and erect (not patent, contrary to Buschmann) capsule teeth. Walters (1989) included both species in the “C. tomentosum group” and did not attempt to distinguish them. Kaleva (1966) is apparently the only person to have studied these species in a scientific and comparative way. He listed eight differences: growth habit; leaf size and shape; petal shape; capsule size; capsule wall anatomy; seed size; pollen grain size; and chromosome number. However, he concluded that the revolution of the margins of the capsule teeth is not a good discriminator (“+ flat” in both), and he did not mention capsule-teeth posture. Nevertheless his photographs showed clearly more patent capsule teeth in C. tomentosum, in agreement with Jalas et al. (1964) but not with Buschmann (1938). MATERIALS AND METHODS Living accessions were obtained as plants or seeds by collecting in the wild, from correspondents in Continental Europe, and by the seed-exchange schemes. Frequently material from the last of these sources was misidentified. Much herbarium material was also examined. Up to 25 measurements were made of each character for each accession. Unfortunately, our observations on C. biebersteinii were restricted to two living accessions and about a dozen herbarium specimens, few of which carried ripe capsules. Seed accessions were germinated at 15—19°C in a 3:7:2 loam:peat:grit mixture under a light regime of 16 hr light/8 hr dark. Germination occurred between 4 and 7 days; seedlings were potted on and after 1-2 months were transferred to an unheated glasshouse where they were kept permanently. Plants flowered much more profusely after vernalization over winter in a cold glasshouse. Vouchers of all accessions at flowering are deposited in LTR. Chromosome counts were made on root-tips obtained from fresh stem-cuttings after 10—15 days; these possessed larger meristems and a higher mitotic index than root-tips gathered from mature plants or from seedlings. They were pretreated in 50% saturated aqueous paradichlorbenzene, washed in tap-water, fixed for 3 hr in ethanoic acid:ethanol (1:3), washed again and stored in 70% IMS. After storage they were rewashed, hydrolysed in SN HCI at room temperature for 3—6 min, washed in 70% ethanol, and transferred to a drop of 45% ethanoic acid on a slide. The meristematic region was dissected from the rest of the tissue and transferred into a drop of 2% aqueous aceto-orcein on another slide. The meristem was dissected apart and squashed in the usual way and then gently heated over a spirit lamp to flatten the cells, and the coverslip was sealed with rubber solution. Suitable spreads were photographed. Thirty characters (Table 2) were measured by us for all the material at our disposal to see whether any discrimination could be achieved. Means per plant, overall ranges or ranges of means were used according to which we considered the most diagnostically useful. RESULTS CHROMOSOME NUMBERS Table 1 shows that all our accessions from the wild in Britain of C. tomentosum var. tomentosum, C. arvense and their hybrids had 2n = 72, representing the octoploid level, except that six of the seventeen accessions of C. tomentosum had up to nine B-chromosomes. Octoploid counts are also the commonest ones reported in the literature for C. tomentosum and C. arvense, but in addition there are tetraploid and dodecaploid counts (2n = 36 and 108) for C. tomentosum, and tetraploid and decaploid (2n = 36 and 90) for C. arvense (Khalaf & Stace 2000). However, the three previous counts of C. arvense and one of C. tomentosum from Britain (Brett 1955, Blackburn & Morton 484 M. K. KHALAF AND C. A. STACE TABLE 1. CHROMOSOME COUNTS OF WILD MATERIAL FROM THE BRITISH ISLES OF CERASTIUM TOMENTOSUM VAR. TOMENTOSUM, C. ARVENSE AND THEIR HYBRID Site Vice-county Grid Ref. Count C. tomentosum var. tomentosum CER 4 Gorebridge Midlothian NT353605 2n = 72 + 1-3B CER 5 Blackford Quarry Midlothian, NT261702 2n = 72+ 5B CER 206621 Shingle, Snettisham W. Norfolk TF648328 2n= 742 CER 25 Shingle, Snettisham W. Norfolk TF648333 2n= 12 CER 26 Shingle, Snettisham W. Norfolk TF648333 2n=72 + 2B CER 31 Sand-dunes, Holme W. Norfolk TF702442 on = 72 CER 103 Near Dumbarton Rock Dunbarton NS37 Dnj=12 CER 104 Sea front, Crosby S. Lancashire SJ39 2ni= 12 CER 105 Behind beach, Southwold E. Suffolk TM507753 7A a CER 106 Wall, Old Town, St Mary’s Scilly SV912013 2n=72+7B CER 118 Wall-top, Llaniestyn Caernarvonshire SH282337 2n = 72 + 9B CER 126 Near railway, Great Heck, York M.W. Yorkshire SE5920 2n= 72 CER 139, 140 & 141. Heathland, near Ipswich E. Suffolk T 14 A CER 143 Hedgerow, Orrisdale Isle of Man S$€325930 2n = 72+ 9B C. arvense CER 27 Grassy roadside, near Snettisham W. Norfolk TF669336 272 CER 29 Sand-dunes, Holme W. Norfolk TF700442 n= 72 C. arvense x C. tomentosum CER 6 Borthwick Bank Midlothian NT369603 n= 12 CER 19 Shingle, Snettisham W. Norfolk TF648328 2n= 72 CER 23 Shingle, Snettisham W. Norfolk TF648330 Pi 12 CER 24 Shingle, Snettisham W. Norfolk TF648333 on =z CER 28 Sand-dunes, Holme W. Norfolk TF700442 2m= 2 CER 30 Sand-dunes, Holme W. Norfolk TF701442 hive 1957) have all been octoploids, as have the numerous counts we have made of British garden material of C. tomentosum. The other members of the C. tomentosum group vary from tetraploid (2n = 36) to 16-ploid (2n = 144), and several species are represented by more than one ploidy level (Khalaf & Stace 2000). C. biebersteinii has been reported as tetraploid (one count of wild Crimean material, Kaleva 1966) and as octoploid (three counts of botanic garden material); this species has been widely misidentified and the true identity of the three cultivated octoploids requires confirmation. We used two accessions of C. biebersteinii, one of wild South Crimean (“Tauria meridionalis”) and the other of Miinchen botanic garden origin; both were tetraploids (2n = 36). MORPHOLOGICAL AND ANATOMICAL CHARACTERS The vast range of morphological features of wild Italian material of C. tomentosum var. tomentosum strongly suggested to us that many of the characters of C. biebersteinii would be covered by those of the former taxon, and that few of them would be likely to distinguish the two species. The results shown in Table 2 can be classified as follows: i - characters showing differences in the means, or in a good part of the ranges, between the two species, but with overlap, viz. sepal length and breadth, pollen grain diameter, capsule length, capsule shape, capsule wall rigidity, capsule tooth length and breadth, capsule tooth margins, and seed length (also seed breadth, not listed in Table 2). ii - characters apparently clearly discriminating the two species, viz. presence of rhizomes and stolons, leaf thickness, filament length, anther length, ovary length, and seed weight. iii - characters not discriminating between the two species, viz. 13 of the other 16 characters (the three hair characters are covered below). CERASTIUM TOMENTOSUM & C. BIEBERSTEINII 485 TABLE 2. DIAGNOSTIC CHARACTERS OF CERASTIUM TOMENTOSUM VAR. TOMENTOSUM AND C. BIEBERSTEINII | | Character C. tomentosum var. tomentosum C. biebersteinii Height of flowering stems to tip of highest 200-450 mm 100-300 mm flower* Presence of rhizomes and stolons Both Neither Length of uppermost internode of flowering stem, 13-3-120 mm 63-7-125 mm i.e. internode between uppermost leaf-pair and lowermost bract-pair Mean length of next 3 lower internodes 7-2-57 mm 8-4-69-5 mm Length and breadth of uppermost pair of leaves 749-7 x 17-8-42 x on flowering shoot 1-5-8-5 mm 1-9-7 mm Thickness of same leaves 0- 1-0-2 mm 0-2—0-3 mm Length and breadth of lowest bracts 4-9.7 x 24 mm 5-9-5 x 2:54 mm Length of scarious tip of lowest bracts 0-2-1-8 mm 0-5-0-9 mm No. flowers per inflorescence 2-6-12-2 4-3-10-5 Pedicel length of terminal flower 12-7-42-8 mm 13-7-47 mm Sepal length and breadth 4-9 x 2-3-4 mm 6-5-9-7 x 2-2-7 mm Length of scarious tip of sepals 0-2-1-3 mm 0-5—1-1 mm Petal length and breadth 11-18 x 5-10 mm 17-17:3 x 6-8-7 mm Ratio petal cleft length/petal length* 0-2-0-3 0-2-0:3 Filament length* 4—5 mm 5-4-5-8 mm Anther length* 1-1-1 mm 0-7—0-9 mm Pollen grain diameter (tetraploid) 34-3-36-5 um 35-8-37 um Pollen grain diameter (octoploid) 35-3—40-1 um - Pollen grain diameter (dodecaploid) 43—45-5 um - Ovary length* 2-2-5 mm 1-7-1-:9 mm Style length* 3-6 mm 34-4 mm Capsule length* 8-12 mm 10-15 mm Capsule shape Straight or bent Straight Capsule wall rigidity Pliable Pliable or rigid Capsule tooth length, 0-7-1-3 x 1-1-5 x and breadth (0-2 mm behind apex)* 0-1—0-3 mm 0-3-0-5 mm Capsule tooth posture Capsule tooth margins Erect to patent mostly revolute Erect to patent mostly + flat Seed length (from hilum to furthest point)* 1-1-9 mm 1-5—2-2 mm Seed weight per 100 seeds 33-75-42 mg 67—70-5 mg Upper leaf-surface hair density (1) 5-62:8 4-426 Lower leaf-surface hair density (1) 6-6-5] 12-8-30-2 Length-ratio terminal/subterminal hair-cells (3-7)4—19(19-8)** (1)1-1-1-7(1-8)** Ranges of means per plant, or (*) overall ranges, or (**) range of means with overall ranges in parentheses (1) Number of hair-bases in circle of diameter of 250 um (0-05 mm”) at midpoint of uppermost pair of leaves on flowering shoot HAIR CHARACTERS Both C. tomentosum var. tomentosum and C. biebersteinii possess only unbranched non-glandular uniseriate-celled hairs. The great range in visible hair density in the former suggested that this character would be of no value in distinguishing between the two taxa, and precise counts bear this out (Table 2). However, a good difference in hair anatomy was discovered between these two taxa (Table 2). The uniseriate hairs consist of three to six (usually four or five) cells, of which the terminal one is very long and the basal one or two (rarely three) are very short. The subterminal cell is much shorter than the terminal cell in C. tomentosum var. tomentosum (and also in C. tomentosum var. minus) but only slightly so in C. biebersteinii; the ratio between the two cells clearly separates the 486 M. K. KHALAF AND C. A. STACE | Z| \ C4 = FIGURE 1. Representative leaf-hairs of Cerastium. A: C. tomentosum. B: C. biebersteinii. two species in all cases (Table 2; Fig. 1). Of the other seven species of the C. tomentosum group examined in this study, none agrees with C. tomentosum in this character, five (C. argenteum M. Bieb., C. decalvans Schloss. & Vuk., C. gibraltaricum Boiss., C. lineare All., C. moesiacum Friv.) agree with C. biebersteinii , and two (C. candidissimum Correns, C. grandiflorum Waldst. & Kit.) differ from all the others in having branched hairs (of quite different form in the two species - dendroid in C. candidissimum, T-shaped in C. grandiflorum). In addition some or all plants of C. gibraltaricum, C. moesiacum and C. decalvans possess unbranched, uniseriate-celled glandular hairs, which are also present in C. tomentosum var. minus. CERASTIUM TOMENTOSUM & C. BIEBERSTEINII 487 DISCUSSION Our chromosome counts lead us to believe that British wild and cultivated material of the above three species mainly or entirely consists of tetraploid C. biebersteinii and octoploid C. tomentosum and C. arvense, although of course some clones at other ploidy levels might exist as yet undetected. In general appearance C. biebersteinii is a shorter, less mat-forming, more sturdy plant than C. tomentosum, with larger leaves, a denser, more matted tomentum and longer sepals. However, the ranges of many characters of C. tomentosum are wider than those of C. biebersteinii and in several cases completely overlap them. This is due not only to the much greater variation shown by C. tomentosum but also to the relatively small number of specimens of C. biebersteinii that were examined. In general our measurements, even the means, show wider ranges than those of Kaleva (1966), presumably due to our greater sample size and possibly to the fact that our samples of C. tomentosum included tetraploids and dodecaploids as well as octoploids. Because of this, some characters previously claimed to separate the two species were found by us to be of virtually no value, viz. leaf length and breadth, leaf-hair density, petal length and breadth, depth of petal cleft and capsule tooth posture. Clapham (1987) and Walters (1989) suggested that there has been hybridization between C. tomentosum and C. biebersteinii in cultivation; their views are plausible, but we are unable to substantiate them. We have found no evidence that any spontaneous hybrids between these two species exist. In our experiments (Khalaf & Stace 2000) no capsules were produced from crosses between tetraploid C. tomentosum and C. biebersteinii. In crosses using octoploid C. tomentosum as the male parent some hybrid seeds were produced, of which 7:7% was viable, but no mature hybrid plants could be obtained. But when octoploid C. tomentosum was used as the female parent 74% of the hybrid seed was viable and mature flowering hybrid plants were obtained. Moreover, these were fertile to some degree, despite being hexaploids (2n = 54). Specimens have been deposited in LTR. All the hybrids obtained in the C. tomentosum group resembled their female parent more closely than their male parent. In the case of the C. tomentosum x C. biebersteinti hybrids, the length ratio of the terminal/subterminal cells of the non-glandular hairs was well within the range of that of C. tomentosum. Hybrids between C. arvense and C. tomentosum occur in the wild in Britain as well as in botanic and other gardens. They have no valid binomial. They are fertile octoploids and have the potential of extensive and repeated back-crossing. Similar fertile hybrids were synthesised reciprocally by us with ease. They can be distinguished from C. tomentosum by their greener appearance, with much less dense non-glandular hairs, frequent possession of glandular hairs, and less sturdy growth habit. These characters are shared by C. tomentosum var. minus, which might have such a hybrid origin. The 16 characters that show some promise as distinctions between C. biebersteinii and C. tomentosum var. tomentosum are briefly discussed below. Presence of rhizomes and stolons Rhizomes and stolons are certainly more evident and far-reaching in C. tomentosum, but we cannot be certain of their absence in C. biebersteinii. Field studies in Crimea are required to elucidate this. Thickness of uppermost pair of leaves on flowering shoot According to our measurements this character separates the two species, but further sampling is required to see whether there is some overlap. In any case this is not a character that is easy to measure for identification purposes. Sepal length and breadth The sepals are usually longer and narrower in C. biebersteinii, but the ranges of both measurements overlap. The length/breadth ratio provides a better discriminator, but there is still overlap. Kaleva’s (1966) sepal length measurements were contained within ours and gave a clear separation of the two species, but with a sample size of only 10. 488 M. K. KHALAF AND C. A. STACE Filament, anther and ovary lengths These all gave good separation in our material, but further sampling is desirable to ensure that there is no overlap. The ovary must be measured at anthesis, before there is any chance of its swelling after pollination. Pollen grain diameter The pollen of the tetraploid C. biebersteinii is similar in size to that of tetraploid and octoploid C. tomentosum, but that of dodecaploid C. tomentosum is slightly larger. Kaleva’s (1966) measurements were very close to ours. Capsule length Again, there is a difference, but with overlap. Kaleva’s (1966) measurements were very close to ours. Capsule shape Whereas C. biebersteinii always has straight capsules, those of C. tomentosum can be straight or bent/curved. Capsule wall rigidity The walls of the ripe empty capsules of C. ftomentosum are rather soft and pliable, whereas those of C. biebersteinii are harder and more rigid, often fracturing when pressed with a needle. Kaleva (1966) showed that this difference is based on the capsule wall anatomy; in C. biebersteinii the lignified outer wall of the outer epidermal layer is at least twice as thick as in C. tomentosum. However, we found that some capsules of C. biebersteinii had pliable walls, although this might be due to the conditions under which the plants grew. Field studies in Crimea are required to check this. Capsule tooth length and breadth The capsule teeth are generally longer and (especially) wider in C. biebersteinii, but there is some overlap in length and ranges of breadths are contiguous. Breadth measurements are difficult to compare as the point at which the measurement is taken is critical. Capsule tooth margins This is the most frequently cited diagnostic character, and we found it reliable (Fig. 2). The greater revolution in C. tomentosum renders the apparent capsule tooth breadth more distinct from that of C. biebersteinii than it would otherwise be. Kaleva (1966), however, found this character to be unreliable, stating that both species possessed “+ flat” tooth margins. But his “C. tomentosum strain B”, which he very plausibly concluded was C. tomentosum x C. arvense, had “revolute” tooth margins. The capsule teeth in C. arvense are even more revolute at the margins than in C. tomentosum, and it is therefore possible that Kaleva differed from us simply in his verbal interpretation of the same structures. Seed length The ranges of the two species differ, but overlap. Kaleva’s (1966) measurements agree closely with ours. Seed weight Seed weight gave a good separation between the two species; this is a combination of seed length, breadth and thickness plus any possible anatomical differences. Length ratio of terminal and subterminal hair-cells A very clear separation was obtained between C. tomentosum and C. biebersteinii (along with all other species in the group) (Fig. 1). The difference is so clear that after some experience it is not necessary to make detailed measurements. Chromosome number Octoploid counts of C. biebersteinii need to be confirmed among wild collections; until then it 1s safer to consider this species a tetraploid. C. tomentosum var. tomentosum is usually octoploid and var. minus always so. Rarely tetraploid and dodecaploid populations of var. tomentosum are reported; all our material of these two cytotypes was of botanic garden material, but according to Favarger (1972) all three ploidy levels occur in the wild in the Italian Apennines. CERASTIUM TOMENTOSUM & C. BIEBERSTEINII 489 FIGURE 2. Dehisced capsules of Cerastium. A: C. tomentosum. B: C. arvense. C: C. biebersteinit. CONCLUSIONS All the material of the C. tomentosum group that we have seen either from the wild or cultivated as ornamentals in Britain has been referable to C. tomentosum var. tomentosum or to hybrids between that and C. arvense. Despite the apparent absence of C. tomentosum var. minus and C. biebersteinii, all three taxa are described here in order to allow future identification: CERASTIUM TOMENTOSUM L. VAR. TOMENTOSUM Plant mat-forming, with stolons and/or rhizomes, almost completely covered with simple non- glandular hairs; length ratio of terminal to subterminal cell (3-7)4—19(19-8); glandular hairs absent; flowering stems 20—45cm, their uppermost internode 13-3—120 (135) mm. Uppermost leaves on flowering shoots linear to oblong- or elliptic-lanceolate, 7-49-7(54) x 1-5-8-5(19) mm, 0-1-0-2 mm thick, acuminate or acute to subobtuse at apex, greyish-green to greyish white, with hair density on upper and lower surfaces (4)5—62-8(68) and (5)6-6—51(55) per 0-05 mm? respectively. Inflorescence contracted to spreading, with (1)2-6—12-2(16) flowers; pedicel of terminal flower (10)12-7-42-8(60) mm; lowest bracts 4—9-7(10) x 2-4 mm, with scarious tip (0-1)0-2—1-8(2) mm. Sepals 4—9(9-2) x 2—3-4(4) mm, lanceolate to oblong- or ovate-lanceolate, with scarious tip 0-2— 1-3(1-5) mm, subacute to obtuse at apex. Petals (10-3)11—18(18-5) x (4-5)5—10(11) mm, with cleft 3-6 mm. Filaments (4)4-2—4-6(5) mm; anthers 1-1-1 mm; pollen grains (32-5)34-3-36-5(38) um in tetraploid, (35)35-3-40-1(42) um in octoploid, and (42)43—45-5(50) um in dodecaploid. Ovary (2) 2-2—2-4(2:5) mm; styles (3)3-2—5-7(6) mm. Capsules 8—11(12) mm, straight to bent, with pliable walls, with teeth (0-7)0-8—1-2(1-3) mm and (0-1)0-2-0-3 mm wide and with revolute lateral margins. Seeds (1)1-19-1-6(1-9) mm. 2n = 36, 72, 108. CERASTIUM TOMENTOSUM VAR. MINUS C. PRESL Plant mat-forming, with stolons and/or rhizomes; flowering stems 5—20cm, their uppermost internode (20)33-42-5(46) mm and with simple non-glandular hairs, simple glandular hairs, or both. Uppermost leaves on flowering shoots linear to lanceolate, (8-3)9-30 x 1—3 mm, 0-2—0-3 mm thick, acute to acuminate at apex, greyish-green to green, with simple non-glandular hairs only, with hair density on upper and lower surfaces (4)4-8-13-4(16) and (5)5-3-6-8(18) per 0-05 mm? respectively; length ratio of terminal to subterminal cell (6)6-5—15-8(16). Inflorescence contracted, with (2)3-3-4-1(7) flowers; pedicel of terminal flower 10—19-3(23) mm; lowest bracts (4)4-5—7-3 (7-5) x 2-3 mm, with scarious tip (0-1)0-2-0-4(0-7) mm. Sepals (4:5)5—8 x 2-2-2(2-5) mm, lanceolate to oblong-lanceolate, with scarious tip (0-2)0-3—1 mm, subacute to acute at apex. Petals (8-9)9-11-7(12) x (5-3)6(7) mm, with cleft 3-4-1(4-5) mm. Filaments (3-8)4—4-1 mm; anthers 1— 1-1 mm; pollen grains (35)35-5—39-2(42) um. Ovary (1-8)1-9-2 mm; styles (3)3-1-3-5(4) mm. Capsules 8—8-1(8-2) mm, straight to bent, with pliable walls, with teeth 1-1-1 mm long and 0-1—0-2 mm wide and with revolute lateral margins. Seeds (0-9)1-18—1-4(1-5) mm. 2n = 72. 490 M. K. KHALAF AND C. A. STACE CERASTIUM BIEBERSTEINII DC. Plant forming tussocks or small mats, without(?) stolons or rhizomes, almost completely covered with simple non-glandular hairs; length ratio of terminal to subterminal cell (1)1-1—1-7(1-8); glandular hairs absent; flowering stems 10-—30cm, their uppermost internode (53)63-7—125(160) mm. Uppermost leaves on flowering shoots lanceolate to oblong-lanceolate, (14)17-8-42(50) x (1-5)1-9-7 mm, 0-2—0-3 mm thick, acute to obtuse at apex, white to greyish white, with hair density on upper and lower surfaces (4)4-4—26-8(29) and (10)12-8-30-2(32) per 0-05 mm? respectively. Inflorescence contracted, with (3)4-3—10-5(12) flowers; pedicel of terminal flower (13)13-7-47 mm; lowest bracts 5-9-5 x 2-5-4 mm, with scarious tip 0-5—0-9 mm. Sepals 6-5—9-7 (10) x 2—2-7(3) mm, lanceolate to oblong- or ovate-lanceolate, with scarious tip 0-5—1-1(1-2) mm, acute to obtuse at apex. Petals (16-5)17—17-3(17-5) x (6-5)6-8-7(7-1) mm, with cleft (3-5)4(4-2) mm. Filaments (5-4)5-5—5-7(5-8) mm; anthers (0-7)0-8—0-9 mm; pollen grains (32-5)35-8—37(37-8) um. Ovary (1-7)1-8-1-9 mm; styles (3)3-1-4-1(4-4) mm. Capsules 10(10-1)-14(15) mm, straight, with usually rigid walls, with teeth (1)1-2—1-4(1-5) mm and (0-3)0-4-0-5 mm wide and with + flat lateral margins. Seeds (1-5)1-83-1-87(2-2) mm. 2n = 36, (?72). IDENTIFICATION Despite the enormous range of variation of C. tomentosum in many morphological characters, after some familiarization with these two species it is possible to distinguish between them in the great majority of cases using the following characters as clues: growth habit; leaf size, thickness and indumentum; sepal size; filament, anther and ovary length; capsule, capsule-tooth and seed size, and capsule wall rigidity. It is very difficult, however, to express these as opposing character-states in a key, and a fully confident identification cannot be guaranteed using them alone. Study of a much wider range of wild material of C. biebersteinii is required before we can be sure of the diagnostic value of the above characters. Notwithstanding the conclusions of Kaleva (1966), we have found the revolution of the lateral margins of the capsule teeth to be absolutely reliable (Fig. 2). Unfortunately most herbarium specimens do not carry ripe capsules, and it seems that these are not always produced. This is not due, however, to self-incompatibility, as was surmised by Walters (1989), but to lack of self- pollination in these highly protandrous but self-compatible plants (Khalaf & Stace 2000). The lack of capsules does, of course, present a problem in certain identification. In such cases the hair anatomy (Fig. 1) will provide an unequivocal answer, since this character always clearly differentiates between the two species. REFERENCES BLACKBURN, K. B. & MorTON, J. K. (1957). The incidence of polyploidy in Caryophyllaceae of Britain and Portugal. New Phytologist 56: 344-351. BOISSIER, E. (1867). Cerastium, in Flora Orientalis 1: 712—730. Herbier Boissier, Geneva. BRETT, O. E. (1955). Cytotaxonomy of the genus Cerastium, 1. Cytology. New Phytologist 54: 138-148. BUSCHMANN, A. (1938). Uber einige ausdauernde Cerastium-Arten aus der Vermandtschaft des C. tomentosum L. Feddes Repertorium 43: 118-143. CLAPHAM, A. R. (1987). Cerastium, in CLAPHAM, A. R., TUTIN, T. G. & WARBURG, E. F. Flora of the British Isles, 3rd ed., pp. 131-136. Cambridge University Press, Cambridge. DE CANDOLLE, A. P. (1822). Sur les plantes rares ou nouvelles du jardin de botanique. Mémoires de la société physique et d’histoire naturelle de Genéve |. FAVARGER, C. (1972). Contribution a |’étude cytotaxonomique de la flore des Apennins, 2. Le groupe de Cerastium tomentosum L. Saussurea 3: 65-71. FENZL, E. (1842). Cerastium L., in LEDEBOUR, C., ed., Flora Rossica 1: 396-416. E. Schweizerbart, Stuttgart. JALAS, J., SELL, P. D. & WHITEHEAD, F. H. (1964). Cerastium L., in TUTIN, T. G., HEYwWoop, V. H. et al., eds., Flora Europaea 1: 136-145. Cambridge University Press, Cambridge. KALEVA, K. (1966). Biosystematic notes on Cerastium Biebersteinii DC. and some ornamental strains of C. tomentosum L. (Caryophyllaceae). Annales Botanici Fennici 3: 100-109. KARLSSON, T. (1997). Forteckning Over svenska karlvaxter. Svensk Botanisk Tidskrift 91: 241-560. KHALAF, M. K. & STACE, C. A. (2000). Breeding systems and relationships of the Cerastium tomentosum group. Preslia, Praha 72: 323-344. LINNAEUS, C. (1753). Species Plantarum. Laurentius Salvius, Stockholm. MARSCHALL VON BIEBERSTEIN, M. (1808). Flora Taurico-caucasica 1. Charkow. CERASTIUM TOMENTOSUM & C. BIEBERSTEINII 49] Moca, G. (1963). Considérations géographiques et systématiques sur la flore d’Italie du sud. Webbia 118: 65-72. Morton, J. K. (1975). Cerastium L., in STACE, C. A., ed., Hybridization and the Flora of the British Isles, pp. 171-174. Academic Press, London. = NYMAN, C. F. (1855). Sylloge Florae Europaeae. N. M. Lindh, Orebro. PERRING, F. H. & WALTERS, S. M. (1962). Atlas of the British Flora. Thomas Nelson, London. SCHISCHKIN, B. K. (1936). Cerastium L., in KOMAROV, V. L. & SCHISCHKIN, B. K., eds., Flora URSS 6: 430-466. Izdatel’stvo Akademii Nauk SSSR, Moscow & Leningrad. SOLLNER, R. (1954). Recherches cytotaxonomiques sur le genre Cerastium. Bulletin de la Société Botanique Suisse 64: 221-354. STOJANOFF, N. A. & STEFANOFF, B. (1924). Flore de la Bulgarie. Minist. Agric. Bulg., Sofia. WALTERS, S. M. (1989). Cerastium, in WALTERS, S. M. et al., eds., The European Garden Flora 3, pp. 191-192. Cambridge University Press, Cambridge. (Accepted March 2001) ' — Watsonia 23: 493-503 (2001) 493 The vegetation associated with Spiranthes romanzoffiana Cham. (Orchidaceae), Irish Lady’s-tresses, on the Isle of Coll, Inner Hebrides S. A. HENDERSON Dept. of Botany, The Natural History Museum, Cromwell Road, London SW7 5BD ABSTRACT This paper details some environmental preferences of the scarce orchid Spiranthes romanzoffiana Cham. on the Isle of Coll in the Inner Hebrides. Samples were categorised into four site types that form a continuum rather than a series of discrete habitat types. High cover of Carex panicea L., and to a lesser extent Molinia caerulea (L.) Moench and Juncus articulatus L., occurs throughout. The habitats appear to be influenced by varying degrees of high water and low nutrient availability, as suggested by the disparate species. There is no apparent difference between former and present habitat of the orchid on the island, but few conclusions can be made from vegetation description alone. As it stands, the habitat as defined in this study is not uncommon, and it may be that other factors, such as seed sterility or availability of mycorrhizal associate, are limiting the species distribution. KEYWORDS: Habitat preference, vegetation analysis. INTRODUCTION Spiranthes romanzoffiana Cham. (Orchidaceae), Irish Lady’s-tresses, is one of the few British natives with an amphi-Atlantic distribution, found widespread in North America and limited to the Western fringes of Europe (Preston & Hill (1997) included it under their “boreal-montane element’). Indeed, within Europe it is almost entirely confined to the west of Scotland and south- western, western and Northern Ireland. Other species that share a similar “asymmetrical” amphi- Atlantic distribution include Potamogeton epihydrus Raft., Najas flexilis (Willd.) Rostkov. & W. Schmidt and Eriocaulon aquaticum (Hill) Druce. Although not discovered until 1810 in Ireland (Hackney 1992), it is widely accepted that S. romanzoffiana is native (e.g. Heslop-Harrison 1953; Love & Love 1958; Stace 1997). Theories explaining the amphi-Atlantic distribution include migration across a former land connection between Europe and North America (Dahl 1963; Hulten 1963) followed by a contraction in the European distribution during glaciation with survival in or near the present localities (see Coxon & Waldren 1995), or long distance dispersal (Heslop-Harrison 1953; Perring 1965). One of the outstanding features of North American elements in the British flora is that they are all aquatic, marsh or lake margin plants, suggesting a similarity in propagule dispersal. While controversy persists over the explanation for the uneven distribution of S. romanzoffiana, its similarity in geography to other species and the difficulty in accounting for seed transport to such relatively remote and more or less natural habitats by human means, are justifications of its native status (see Webb 1985). The main populations of $. romanzoffiana occur round the Galway-Mayo lakes area and Lough Neagh in Ireland, and on Colonsay, Coll and the Outer Hebrides of Scotland, and it has one outlying post in South Devon, discovered in 1957 (Anon. 1958). Stewart, Pearman & Preston (1994) categorise the species as Scarce, found in only eighteen 10-km squares in Britain. As the only widely recognised orchid species of Britain not to be found elsewhere in Europe, and with a stronghold in Scotland, Scottish Natural Heritage has designated S$. romanzoffiana a priority species for conservation. None of the Scottish localities recorded before 1981 now show the orchid, although several new sites have since been discovered (UK Biodiversity Group 1999). While little is known of the natural fluctuations of the populations, these results have increased concern over the orchid’s status in the British Isles. 494 S. A. HENDERSON S. romanzoffiana coexists with an agricultural system based on extensive grazing. The orchid is supposedly sensitive to subtle changes in the type, timing and intensity of grazing (e.g. Summerhayes 1951; Ferreira 1978; Horsman 1994). Traditionally, pastoral agriculture in the Hebrides entails cattle grazing from September to May followed by a short period allowing for plant growth and a subsequent harvest for hay. The land management is conducive to both survival and fecundity of S. romanzoffiana, the two factors that determine population status. Cattle grazing enables survival by keeping more competitive tussock-forming species low while the period free from grazing coincides with the orchid’s flowering period, thus ensuring fecundity. In Coll, a general trend in agricultural intensification is confounded by the heavy conservation priority placed on the corncrake, Crex crex, a bird which has now disappeared from most other parts of Britain (Stroud 1992). By encouraging “‘corncrake friendly” extensification of grazing on some nesting areas, the RSPB has forced intensification in other areas. Ironically, the conservation measure may pose a threat to populations of S$. romanzoffiana. In North America typical habitats are moist: in bogs, marshes, meadows, salt flats, muskegs, thickets, on sandy-gravelly beaches, but also occasionally on dry woods and dry open hillsides (Luer 1975; Correll 1978; Case 1987; Smreciu & Currah 1989; Homoya 1993). Typically a northern species, in the more southern parts of its range in America it is a mountain plant where it occurs up to 3000 m asl and flowers until October (Correll 1978). In the British Isles, however, it occurs within a narrower range of habitats, suggesting that here the populations are ecologically as well as geographically marginal. In Ireland it grows in damp meadows, close to rivers and lake-shores and in cut-over bogs (Praeger 1934; Harron 1986; Hackney 1992). In Britain, the orchid is similarly described as occurring on low-lying sites that are at least periodically irrigated and sometimes inundated (Scottish Natural Heritage 1995). Horsman (1994) described Spiranthes romanzoffiana as having a distinct habitat of Molinia caerulea carpet on old cattle-grazed lazy-beds. The habitats of S. romanzoffiana have been surveyed on Coll, Colonsay (Scottish Natural Heritage 1995; Gulliver 1996) and Loch Sheil (Martin & Milnes 1993) using the National Vegetation Classification (NVC) system (Rodwell 1991). All fall into the categories for mire, wet heath and rush pasture (Table 1). Collectively, and on a national scale, these communities are not infrequent. However, as NVC categories are based on the means of nation-wide data, they fail to describe the particular idiosyncrasies of an actual plant community (see Legg 1992). Multivariate ordination and classification techniques are preferable for the level of detail required in the study of rarities. For example Prober & Austin (1991) used the ordination technique of correspondence analysis to compare floristically sites where a rare species occurred with similar sites where it does not. Bowles et al. (1993), Ratcliffe, Birks & Birks (1993), Vazquez & Norman (1995) and Read & © Tweedie (1996) all used numerical classification techniques to characterise the habitats of rare species. In order to investigate the floristic component of S. romanzoffiana’s habitat, a multivariate analytical approach was used, with data from samples taken at two spatial scales. The two-scale TABLE 1. NATIONAL VEGETATION CLASSIFICATION COMMUNITIES AND SUB- COMMUNITIES IN WHICH SPIRANTHES ROMANZOFFIANA HAS BEEN FOUND Code NVC Community and sub-community (Rodwell 1991) M25 Molinia caerulea-Potentilla erecta mire M25 a Erica tetralix sub-community M25 b Anthoxanthum odoratum sub-community M23 Juncus effusus-Galium palustre rush pasture M23 a Juncus acutiflorus sub-community M23 b Juncus effusus sub-community Ml5a Scirpus caespitosus-Erica tetralix wet heath Carex panicea sub-community M10 Carex dioica-Pinguicula vulgaris mire M6 _ Carex echinata-Sphagnum recurvum/auriculatum mire Mé6d Juncus acutiflorus sub-community | ! | | | SPIRANTHES ROMANZOFFIANA ON COLL 495 approach enables a distinction to be made between the habitat preferences at the scale of the individual plant and those of the typical mire community; habitat description is scale-dependent due to spatial heterogeneity. Further, the identification of associated single species at a small scale may facilitate fine-grained ecological insights into the nature of the rare species’ requirements. The study was carried out on the Isle of Coll, in the Inner Hebrides, where the majority of the Scottish S$. romanzoffiana populations are found, and where there is a relatively long history of recorded individuals (Ferreira 1978; Horsman 1990, unpublished; Scottish Natural Heritage 1995). METHODS FIELD SURVEY All 107 sites (numbering 16 locations, although these do not necessarily form discrete populations) with accurate records of the orchid were visited, including two new sites found in 1996. Grid references, sketch maps (Ferreira 1978) and photographs (Horsman 1990, unpublished; Scottish Natural Heritage 1995) were used to relocate individuals or sites with previously recorded individuals which were, in 1996, apparently absent. The locations were also thoroughly searched for any unrecorded individuals. 1 x 1 m quadrats (““meso-scale samples’) were placed around the orchid(s), including only visually homogenous areas of vegetation. The equivalent area was taken where the homogenous vegetation did not fit to a 1 m*. Similarly, for those samples not containing S. romanzoffiana, the quadrat was placed around the site of the absent orchid (this was made possible by having relatively detailed and accurate records). Within each sample the number of individuals of S$. romanzoffiana and the abundance of all species were recorded using the Domin scale. Percentage bare ground was also estimated. 10 x 10 cm quadrats (“micro-scale samples’’) were then placed around each individual of S. romanzoffiana found and presence/absence of all species was recorded. This area was thought small enough to investigate the more specific requirements of the individual orchid but large enough to incorporate other associated species. Plant nomenclature follows Stace (1997) for vascular plants and Watson (1981) for bryophytes. ANALYSIS Ordination was used to summarise the sample data collected at both scales. Analysis of data used detrended correspondence analysis (DCA) (Hill 1979a; Hill & Gauch 1980), a modification of correspondence analysis (ter Braak 1987). Two-way indicator species analysis (TWINSPAN) (Hill 1979b; Gauch & Whittaker 1981) was used to classify the meso-scale samples. The computer packages CANOCO (ter Braak 1988) and CANODRAW (ter Braak 1992) were used for DCA analysis and ordination plots and VESPAN (Malloch 1995) for TWINSPAN analysis, raw data tables and descriptive statistics. RESULTS MESO-SCALE SITE TYPES Spiranthes romanzoffiana was present in 54 of the 107 meso-scale sites, (distributed over 11 of the 16 locations) and was absent in 53 of the sites (in 8 locations). In total 93 species were found. Carex panicea and Molinia caerulea were constant occurring in 94% and 86% of samples respectively, and 11 other species were frequent, occurring in 50% or more of samples: Juncus articulatus, Carex nigra, Hydrocotyle vulgaris, Nardus stricta, Ranunculus flammula, Holcus lanatus, Anagallis tenella, Succisa pratensis, Anthoxanthum odoratum, Carex echinata and Eriophorum angustifolium. The indicator species at divisions | and 2 of TWINSPAN are shown in Fig. 1. The first division distinguishes samples containing Erica tetralix from samples with Holcus lanatus, Galium palustre, Ranunculus flammula and Ranunculus acris. The characteristics of each of the four final groups of division 2 are summarised in Table 2. The sample ordination by DCA with the TWINSPAN groups superimposed (Fig. 2) shows that 496 S. A. HENDERSON TABLE 2. MOST FREQUENT SPECIES AT DIVISION TWO OF TWINSPAN. PERCENTAGE FREQUENCY IN 1M’? SAMPLES IS DENOTED IN PARENTHESES Species present in 80% or more samples Species present in 50%-80% samples Site type Al Spiranthes romanzoffiana (70) Carex nigra (100) Equisetum fluviatile (75) Hydrocotyle vulgaris (100) Carex echinata (70) Ranunculus flammula (100) Filipendula ulmaria (70) Carex panicea (95) Holcus lanatus (70) Juncus articulatus (95) Potentilla palustris (70) Galium palustre (90) Mentha aquatica (65) Caltha palustris (80) Calliergon cuspidatum (65) Eriophorum angustifolium (80) Leontodon autumnalis (55) Myosotis laxa (55) Senecio aquatica (55) Anthoxanthum odoratum (50) Pedicularis palustre (50) Site type A2 Spiranthes romanzoffiana (49) Carex panicea (97) Hydrocotyle vulgaris (78) Molinia caerulea (95) Ranunculus flammula (78) Juncus articulatus (92) Carex nigra (76) Holcus lanatus (89) Ranunculus acris (76) Nardus stricta (81) Anthoxanthum odoratum (73) Succisa pratensis (70) Trifolium repens (62) Potentilla erecta (58) Agrostis canina (51) Carex echinata (51) Site type B1 Spiranthes romanzoffiana (50) Carex panicea (100) Nardus stricta (77) Molinia caerulea (95) Anagallis tenella (75) Juncus articulatus (66) Carex nigra (61) Potentilla erecta (57) Succisa pratensis (57) Agrostis canina (55) Hydrocotyle vulgaris (55) Carex echinata (53) Eriophorum angustifolium (53) (note, Erica tetralix 39) Site type B2 Spiranthes romanzoffiana (11) Erica tetralix (100) Anagallis tenella (67) Molinia caerulea (100) Carex panicea (67) Carex hostiana (89) Selaginella selaginoides (56) Schoenus nigricans (89) Scorpidium scorpioides (56) | (note Drosera rotundifolia, Myrica gale and if Narthecium ossifragum at 44) | Those species characteristically present in high abundance are shown in bold. While subjective. this aids interpretation. SPIRANTHES ROMANZOFFIANA ON COLL 497 Divison | A (357) B (50) Holcus lanatus, Galium palustre, Erica tetralix Ranunculus acris, Ranunculus flammula Divison 2 Al (20) A2 (37) B1 (41) B2 (9) Caltha palustris, Nardus stricta without Schoenus nigricans Schoenus nigricans Equisetum fluviatale, Potentilla palustris FIGURE 1. Indicator species of the TWINSPAN hierarchy for the first two divisions. The number of samples in each category is shown in parentheses. 3 Axis 2 Axis 1 FIGURE 2. Sample ordination plot for the first two axes of Detrended Correspondence Analysis (DCA) of meso-scale habitat data for Spiranthes romanzoffiana. TWINSPAN-defined groups are superimposed. the TWINSPAN-defined site types are not readily separated, either at the first or the second divisions. The first four axes extracted in DCA explain only 22.8% of the total variation in species data, and the first two axes of the ordination (which are used in the ordination plot) explain only 17% of floristic variation between samples. There is thus a large proportion of variation that is not explained by the ordination plot. The sample ordination plot shown with S. romanzoffiana presence/absence data superimposed (Fig. 3), shows for each group an ellipse within which, on average, 68% of the individuals will lie. Considerable overlap is shown which indicates little pattern of differentiation between current and former sites. The full set of samples were consequently analysed as potentially suitable. MICRO-SCALE SITE TYPES A total of 81 micro-scale samples were taken, all of which contained at least one individual of S. romanzoffiana. A total of 68 species were recorded and the mean per sample was 9 (standard error 0-29). Sample ordination by DCA showed that species composition reflects that of the larger scale though inevitably fewer species are present. The proportion of samples within which the 13 most constant species were present is shown in Table 3. Carex panicea was the most associated 498 S. A. HENDERSON TABLE 3. THE 13 MOST ASSOCIATED SPECIES WITH SPIRANTHES ROMANZOFFIANA AT THE MICRO-SCALE, AS MEASURED BY THE PERCENTAGE OF THE 10 CM? SAMPLES WITHIN WHICH THEY WERE PRESENT Species % constancy [Spiranthes romanzoffiana 100] Carex panicea 84 Molinia caerulea 63 Carex nigra 44 Hydrocotyle vulgaris 43 Anagallis tenella 38 Nardus stricta 6s) Eriophorum angustifolium Zi Ranunculus flammula 2] Carex echinata 26 Trifolium repens PG) Holcus lanatus 21 Carex hostiana 20 Juncus articulatus 20 species at the micro-scale occurring in 84% of the samples, Molinia caerulea occurred in only 63% of the samples. No other species was present in more than 50% of samples. Juncus articulatus, a species occurring with S. romanzoffiana in 76% of the meso-scale samples, is present in only 20% of the samples at the micro-scale. DISCUSSION SAMPLING CRITIQUE The accessibility of all the sites relocated opens to question the extent to which the orchid occurs in relatively inaccessible areas. Known sites tend to occur close to roads - perhaps a sampling artifact (as it is assumed that not all individuals in the population will have been recorded, perhaps the unrecorded individuals grow in less accessible areas), indeed “road map recording” is a well known phenomenon. Alternatively, the distribution may reflect true habitat preference for the marginal land between dune and moor, generally known in the Western Isles as the “blackland” (Hambrey 1986). The transitional zone between lime-rich and acid soils is not only important for many species of plant but also tends to provide the most suitable terrain for roads. Sampling was biased towards flowering spikes; vegetative individuals were normally only found when searched for thoroughly in a small site where an individual was known to occur, or in short open turf. Bias was additionally encountered by the relative ease of spotting plants (both vegetative and flowering) in short vegetation. Sites within taller vegetation may thus be under- represented. It is difficult to determine whether an individual is truly absent from a former site or whether it has just been missed. The difficulty in distinguishing absences which are a result of poor sampling from those which are genuine is a serious problem in the study of rarity in general (McArdle & Gaston 1993: Gaston 1994) and accentuated here by the potential for S. romanzoffiana to lie dormant underground (Gulliver et al. 2000). Finally, there is sample bias towards those locations that have a longer history of recorded individuals; at the two new sites, absences cannot be recorded for the obvious reason that there is no comparative data from previous years. THE HABITATS OF SPIRANTHES ROMANZOFFIANA Horsman (1994) described Spiranthes romanzoffiana in Scotland as having a distinct habitat of Molinia caerulea.carpet on old lazy beds grazed by cattle. M. caerulea is also listed as the closest associate in 17 Scottish sites surveyed in 1995 (Scottish Natural Heritage 1995). The present SPIRANTHES ROMANZOFFIANA ON COLL 499 3 : @ Present : © Absent p N & x< = < eee? 1 @ ®@ O 8 0 1 2 3 4 Axis 1 FIGURE 3. Ordination plot for all meso-scale samples for the first two axes of Detrended Correspondence Analysis (DCA), with presence or absence of Spiranthes romanzoffiana in each sample superimposed. 68% confidence ellipses are shown. author, in agreement with Gulliver’s study on Colonsay (1996), found only one site on an old lazy bed. Further, despite it being in general abundant, and the second most associated species with the orchid at the micro-scale, it was found that in a number of samples, typically those of Site Type Al (the habitat in which S. romanzoffiana was most frequent (Table 2)), M. caerulea was not abundant or even present. All the NVC habitats previously described for the orchid include M. caerulea as a typical species at intermediate to high frequency and abundance. However, to compare the results directly to equivalent NVC categories would be inappropriate without considering the effects of scale and pattern within the vegetation. NVC demands larger samples (2 x 2m or over) and treats minute mosaics as a single vegetation type (Rodwell 1991). The resulting NVC classification is therefore at a coarser grain than that presented here, which describes the floristics immediately surrounding the orchid. It would be similarly misleading to convert the present data into NVC classes. All site types have an abundance and constancy of the generally low-growing Carex panicea and an abundance and frequency of Hydrocotyle vulgaris, Ranunculus flammula and/or Anagallis tenella. The species are characteristic of unshaded, soligenous mire on peaty, mildly acidic soils where growth of potential dominants is suppressed by low fertility and grazing pressure (Grime, Hodgson & Hunt 1990). While the soil characteristics were not analysed, grazing was evident in all sites. In these aspects, the habitats are similar. However, TWINSPAN and ordination by DCA indicate differential abundances of the dominants and distributions of more minor species between the four species groups. The habitats are distinguished by these lesser species which have quite different ecological preferences and may reflect differences in water and nutrient levels. Site-type Al contains a species component tolerant of a degree of waterlogging and typical of neutral/mildly acid conditions (Grime, Hodgson & Hunt 1990): Caltha palustris, Potentilla palustris and Equisetum fluviatile define the habitat and are largely restricted to it. Other wetland species present in these sites and less abundant or absent in others include Mentha aquatica, Myosotis laxa, Filipendula ulmaria, Senecio aquaticus and Iris pseudacorus. Carex nigra, present and abundant in all samples within this site type, has a wide habitat tolerance but is usually found in mires with some degree of water movement or mineral enrichment (Jermy, Chater & David 1982). Hydrocotyle vulgaris is constant, and is a wetland species suggested to have an inability to exploit sites which are waterlogged all year (Grime, Hodgson & Hunt 1990), while Ranunculus flammula, also a constant, is adapted to both submergence and desiccation (Cook & Johnson 500 S. A. HENDERSON 1968). The evidence suggests that Al is composed of sites that are periodically flooded. Indeed, samples of this type are primarily found around a periodically flooded pasture and a flat, poorly drained area beside a lochan. Al grades into site-type A2, a drier, slightly more acidic habitat, as indicated by the significant contribution to cover of some graminoids less tolerant of inundation such as Nardus stricta, Molinia caerulea and Holcus lanatus, but one which still contains an element of species of Al in the wetter samples, such as Galium palustre, Senecio aquaticus and Filipendula ulmaria. Nardus stricta, the TWINSPAN indicator species which distinguishes the habitat from A1, is a grass that is characteristic of free-draining, acid soils (Grime, Hodgson & Hunt 1990). Species found almost exclusively in this habitat, such as Cynosurus cristatus, a species susceptible to both drought and waterlogging (Grime, Hodgson & Hunt 1990) and Achillea ptarmica, again suggest a habitat that is moist but not waterlogged. Site type A2 thus appears to be of a wet, slightly acid graminoid and herb-rich turf. Site type A2 (and to a lesser extent Al) grades into site type B1. Species that are common to both habitats A2 and B1 but not the others include frequent species such as Succisa pratensis, Potentilla erecta, Carex echinata as well as less abundant species such as Plantago maritima, P. lanceolata, Carex dioica and Euphrasia scottica. B1, however, contains elements typical of wet acid heath (Grime, Hodgson & Hunt 1990; Rodwell 1991; Stace 1997), primarily shown by the presence of the indicator species at division one, Erica tetralix, but also the other woody shrubs Calluna vulgaris and, occasionally, Salix repens and Myrica gale. The low cover and presence of these species in the samples indicates a habitat where their dominance is restricted to some extent. Further, the abundance of these subshrubs is sufficiently low to allow a relatively high cover and constancy of the creeping species Anagallis tenella, and the low-growing Drosera spp., typical of unshaded and wet, nitrogen-poor sites of heaths and bogs (Stace 1997). The wet, acidic conditions of the site type are further suggested by the presence of Sphagnum auriculatum and, more frequently, S. recurvum. Eleocharis quinqueflora, characteristic of damp, peaty places and wet flushes in the west of Scotland (Phillips 1994), also appears in this habitat. It seems therefore that these habitats may comprise flush vegetation within more enclosed communities where competition from low-growing shrubs would otherwise suppress the orchid. The constancy of Erica tetralix and presence of species such as Myrica gale, Pinguicula vulgaris, Selaginella selaginoides, Drosera spp., and Narthecium ossifragum render site type B2 similar to B1; the habitat still appears to be a flush in wet, nutrient-poor heath. However, B2 is distinguished from B1 by the presence of the tuft-forming Schoenus nigricans, occurring in all samples but one, and also the bryophyte Scorpidium scorpioides. The combined presence of these species suggests a habitat of wet flushes where the water content is fairly base-rich (see Phillips 1994; Sparling 1968; Watson 1981). Such base and cation enrichment may result from flushed shell-sand or directly from the bedrock (Rodwell 1991). Carex panicea is the species most associated with S. romanzoffiana at both the large and small scale. C. panicea is characteristically a wetland species, relatively low-growing, with a pH range from 4-7-5 (Grime, Hodgson & Hunt 1991); in vitro seed germination experiments have revealed that S. romanzoffiana prefers a soil pH from 4-7—5 (Arditti 1992). C. panicea is widespread throughout the British Isles, occurring in all British vice-counties, and has a capacity to exploit such diverse habitats as soligenous mire and limestone grassland (Grime, Hodgson & Hunt 1991) as well as dwarf-shrub heaths (Jermy, Chater & David 1982). The occurrence of C. panicea in a wide range of vegetation types outside those described here make it an unsuitable indicator species for potential S. romanzoffiana habitat. Associated species are therefore not necessarily good indicator species. However, the results suggest that, on Coll at least, habitats without C. panicea are unlikely to support S. romanzoffiana. The low association of Juncus articulatus with S. romanzoffiana at the micro-scale, despite the presence of the rush in 76% of the samples at the meso-scale, may reflect an ecological preference of the orchid. Grime, Hodgson & Hunt (1990) claim that J. articulatus is morphologically variable (ranging in height from 20 mm to 600 mm), often depending on the level of grazing. It is possible that S. romanzoffiana cannot grow in the shade that the tall-growing rush provides, but can survive when J. articulatus is relatively short. Alternatively, the tufted nature of the rush may make survival and growth of the orchid amongst its stems difficult. SPIRANTHES ROMANZOFFIANA ON COLL 501 CONCLUSIONS Overall vegetation description alone cannot provide a comprehensive summary of all habitat variables. The structure of vegetation may be one of the most important conditions determining habitat suitability and may be altered in a relatively short time by grazing. A period of inertia before species composition responds would then render floristic measurements inadequate. This may explain the apparent similarity in the current study of former and present habitats of S. romanzoffiana. Vegetation description also fails to account for environmental variables that affect only one species, in particular the requirement for orchid mycorrhizal infection before germination (Wells 1981). It would therefore be interesting to examine the presence of the specific symbiotic fungus in the soil in suitable and potentially suitable habitats for S$. romanzoffiana. There are other reasons why an individual may appear absent from a potentially suitable site. Dormant individuals will be recorded as absent unless destructive methods are used (hardly an option for a scheduled species!). The related orchid S. spiralis can lie dormant for at least a year before flowering (Wells 1967) and S. romanzoffiana has been documented to subsist for five years underground before producing above-ground growth (Gulliver et al. 2000). The role of stochastic events is also important; the processes of population extinction and colonisation may cause suitable sites to become unoccupied, at least temporarily. For this reason, all habitats in which S. romanzoffiana was known to occur, however infrequently, were included in the range of vegetation types designated suitable. These results provide a more accurate depiction of the immediate habitat of S$. romanzoffiana than do the NVC descriptions (Rodwell 1991). They demonstrate the orchid’s ecological affinity to C. panicea and avoidance of tussock-forming species at the micro-scale. They also show that the orchid does not apparently have an affinity to M. caerulea, at least not at the spatial scale examined here. However, the plant associations recognised in this study are not uncommon. Gaston (1994) recognised two broad ranges of factors causing limited ranges and/or low abundances: environmental limitations and colonisation abilities. The results here suggest that factors other than the environment may be limiting the distribution of $. romanzoffiana. There is evidence that the species is reproductively limited where it is found in the British Isles; Heslop- Harrison (1953) found that, while it was an active colonist of cut-away bogs around Lough Neagh, there were few years when it set seed. While flowering spikes are produced, it appears that Scottish populations may produce fertile seed rarely (Gulliver 1997; Gulliver et al. 2000) and may even reproduce entirely vegetatively (Horsman 1994; UK Biodiversity Group 1999). Whether or not this accounts for the limited distribution of S$. romanzoffiana in Britain is not known; further studies into the dispersal and colonisation abilities of S. romanzoffiana are therefore required. ACKNOWLEDGMENTS I am indebted to Neil Cowie at the R.S.P.B. for his suggestions and support, and to the farmers on the Isle of Coll for their co-operation. Thanks also to Fred Rumsey, Bob Press and Richard Bateman for valuable comments on earlier drafts of the manuscript, and to Nancy Garwood for help with the graphics. The work was provided with financial support from Scottish Natural Heritage. REFERENCES ANON. (1958). Spiranthes romanzoffiana in South Devon. Proceedings of the Botanical Society of the British Isles 3: 37-38. ARDITTI, J. (1992). Fundamentals of orchid biology. John Wiley, Chichester. BOWLES, M. L., FLAKNE, R., MCEACHERN, A. K., & PAVLOVIC, N. B. (1993). Recovery planning and reintroduction of the Federally threatened Pitcher’s thistles (Cirsium pitcheri) in Illinois. Natural Areas Journal 13: 164-170. CASE, F. W. (1987). Orchids of the western Great Lakes region. Edward Brothers, Inc. Cook, S. A. & JOHNSON, M. P. (1968). Adaptation to heterogeneous environments. 1: Variation in heterophylly in Ranunculus flammula L. Evolution 22: 495-516. 502 S. A. HENDERSON CORRELL, D. S. (1978). Native orchids of North America north of Mexico. 2nd edition. Waltham, Massachusetts. COXON, P. & WALDREN, S. (1995). The floristic record of Ireland’s Pleistocene temperate stages, in PREECE R. C., ed., Island Britain: a Quarternary perspective pp. 243-267. Geological Society Special Publication 96. Geological Society, England. DAHL, E. (1963). Plant migrations across the North Atlantic ocean and their importance for the paleogeography of the region, in LOVE, A. & LOVE, D., eds., North Atlantic biota and their history pp. 173-188. Pergamon Press. FERREIRA, R. E. C. (1978). Spiranthes romanzoffiana sites on Coll. Nature Conservancy Council. GASTON, K. J. (1994). Rarity. Chapman & Hall, London. GAUCH, H. G. & WHITTAKER, R. H. (1981). Hierarchical classification of community data. Journal of ecology 69: 537-557. GULLIVER, R. L. (1996). The status of Spiranthes romanzoffiana Cham. (Orchidaceae) Irish Lady’ s-tresses, on Colonsay, (v.c. 102) in 1995; with special reference to associated plant communities. Watsonia 21: 202- 204. GULLIVER, R. L. (1997). Irish Lady’s Tresses (Spiranthes romanzoffiana) on Colonsay (v.c. 102) Glasgow Naturalist 23(2): 55—56. GULLIVER, R. L., KEIRNEN, M., GULLIVER, M. & SYDES, C. (2000). Observations on Irish Lady’s Tresses orchid (Spiranthes romanzoffiana) on Colonsay (v.c. 102) Glasgow Naturalist 23(5): 9-12. GRIME, J. P., HODGSON, J. G. & HUNT, R. (1990). The abridged comparative plant ecology. Chapman & Hall, London. HACKNEY, P. (1992). Stewart and Corry’s Flora of the North-East of Ireland. 3rd ed. Queens University, Belfast. HAMBREY, J. (1986). Agriculture and environment in the Western Isles. Nature Conservancy Council, Peterborough. HARRON, J. (1986). Flora of Lough Neagh. University of Ulster, Northern Ireland. HESLOP-HARRISON, J. (1953). The North American and Lusitanian elements in the flora of the British Isles, in LOUSLEY, J. E., ed., The changing flora in Britain pp. 105-123. Macmillan, London HILL, M. O. (1979a). DECORANA - a FORTRAN Program for Detrended Correspondence Analysis and Reciprocal Averaging. Cornell University, Department of Ecology and Systematics, Ithaca, New York. HILL, M. O. (1979b). TWINSPAN - a FORTRAN Program for arranging multivariate data in an ordered two-way table by classification of the individuals and the attributes. Cornell University, Department of Ecology and Systematics, Ithaca, New York. HILL, M. O. & GAUCH, H. G. (1980). Detrended correspondence analysis, an improved ordination technique. Vegetatio 42: 47-48. Homoya, M. A. (1993). Orchids of Indiana. Indiana University Press. HORSMAN, F. C. (1994). Spiranthes romanzoffiana in STEWART, A., PEARMAN, D. A. & PRESTON, C. D., eds., Scarce plants in Britain p. 398. Joint Nature Conservancy Council, Peterborough. HULTEN, E. (1963). Phytogeographical connections of the North Atlantic, in LOVE, A. & LOVE, D., eds., North Atlantic biota and their history pp. 45-72. Pergamon Press. JERMY, A. C., CHATER, A. O. & DAVID, R. W. (1982). Sedges of the British Isles. Botanical Society of the British Isles, London. LEGG, C. J. (1992). Putting concrete vegetation into abstract boxes. Bulletin of the British Ecological Society 23: 28-30. Love, A. & LOvE, D. (1958). The American elements in the flora of the British Isles. Botaniska Notiser 3: 376-388. LUER, C. A. (1975). The native orchids of the United States and Canada. New York Botanical Gardens, New York. MALLocH, A. J. C. (1995). Vespan III. Routines for vegetation analysis and species distribution. Unit of Vegetation Science, University of Lancaster. MARTIN, D. & MILNES, K. (1993). An investigation into scarce plants in the locality of the river and Loch Shiel. Scottish Natural Heritage report, Fort William. MCARDLE, B. H. & GASTON, K. J. (1993). The temporal variability of populations. Oikos 67: 187-191. PERRING, F. H. (1965). Advance and retreat of the British flora, in JOHNSON, G. C. & SMITH, L. P., eds., The biological significance of climate changes in Britain pp. 51-59. Institute of Biology Symposia 14. Academic Press, London. PHILLIPS, R. (1994). Grasses, ferns, mosses and lichens of Great Britain and Ireland. 2nd ed., Macmillan, London. PRAEGER,R. L. (1934). The botanist in Ireland. Hodges, Figgis & Co., Dublin. PRESTON, C. D. & HILL, M. O. (1997). The geographical relationships of British and Irish vascular plants. Botanical journal of the Linnean Society 124: 1-120. PROBER, S. M. & AUSTIN, M. P. (1990). Habitat peculiarity as a cause of rarity in Eucalyptus paliformis. Australian journal of ecology 16: 189-205. SPIRANTHES ROMANZOFFIANA ON COLL 503 RATCLIFFE, D. A. BIRKS, H. J. B. & BIRKS, H. H. (1993). The ecology and conservation of the Killarney fern Trichomanes speciosum Willd in Britain and Ireland. Biological conservation. 66: 231—247. READ, D. G. & TWEEDIE, T. D. (1996). Floristics of habitats of Pseudomys oralis (Rodentia, Muridae). Wildlife research 23: 485-493. RODWELL, J. S. (1991). British plant communities volume 2. Mires and heaths. Cambridge University Press, Cambridge. SCOTTISH NATURAL HERITAGE (1995). Species dossier: Spiranthes romanzoffiana. SMRECIU, E. A. & CURRAH, R. S. (1989). A guide to the native orchids of Alberta. University of Alberta, Canada. SPARLING, J. H. (1968). Biological Flora of the British Isles: Schoenus nigricans L. Journal of ecology 56: 883-899. STACE, C. A. (1997). New Flora of the British Isles. Cambridge University Press, Cambridge. STEWART, A., PEARMAN, D. A. & PRESTON, C. D., eds., (1994). Scarce plants in Britain. Joint Nature Conservancy Council, Peterborough. STROUD, D. A. (1992). Crofting and bird conservation on Coll and Tiree. British wildlife 3: 340-349. SUMMERHAYES, V. S. (1951). Wild orchids of Britain. Collins, London TER BRAAK, C. J. F. (1987). Ch. 5. Ordination, in JONGMAN, R. H. G., TER BRAAK, C. J. F. & VAN TONGEREN, O. F. R., eds., Data analysis in community and landscape ecology pp. 91-173. Pudoc, Wagineningen. TER BRAAK, C. J. F. (1988). CANOCO - a FORTRAN program for canonical community ordination by (partial) (detrended) (canonical) correspondence analysis. TNO Institute of Applied Computer Science, Wageningen. TER BRAAK, C. J. F. (1992). CANODRAW. TNO Institute of Applied Computer Science, Wageningen. UK BIODIVERSITY GROUP (1999). Tranche 2 Action Plans Vol. III: Plants and Fungi. English Nature, Northminster House. VAZQUEZ, J. A. & NORMAN, G. (1995). Identification of site-types important for rare ferns in an area of deciduous woodland in northwest Spain. Vegetatio 116: 133-146. WATSON, E. V. (1981). British Mosses and Liverworts. 3rd ed. Cambridge University Press, Cambridge. WEBB, D. A. (1985). What are the criteria for presuming native status? Watsonia 15: 231-236. WELLS, T. C. E. (1967). Changes in a population of Spiranthes spiralis (L.) Chevall. at Knocking Hoe National Nature Reserve, Bedfordshire, 1962-1965. Journal of ecology 55: 83-89. WELLS, T. C. E. (1981). Population ecology of terrestrial orchids, in SYNGE, H., ed., The biological aspects of rare plant conservation pp. 281—295. John Wiley, Chichester. (Accepted November 2000) wee. fa) fi a: a i yrA Al See oe a: es #9) “vate § Lie st b na Watsonia 23: 505—516 (2001) 505 A re-assessment of the putative Carex flava agg. (Cyperaceae) hybrids at Malham Tarn (v.c. 64): A morphometric analysis N. BLACKSTOCK and P. A. ASHTON Dept. of Natural and Applied Sciences, Edge Hill College of Higher Education, St. Helens Rd, Ormskirk, Lancashire L39 4QP ABSTRACT Carex flava L., a rare species in the British Isles, is currently only recognised from one extant population. The presence of three putative C. flava x C. viridula Michx. subsp. brachyrrhyncha (Celak.) B. Schmid hybrid (= C. x alsatica Zahn.) populations suggests that it may have formerly been more widespread. Alternative taxonomic treatments have been considered populations, including the possibility that they could be C. flava subsp. jemtlandica (Palmgr.) P. D. Sell, a Fennoscandian taxon not previously recorded in this country. Uni- and multivariate statistical analyses were used to elucidate the taxonomic affinity of the largest of these populations. A new combination C. flava subsp. jemtlandica to C. virtdula Michx. subsp. brachyrrhyncha (Celak.) B. Schmid var. jemtlandica (Palmgr.) Blackstock & P. A. Ashton, is made to bring the nomenclature of this taxon in line with the current nomenclature of this group. The presence of a second extant population of C. flava in the British Isles is also reported. KEYWORDS: Carex viridula subsp. brachyrrhyncha var. jemtlandica, Principal Components analysis (PCA), ANOVA, Discriminant Analysis (DA). INTRODUCTION In the British Isles Carex flava L. (Large Yellow Sedge) is currently only recognised from one extant population and one historic record. In addition, populations generally considered to consist of C. flava x C. viridula Michx subsp. brachyrrhyncha (Celak.) B. Schmid hybrids (= C. x alsatica Zahn) have been recorded from three sites; the Tarn Moss, Malham (v.c. 64) (Shaw 1946), Greywell Moors, Hants (v.c. 11) (Brewis et al. 1996) and by the River Corrib near Menlough, north-east of Galway City (v.c. H17) (Perring 1970). C. flava is considered to have formerly been extant at these localities but is now believed to be extinct, probably through hybridisation with C. viridula. subsp. brachyrrhyncha (Jermy et al. 1982). C. flava is categorised in The British Red Data Book: I. Vascular Plants (Taylor 1999) as vulnerable in Britain, but not threatened in Europe. Populations are also found in Western and Eastern North America at latitudes similar to those in Europe. Although early Floras covering the various regions of the British Isles are littered with accounts for C. flava s. I. (e.g. Goodenough 1792; Green 1933), the first authenticated specimen for C. flava s. s. was collected by J. Dickinson in 1836, somewhere in the Ennerdale area of Cumbria (v.c. 70). This population is now only known from one voucher specimen (LIV), although doubts as to the authenticity of the site locality have been raised (Halliday 1997). Evidence of a population previously unaccounted for in the British Floras comes from a herbarium specimen collected by T. W. Edmondson from Hebden, W. Yorks (v.c. 64) in 1906. This population is also only known from one voucher specimen (GH) which has been examined by W. J. Crins and has been determined by him to be C. flava s. s. (Crins 1985). The only currently recognised extant population of C. flava s. s. is at Roudsea Wood, Haverthwaite, Cumbria (v.c. 69) being discovered by D. Lumb in 1913 (B.E.C. Rpt vol Il pt VI p. 504). The Roudsea Wood population is well known and the number of plants there appears to be relatively stable at around 1500-2000 presumed genets (Blackstock, unpublished data). In 1946 G. A. Shaw reported C. flava from the Tarn Moss, Malham (v.c. 64). This population was also referred to as C. flava s. s. by Davies (1953). However, doubts as to the true identity of this population were raised by Jermy & Tutin (1968). They state that “the only typical material” of 506 N. BLACKSTOCK AND P. A. ASHTON C. flava in Britain “is from N. Lancs” i.e. Roudsea Wood. Jermy ef al. (1982) state categorically that “plants intermediate to C. flava and C. lepidocarpa Tausch. (= C. viridula subsp. brachyrrhyncha) ...occur at Malham Tarn, Lough Corrib and N. Hants, where C. flava presumably once occurred but is now extinct’. Stace (1997) reiterates this view by formally referring to these populations as C. x alsatica. A different approach to the putative C. x alsatica populations found at Malham Tarn, Lough Corrib and Greywell Moors has been taken by Sell & Murrell (1996) who classify them as C. flava subsp. jemtlandica (Palmgrn.) P. D. Sell (= C. jemtlandica Palmgrn.). This taxon is generally regarded as a Fennoscandian endemic (Hedrén & Prentice 1996; Pykalé & Toivonen 1994) and has not previously been recognised in the British Isles. Clearly the Malham Tarn, Lough Corrib and Greywell Moors populations of yellow sedges have been subject to a degree of taxonomic confusion. Davies (1953) considers them to be C. flava s. s.; Jermy et al. (1982) consider them to be C. x alsatica, a C. flava x C. viridula subsp. brachyrrhyncha hybrid; while Sell & Murrell (1996) treat them as a subspecies of C. flava comparable with their classification of other C. flava subspecies. Multivariate statistical analyses have been successfully used to clarify taxonomic problems in Carex complexes e.g. sects. Vesicariae (O. F. Lang) Christ. (Shepherd 1975), Phyllostachys (J. Carey) L. H. Bailey (Naczi et al. 1998) and Ceratocystis Dumort. (Crins 1985). Studies within sect. Ceratocystis have utilised multivariate statistical methods such as Detrended Correspondence Analysis (DCA) (Schmid 1980), Principal Components Analysis (PCA) (Crins 1985) and Discriminant Analysis (DA) (Crins 1985) based upon herbarium specimens. The use of two of these methods (PCA and DA) on living specimens of C. flava, C. viridula subsp. brachyrrhyncha, the putative C. x alsatica from Malham Tarn and C. jemtlandica has been adopted here to elucidate the taxonomic boundaries between these taxa. This study includes plants from North America, the British Isles and continental Europe and seeks to place the British and Irish taxa into a wider context, while clarifying the taxonomic status of the problematic Malham Tarn yellow sedges. MATERIALS AND METHODS Specimens were collected in the field, either as ramets or seed, from clumps situated at least 3 m apart from each other. Members of the C. flava agg. have a caespitose growth habit, so each sample was presumed to represent an individual genet. The individual specimens were cultivated under similar conditions at Edge Hill for 1-2 years until the plants had reached maturity. Specimens of C. flava from Canada provided a valuable comparison, as these populations are approximately 700 km away from the nearest C. viridula subsp. brachyrrhyncha population. Voucher specimens are currently held at Edge Hill College of Higher Education and it is intended that they will be deposited at LIV upon completion of the associated work. Details of the specimen localities are given in Table 1. Four a priori groups (Table 1) were identified for analysis: 1. C. flava, (40 specimens, eleven populations); 2. The putative C. x alsatica from Malham Tarn (MT) (ten specimens, one population); 3. C. viridula subsp. brachyrrhyncha (40 specimens, 13 populations) and 4. C. jemtlandica (ten specimens, four populations). All of the C. jemtlandica sites were included in a study of C. lepidocarpa s. 1. (Hedrén & Prentice 1996). The specimens collected from Dalbyn, near Ore, Dalarna, Sweden are from the same site as specimens collected and photographed by Palmgren for his monograph of the C. flava agg. (Palmgren 1959). As no single classification recognises all of these groups, they will be referred to as listed above until their status is re-assessed later in this paper. The putative C. x alsatica from Greywell Moors is only known from a single plant and this has not been seen in recent years. Therefore this population has not been included in our analyses. In addition, it has not been possible to obtain samples of the putative C. x alsatica from Ireland. Multivariate analyses provide a preliminary re-assessment of the a priori classification and of the phenetic relationships between the taxa. Eleven characters (Table 2) were included in the morphometric analysis. All morphometric data were derived from living plants that were producing a fertile culm with perigynia containing well-developed nutlets. This eliminated the possibility that Fl-hybrids could be included in the data set and ensured that only complete data CAREX FLAVA AGG. AT MALHAM TARN 507 TABLE 1. LOCALITIES OF THE POPULATIONS OF YELLOW SEDGES INCLUDED IN THE PCA. TAXON REFERS TO THE A PRIORI CLASSIFICATION OF THE GROUPS Number of Species Location Country Location individuals Taxon sampled Nr. Campbellville, Halton Co., Canada 43°31’N 79°59°W 2 C. flava Ontario Nr. Orangeville, Peel Co. OntarioCanada 43°53’N 80°06’ W 1 C. flava 8 km NW Tranas, Smaland Sweden 58°06’N 14°58’E D) C. flava Tannas, Harjedalen Sweden 62°40’N 12°41’E 2 C. flava Sjugelmyren, S6rboda, Dalarna Sweden 61°07’N 15°13’E 1 C. flava Lake Essen, Nr. Rattvik, Dalarna Sweden 60°57’N 15°13’E 4 C. flava Vastana, Boda, Dalarna Sweden 60°59’N 15°12’E i C. flava Romedenne Belgium 50°10’N 04°42’E 5 C. flava Roudsea Wood, Cumbria England 54°14’N 03°08’ W 6 C. flava Nittedal Hakadal Norway 60°05’N 10°50’°E 9 C. flava K4arnten Austria 47°N 13°E 1 C. flava Malham Tarn Fen, England 54°06’N 02°11’ W 10 MT Mid-west Yorkshire Stromyren, Boda, Dalarna Sweden 61°01’N 15°13’E 2 C. jemtlandica Vadeermyran, Hammerdal, Sweden 62°52’N 15°16’E B) C. jemtlandica Jamtland Ljusnedal, Harjedalen Sweden 62°48’N 12°36’E 2 C. jemtlandica Dalbyn,Ore, Dalarna Sweden 61°08’N 15°10’E 3 C. jemtlandica Benestads Bakan, Tomelilla, Sweden 99°57 N13" 53; EF 4 C. viridula subsp. brachyrrhyncha Skane Lyngsjon Lake, Skane Sweden 55°48’N 13°53’E 4 C. viridula subsp. brachyrrhyncha Hiiumaa Island Estonia » 59°10°N'22°30°E po C. viridula subsp. brachyrrhyncha Locheres France 47°41’N 04°5S’E 1 C. viridula subsp. brachyrrhyncha Crummack Farm, Inglebrough, England 54°08’N 02°21’W 1 C. viridula subsp. brachyrrhyncha N. Yorkshire Crummack Dale, Inglebrough, England 54°08’N 02°21’ W 4 C. viridula subsp. brachyrrhyncha N. Yorkshire Malham Tarn outflow, England 54°06’N 02°11’>W 6 C. viridula subsp. brachyrrhyncha Mid-west Yorkshire Woodbastwick, E. Norfolk England 52°41N 01°27’E 3 C. viridula subsp. brachyrrhyncha Greywell Moor, Hants. England 51°15’N 00°58’ W 1 C. viridula subsp. brachyrrhyncha Hawes Water, Lancashire England 54°11’N 02°48’ W 6 C. viridula subsp. brachyrrhyncha Fearnan, Loch Tay, Scotland 56°47’N 03°48’ W 1 C. viridula subsp. brachyrrhyncha Perth & Kinross Glen Fender, Perth & Kinross Scotland 56°34’N 04°05’ W 5 C. viridula subsp. brachyrrhyncha Prission, S-Tirol Austria 47°N 12°E 2 C. viridula subsp. brachyrrhyncha sets were included for subsequent analysis. Measurements were taken using Rabone vernier calipers graduated to 0-1 mm, with the exception of a coarse measurement (character VEGLEAF) which was measured to the nearest mm using a standard 30 cm line ruler. Previous morphometric studies, either within the C. flava agg. or Carex sub-genus Carex, have utilised a greater number of morphological characters (e.g. Shepherd 1975; Schmid 1980; Crins 1985; Hedrén 1998; Naczi et al. 1998). However, many of the characters included in the above studies are associated with the position of the pistillate spikelets. Such spikelets are not necessarily equivalent within the C. flava agg. Therefore, they and their associated characters have not been included in this study. Characters BEAKL, PERLE and BRISTLE are all associated with the perigynia. Crins (1985) and Rothrock et al. (1997) noted that there was greater variability in perigynia from the lower parts of the pistillate spikelets. Therefore perigynia from the mid-section of the pistillate spikelets were used for the measurements of all perigynia characters. All characters were then used in PCA to summarise variation patterns within all of the taxa studied. Data were standardised for PCA so that 508 N. BLACKSTOCK AND P. A. ASHTON TABLE 2. MORPHOLOGICAL CHARACTERS, WITH THEIR ABBREVIATIONS, MEASURED FROM LIVING PLANT SPECIMENS OF THE CAREX FLAVA AGG. ALL CHARACTERS WERE INCLUDED IN THE PCA AND ANOVA Character Abbreviation Vegetative leaf length/width ratio VEGLEAF Vegetative leaf width at base (mm) VEGBAS Ligule length (mm) LIGL Male spikelet length (mm) MSPIKE Male spikelet peduncle length (mm) MPEDL Beak length BEAKL Perigynum length (mm) PERLE Male glume length/width ratio MGLUM Female glume length/width ratio FGLUM No. bristles on one side of the perigynum BRISTLE Nutlet length excluding beak (mm) NUTL each variable would have a mean of 0 and a standard deviation of 1. PCA was carried out using the Multivariate Statistical Package version 3.0 from Kovach Computing Services (Kovach 1998). Data were further analysed using a stepwise DA (S.P.S.S. 1998). Univariate statistical analyses (mean, range and standard deviation) were performed on all characters using Microsoft Excel 97. Differences in the mean values for single characters between all the taxa were tested using one way analysis of variance (ANOVA) with the data first being tested for variance equivalence using the F-test. RESULTS Scatter plots of the principal components PC I vs PC II (Fig. 1) and for PC I vs PC III (Fig. 2) for all the taxa and individuals depict two clusters. These can be identified as a C. flava/MT cluster and C. viridula subsp. brachyrrhyncha/C. jemtlandica cluster. No sub-clusters were evident within the C. flava/MT cluster. All individuals from the MT cluster with the C. flava individuals and are not intermediate between C. flava and C. viridula subsp. brachyrrhyncha. From Figs 1 and 2, two sub-clusters within the C. viridula subsp. brachyrrhyncha/C. jemtlandica cluster associated with C. viridula subsp. brachyrrhyncha and C. jemtlandica may be identified. However, the separation is not complete. The first two principal components account for 46-6% of the variance within the data set, with PC I accounting for 32-5% and PC II accounting for 14-1%. PC III accounts for a further 11-5% of the variance within the data set. The variables with the highest component loadings (Table 3) on PC I are, in descending order, variables PERL, BEAKL, BRISTLE, MPEDL and LIGL. The first three variables are all associated with the perigynium. The highest component loading on PC II and PC III are for characters MSPIKE, VEGBAS, LIGL and NUTL, FGLUM respectively. Morphometric data from the 100 specimens of the C. flava agg. were further analysed using a stepwise DA. The initial DA was constructed using the four a priori groupings listed above. The program selected eight differentiating variables in the following descending order: MPEDL, BRISTLE, NUTL, BEAKL, MGLUM, VEGLEAF, LIGL and FGLUM. DA succeeded in correctly classifying 94% of the group cases (Fig. 3). The first canonical discriminant function (CDF I) accounted for 76-1% of the variance, with 17-4% and 6-5% of the variance being accounted for on CDF II and CDF III (not shown) respectively. The greatest contribution to variate one is given, in descending order, by MPEDL, BEAKL, BRISTLE and FGLUM. Along variate two NUTL, VEGLEAF and LIGL give the greatest contribution. A second DA was then carried out on the same data set with the exception that the C. flava and MT populations were condensed into one group. The discrimination efficiency improved to classify 99% of the group cases correctly. However, condensing the groups gave a similar pattern and spatial arrangement to that seen in Fig. 3. The first canonical discriminant function (CDF I) PC Il PC Ill CAREX FLAVA AGG. AT MALHAM TARN -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 PC | aC. flava AMTP @C. viridula subsp. brachyrrhyncha — © C. jemtlandica FIGURE 1. Scatter plot of PC I vs PC II from PCA of the measurements of C. flava , MTP, C. viridula subsp. brachyrrhyncha and C. jemtlandica. -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 PC | 4C. flava SMTP @C. viridula subsp. brachyrrhyncha — © C. jemtlandica FIGURE 2. Scatter plot of PC I vs PC III from PCA of the measurements of C. flava , MTP, C. viridula subsp. brachyrrhyncha and C. jemtlandica . 0.5 0.5 509 0.6 0.6 510 N. BLACKSTOCK AND P. A. ASHTON TABLE 3. LOADINGS FOR THE FIRST THREE PRINCIPAL COMPONENTS FROM PCA OF SPECIMENS OF C. FLAVA, THE MT POPULATION, C. VIRIDULA SUBSP. BRACHYRRHYNCHA AND C. JEMTLANDICA. Character PCI PC II PC Ill VEGLEAF 0-160 0-294 0-230 VEGBAS 0-277 0-452 -0-084 LIGL 0-328 0-422 -0-126 MSPIKE -0-197 0-550 0-131 MPEDL -0-399 0-234 -0-071 BEAKL 0-415 -0-134 -0-016 PERLE 0-446 0-034 0-041 MGLUM 0-209 -0-203 0-288 FGLUM 0-075 -0-258 -0-633 BRISTLE 0-407 0-023 -0-118 NUTL 0-077 -0-217 0-634 Abbreviations follow Table 2. accounted for 83-5% of the variance, with 16-5% of the variance being accounted for on CDF II. The greatest contribution to variate one is given, in descending order, by MPEDL, BRISTLE , BEAKL and FGLUM. Along variate two NUTL, VEGLEAF and LIGL give the greatest contribution. Results for the univariate analyses of the mean, standard deviation and range for each character are given in Table 4. A one-way ANOVA of each individual character showed no significant differences in the mean values between the taxa included in the C. flava and MT cluster. Significant differences were identified between the taxa in the C. flava/ MT cluster and the taxa included in the C. viridula subsp. brachyrrhyncha/C. jemtlandica cluster. However no single character could be identified that clearly delimited the taxa as all characters had some degree of overlap in their range. This is consistent with previous studies on the C. flava agg. (e.g. Schmid 1980; Havlickova 1982; Crins 1985; Pykaéla & Toivonen 1994). The perigynium characters BEAKL, PERLE and BRISTLE were found to be significantly different (p<0-05) between the C. flaval MT cluster and the C. viridula subsp. brachyrrhyncha!/C. jemtlandica cluster. Although the sample size of C. jemtlandica from Sweden is very small, PCA gives a clear disjunction along PC I between these and the MT plants. DA gives a similar disjunction along > CDF I. This is further supported by results of the ANOVA, which reveals statistically significant differences in the mean values for characters MSPIKE, BEAKL, BRISTLE, PERLE and NUTL (p<0-05) between the two groups. Within the C. viridula subsp. brachyrrhynchal/C. jemtlandica cluster the mean length of characters MPEDL and NUTL gave the clearest differentiation (p<0-001) between C. viridula subsp. brachyrrhyncha and C. jemtlandica with characters PERLE and BEAKL also being significantly different (p<0-01). DISCUSSION Morphometric analysis does not support the classification of C. jemtlandica as a subordinate taxon of C. flava. From PCA there is marked separation along PC I between C. jemtlandica and C. flava and between C. jemtlandica and the MT plants. However, C. jemtlandica is not clearly differentiated from C. viridula subsp. brachyrrhyncha although it does form a sub-cluster within this group. Differentiation between C. jemtlandica and C. flava and between C. jemtlandica and the MT plants is maintained in the results of the DA with marked separation along CDF I. Unlike the results from PCA C. jemtlandica forms a distinct cluster when CDF I is plotted against CDF I. However, this cluster is not clearly separated from the C. flava/MT or C. viridula subsp. brachyrrhyncha clusters along either CDF I or CDF II individually. These results support the close phenetic relationship that exists within the C. flava agg. C. viridula subsp. brachyrrhyncha and C. jemtlandica have been recognised as two distinct species under the names C. lepidocarpa (= C. viridula subsp. brachyrrhyncha) and C. jemtlandica respectively (Palmgren 1938; CAREX FLAVA AGG. AT MALHAM TARN Sil Canonical discriminant function II 6 -4 2 0 2 4 6 Canonical discriminant function | 4 C. flava 4 MTP @ C. viridula subsp. brachyrrhyncha © C. jemtlandica x Centroid for Carex flava © Centroid for MTP + Centroid for Carex viridula subsp. brachyrrhyncha x Centroid for Carex jemtlandica FIGURE 3. Scatter plot of CDF I vs CDF II for the measurements of C. flava , MTP, C. viridula subsp. brachyrrhyncha and C. jemtlandica . Halkka et al. 1992); however, they are generally united by Scandinavian botanists under the name C. lepidocarpa and accorded subordinate rank as C. lepidocarpa subsp. lepidocarpa and C. lepidocarpa subsp. jemtlandica (Palmgr.) Palmgr. (Palmgren 1926, 1959; Pykala & Toivonen 1994; Hedrén & Prentice 1996). C. lepidocarpa subsp. jemtlandica is described by Hedrén (1996) as endemic to Fennoscandia, largely replacing C. lepidocarpa subsp. lepidocarpa north of the limes norrlandicus. This northern distribution is reflected in the original name (nomen nudum) given to this taxon in 1910 by Palmgren as C. lepidocarpa Tausch * septentrionalis Palmgr. forma jemtlandica. Extensive studies on morphology (Hedrén 1994, 1996) and genetic differentiation using allozymes (Hedrén & Prentice 1996) support the recognition of this taxon as a subordinate of C. lepidocarpa (=C. viridula subsp. brachyrrhyncha). On the basis of work on demography, ecology, karyology, hybridisation and morphology of Swiss material Schmid (1983) reduced the status of C. lepidocarpa to that of a subspecies of C. viridula Michx. He further considered that the name of brachyrrhyncha Celakovsky to have priority over that of /epidocarpa. Former subspecies of C. lepidocarpa were thus recognised as subordinate taxa of C. viridula subsp. brachyrrhyncha and given varietal status. The exception to this was C. jemtlandica, which Schmid did not recognise at any level (cf. Schmid 1983). Based on biometric analysis of herbarium specimens Crins (1985) similarly concluded that C. jemtlandica did not merit recognition at any taxonomic level. However, the studies by Schmid (1983) and Crins (1985) included specimens of C. jemtlandica collected from a greater geographical range, extending as far south as Switzerland, than the range of the taxon recognised by Hedrén (1996). Crins further renamed C. viridula subsp. brachyrrhyncha var. lepidocarpa as C. viridula subsp. brachyrrhyncha var. elatior (Schltdl.) Crins. Studies of allozyme variation by Bruederle & Jensen (1991) within the C. flava agg. concluded that C. flava was enzymatically distinct from C. viridula, but that within C. viridula agg. clear differentiation was not possible. However, they noted that C. viridula subsp. brachyrrhyncha formed a relatively distinct group within the aggregate. They concluded that their data were insufficient to justify a reclassification of this taxon and so they retained the classification of Schmid. They also reported that C. jemtlandica was not separable from C. flava using allozymes. This is in contrast to Hedrén & Prentice (1996) who found in their larger study of C. jemtlandica that it was enzymatically distinct. Pykéla & Toivonen (1994) suggest that Bruederle & Jensen may have included specimens of C. flava in their study rather than C. jemtlandica. S12 N. BLACKSTOCK AND P. A. ASHTON TABLE 4. MEAN, RANGE AND + 1 SD FOR MORPHOLOGICAL CHARACTERS MEASURED FOR SPECIMENS OF C. FLAVA, THE MT POPULATION, C. VIRIDULA SUBSP. BRACHYRRHYNCHA AND C. JEMTLANDICA. Character C. flava MT C. viridula subsp. C. jemtlandica (n=40) (n=10) brachyrrhyncha (n=40) (n=10) VEGLEAF 53-41 (+18-48) 39-28 (45-97) 40-95 (+19-17) 55-82 (427-70) 21-41-105-05 31-77-52-79 11-22-107-76 21-27-109-83 VEGBAS 5-46 (+1-48) 5-10 (41-24) 4-23 (+1-28) 3-69 (+0-47) 3-02-11-78 3:50-7:29 2:30-8:79 2:89-4:54 LIGL 3-68 (+1-57) 2°97 (41-07) 2:04 (41-58) 1-21 (+0-37) 1-15-8-25 1-50-5-21 0-81—9-42 0-70-1-80 MSPIKE 11-48 (44-25) 9-72 (+1-80) 15-17 (45-18) 15-64 (+5-01) 3-81-22-01 6-79—13-12 5-32-24-89 9-63—25-51 MPEDL 1-61 (+1-85) 1-03 (41-21) 13-86 (+7-50) 6-41 (+3-06) 0:00-6:58 0-00-2-42 0-00-36-34 3-81-11-24 BEAKL 2-18 (+-038) 252 (20-27) 1-45 (40-31) 1-82 (+0-17) 1-17—3-00 1-94—2-82 0-73—2-38 1-61-2-01 PERLE 5-08 (+0-73) 5-14 (+0-46) 3-80 (40-27) 4-49 (+0-57) 4-02-6-63 4-51-5-74 3-12-4-48 3:72-5:28 MGLUM 3-19 (+0-44) 3-11 (+0-39) 2-73 (+0-38) 3-01 (40-51) 2-27-4-23 2-11-4-01 2-06-3-45 2-35—3-65 FGLUM 2-09 (40-29) 2-00 (40-31) 1-92 (+0-36) 1-69 (+0-21) 1-53-3-10 1:72-2:61 1-14—2-65 1-38—1-99 BRISTLE #33 (42-03) 8-40 (+1-71) 3-50 (+1-92) 4-10 (+0-74) 3-11 5-11 0-8 3-5 NUTL 1-60 (+0-13) 1-73 (40-11) 1-56 (+0-19) 1-96 (+0-17) 1-37-1-91 1-61-1-91 0-92-1-38 1-68-2:-21 Abbreviations follow Table 2. Although the present study shows that morphological differences between C. viridula subsp. brachyrrhyncha and C. jemtlandica are not fully developed there is a degree of separation indicated by the PCA and DA analyses. The same chromosome number of 2n=68 and limited morphological distinction supports the recognition of these taxa at a very close taxonomic level. However, caricologists working in Fennoscandia, where both C. viridula subsp. brachyrrhyncha and C. jemtlandica occur, consistently classify them as distinct taxa (e.g. Palmgren 1959; Pykala & Toivonen 1994; Hedrén 1996). We support the recognition of C. jemtlandica at a subordinate level to that of C. viridula subsp. brachyrrhyncha. In order to bring the current nomenclature for C. jemtlandica into line with that used by Crins (1985) it is proposed that this taxon be renamed C. viridula Michx subsp. brachyrrhyncha (Celak.) B. Schmid var. jemtlandica (Palmgr.) Blackstock & P. A. Ashton comb. et stat. nov. The hypothesis that the MT plants are of hybrid origin is not supported by this study. Jermy et al. (1982) preferred to recognise hybrids as being both morphologically intermediate between the parents and also being wholly sterile. However, they acknowledged that the MT plants are fertile. It is implied that this fertility may be due to the population being a stabilised introgressed hybrid between C. flava and C. viridula subsp. brachyrrhyncha, with the former species now being extinct at this site. In Switzerland natural and cultivated hybrid populations between C. flava and C. viridula subsp. brachyrrhyncha have been studied by Schmid (1983). Fl-hybrids were found to be almost completely sterile with 0% seed set and pollen fertility of only 0-c.3% (Schmid 1983). When directional backcrosses do arise, forms identical to the parent species in both morphology and fertility may eventually result. In these populations it is C. viridula subsp. brachyrrhyncha that may be out-competed and eventually ousted by C. flava, the Fl-hybrid or backcrosses with either parent (Schmid 1980). This does not fit with the model cited by Jermy et al. (1982) who suggested that C. flava was probably eliminated by introgression with C. viridula subsp. brachyrrhyncha, and the putative hybrids at Malham Tarn, Greywell Moors and Lough Corrib being a product of CAREX FLAVA AGG. AT MALHAM TARN 513 TABLE 5. PERIGYNIUM LENGTHS (MM) FOR C. FLAVA Author and Year Area Perigynium length (mm) Robertson (1984) North America 3-0-6-0 Hedrén, in itt. Sweden 3-5+ Havlickova (1982) Czechoslovakia 3-8-6:3 Crins (1985) North America 4-0-6:3 Mackenzie (1935) North America 4-5-6-0 Schmid (1985) Europe 4-5-6-0 Palmgren (1958) Scandinavia 4-5-6-7 Hultén (1968) North America 5-0-6-0 Chater (1980) Europe 5-0-7-0 Nelmes (1945) UK 5-0-7-0 Sell & Murrell (1996) UK 5-0-7-0 Davies (1953) UK 5-25-7-0 Stace (1997) UK 5-5-6-°5 Jermy et al. (1982) UK 6-6-5 Rose (1989) UK 6-6-5 this process. The hypothesis that the MT plants are intermediate in morphology is also unsupported, as there is marked separation in the PCA along PC I between C. viridula subsp. brachyrrhyncha and the MT plants, but no separation between C. flava and the MT plants. A similar separation occurs along CDF I using DA. Based on the biometric data presented here the MT plants are aligned with C. flava s. s. A chromosome count of a specimen from the MT population by Davies (1953) gave a haploid number of n=30. This is consistent with other chromosome counts for C. flava s. s. (Davies 1953; Halkka et al. 1992; Schmid 1982) but inconsistent for chromosome counts for C. viridula subsp. brachyrrhyncha (n=34) (Davies, 1953; Halkka et al. 1992; Schmid 1982) and C. x alsatica (n=irregular) (Schmid 1982). Difficulties over the correct classification of the MT plants may be due, in part, to its apparently obvious morphological dissimilarity to the Roudsea Wood population of C. flava. It should be noted that although both populations are situated at approximately the same latitude, the Roudsea Wood population is situated in a semi-shaded wood near sea level whereas the MT population is situated on an exposed site at approximately 380 m. It would appear that these two populations are approaching the opposite extremes of morphological range within this species. When continental European and North American material is taken into consideration it becomes apparent that both populations fit within the phenotypic range of this species. Table 5 indicates that there may be an over reliance upon large utricle size to separate C. flava from its closely related allies within the British Isles. In conclusion the population of yellow sedges discovered by G. A. Shaw in 1946 at the Tarn Moss, Malham should be recognised as a second extant population of C. flava s. s. in the British Isles. Furthermore, problematic populations whose identity is uncertain such as those already identified as C. flava hybrids at Greywell Moors, Hants and by the River Corrib near Menlough north-east of Galway City need to be reviewed as soon as material is available. In light of the evidence presented here it is possible that C. flava is not as rare as currently thought in the British Isles but has been overlooked or misidentified in the past. TAXONOMIC TREATMENT The following key should serve to discriminate between C. flava, C. viridula subsp. brachyrrhyncha var. elatior and C. viridula subsp. brachyrrhyncha var. jemtlandica . However, it should be noted that no single character can satisfactorily discriminate between the taxa and where two or more of the taxa are sympatric the taxonomic affinities may be further confused. Where possible several specimens from one population should be carefully compared and mean character values used. Only perigynia from the middle of a pistillate spikelet should be used. 514 N. BLACKSTOCK AND P. A. ASHTON KEY TO TAXA: 1 Utricles straight or curved, usually <5 mm long, abruptly contracting into a conspicuous beak usually <2 mm long with usually <5 bristles on each side of the beak margin. Leaves of the fertile tiller usually <4-5 mm wide, longer or shorter than the culm. Ligule usually <2-5 mm long acute to rounded. Solitary staminate spikelet + sessile to conspicuously pedunculate, PELMINAL ose.c 500k 55 vacantncs ivceenewssananagtevsicervecderiinectea Ueius enecesunebzoneccececnet heme auansnes Remnenee ees eae ee 2 1 Utricles curved usually >3-8 mm long, gradually tapering into a long 1-5-3 mm beak with a conspicuously scabrous margin of usually 5—11 bristles on each side. Leaves of fertile tiller usually >3 mm wide, usually + as long as the culm. Ligule usually >3 mm, acute. Solitary Staiminate spikelet + sessile; terminal <.....:.0.1.:5:chccscienessscceeveeeese eer 1. C. flava 2 Utricles usually 3-5-5 mm long; beak 1-5-2 mm long. Leaves of fertile tiller usually <3 mm WICC vis sciatic dcnancantccccess cada cedsiecedtanceaanebcdanddunetnudeeeciehlbebiaccespeaceeadeagecapeteceeees:aee eee eee ee 3 2 Utricles usually <4 mm long: beak usually <1-5 mm long. Leaves of fertile tiller 1-5—-4-5 mm WYN sieora ices tice tynseisaauh tater tees ctipnceveas ee Renee aeem tenes 2. Other members of the C. viridula agg. 3 Utricles + dense; beak usually deflexed. Staminate spikelet usually conspicuously pedunculate. Pistillate spikelets usually 2—3, + separate, ovate to cylindrical. Leaves of fertile tiller usually shorter than half of the Culm scssssecececesesesenss 2aa. C. viridula subsp. brachyrrhyncha var. elatior 3 Utricles + lax; beak usually straight. Staminate spikelet usually sub-sessile to shortly pedunculate. Pistillate spikelets usually 1-2 + contiguous to approximate, globose to ovate. Leaves of fertile tiller usually longer than half of the culm ee eee ere en er ere 2ab. C. viridula subsp. brachyrrhyncha var. jemtlandica The following is a brief taxonomic treatment and includes only some of the most important synonyms. For a fuller taxonomic treatment see Schmid (1983) and Crins (1985). 1. C. flava L., Sp. Pl. 2: 975 (1753). Type: LINN Savage Cat. No. 1100.40 (Holotype: LINN). Synonyms: C. patula Host., Icon. Descr. Gram. Austriac. 1: 48-49 (1801). Type: Austria (Holotype: W, destroyed). C. flava subsp. gaspensis Fernald, Rhodora 8: 200 (1906). Type: Canada, Quebec, gravelly banks of Bonaventure River, 8 August 1904, J. F. Collins, M. L. Fernald & A. S. Pease s.n. (Lectotype: GH, n.v. selected by Crins 1988). 2. C. viridula Michx. Fl. Bor-Amer. 2: 170 (1803). Type: Canada, between Montréal and Les Trois Rivieres, Michaux in herb. Michaux (Holotype: P) 2a. C. viridula subsp. brachyrrhyncha (Celakovsky) B. Schmid, Watsonia 14: 317 (1983). Basionym: , C. flava subsp. brachyrrhyncha Celakovsky, Prodr. Fl. Bom. 1: 71 (1876). Synonyms: C. lepidocarpa Tausch, Flora (Regensb.) 17: 179 (1834). Type: Czechoslovakia, Praha, (no collector), no. 1636 (lectotype: PRC, isolectotype: PRC, selected by Crins 1985). 2aa. C. viridula subsp. brachyrrhyncha var. elatior (Schlectendal) Crins, Canad. J. Bot. 67(4): 1058 (1989). Basionym: C. flava var. elatior Schltdl., Fl. Berol. 1: 477 (1823). Type: Germany, Berolinensi (Holotype: B n.v.). Synonyms: C. viridula subsp. brachyrrhyncha var. lepidocarpa (Tausch) B. Schmid Watsonia 14: 317 (1983). C. lepidocarpa Tausch, Flora (Regensb.) 17: 179 (1834). C. lepidocarpa var. lepidocarpa (Tausch) Petermann, Anal. Pflanzenschl. Leipzig 509 (1846), pro autonym. C. flava subsp. lepidocarpa (Tausch) Godr. Fl. Lorraine. 1: 118 (1843). C. flava subsp. lepidocarpa (Tausch) Nyman, Consp. 771 (1882). C. flava subsp. lepidocarpa (Tausch) Lange, Haandb. Danske. FI., 4th ed., 144 (1886). CAREX FLAVA AGG. AT MALHAM TARN 515 2ab. C. viridula subsp. brachyrrhyncha var. jemtlandica (Palmgr.) Blackstock & P. A. Ashton comb. et stat. nov. Basionym: C. lepidocarpa Tausch subsp. jemtlandica Palmgr. in Lindman. Sv. Fanerogamfl., ed. 2: 153 (1926). Type: Sweden, Jamtland, Ostersund, Frés6n, 13 August 1910, A. Palmgren, Caric. Fulv. No. 53 (lectotype: H, selected by Crins 1985). Synonyms: C. lepidocarpa Tausch * septentrionalis Palmgr. forma jemtlandica Palmegr., Caric. Fulvellae Exs. No. 52-54 (1910) (nom. nud.). C. jemtlandica (Palmgr.) Palmgr., Mem. Soc. Fauna Flora Fenn. 13: 126 (1937). C. jemtlandica (Palmgr.) Palmgr. var. gotlandiae Palmgr., Comment. Biol. 20(3): 12 (1958). Type: Sweden, Gotland, par. Othem, 14 June 1910, A. Palmgren (Holotype: H, Isotype: H, photo!). C. jemtlandica (Palmgr.) Palmgr. var. kainuensis Palmgr., Comment. Biol. 20(3): 13 (1958). Type: Finland, Ostrobottnia kajanensis, Suomussalmi, Ruhtinaansalmi, Rytisuo, 10 July 1937, O. Kyyhkynen (Holotype: H, photo!). C. flava L. subsp. jemtlandica (Palmgr.) P. D. Sell Fl. of Great Britain and Ireland 5: 110 (1996). ACKNOWLEDGMENTS This research has been carried out as part of a research studentship funded by Edge Hill College of Higher Education. The authors would like to thank English Nature and The National Trust for granting permission to collect samples from Roudsea Wood and Malham Tarn respectively. We would also like to thank everyone who supplied seed and ramets for cultivation. Special thanks go to Mikael Hedrén for suggesting localities for material from Sweden and his input on the C. jemtlandica problem. REFERENCES BREWIS, A., BOWMAN, P. & ROSE, F. (1996). The Flora of Hampshire. Harley Books, Colchester. BRUEDERLE, L. P. & JENSEN, U. (1991). Genetic differentiation of Carex flava and Carex viridula in Western Europe (Cyperaceae). Systematic botany 16: 41-49. CHATER, A. O. (1980). Carex L., in TUTIN, T. G. et al., eds. Flora Europaea 5: 254. Cambridge University Press, Cambridge. CRINS, W. J. G. (1985). A Taxonomy of Carex section Ceratocystis in North America and Northern Eurasia. Unpublished Ph.D. thesis, University of Toronto. DAVIES, E. W. (1953). An experimental taxonomic study of some species of Carex with special reference to the Carex flava agg. Unpublished Ph.D. thesis, University of London. GOODENOUGH, S. (1792). Observations on the British species of Carex. Transactions of the Linnean Society of London 2: 126-211. GREEN, C. T. (1933). The Flora of the Liverpool District. T. Buncle & Co., Arbroath. HALKKA, L., TOIVONEN, H., SAARIO, S. & PYKALA, J. (1992). Chromosome counts in the Carex flava complex (Cyperaceae) in Finland. Nordic journal of botany: 12: 651-655. HALLIDAY, G. (1997). A Flora of Cumbria: comprising the vice-counties of Westmorland with Furness (v. c. 69), Cumberland (v. c. 70) and parts of North-west Yorkshire (v. c. 65) and North Lancashire (v. c. 60). Centre for North-west regional studies University of Lancaster, Manchester. HAVLICKOVA, J. (1982). Carex flava complex in the Czech lands I. Analysis of the variability of morphological characters. Preslia (Praha) 54: 201-222. HEDREN, M. (1994). Morfologisk variation inom nabbstarr och jamtstarr (Carex lepidocarpa s.l.) i Sverige. Svensk Botanisk Tidskrift 88: 129-141. HEDREN, M. (1996). Genetic differentiation among Finnish, Norwegian and Swedish populations of Carex lepidocarpa s. lat. (Cyperaceae). Symbolae botanicae Upsaliensis 31: 105-113. HEDREN, M. (1998). Status of Carex bergrothii (Cyperaceae) on Gotland, S.E. Sweden. Nordic journal of botany 18: 41-49. HEDREN, M. & PRENTICE, H. C. (1996). Allozyme variation and racial differentiation in Swedish Carex lepidocarpa s. I. (Cyperaceae). Biological journal of the Linnean Society 59: 179-200. HULTEN, E. (1968). Flora of Alaska and neighbouring territories. Stanford University Press, Stanford. JERMY, A. C. & TUTIN, T. G. (1968). British Sedges. Botanical Society of the British Isles, London. 516 N. BLACKSTOCK AND P. A. ASHTON JERMY, A. C., CHATER, A. O. & DAVID, R. W. (1982). Sedges of the British Isles. Botanical Society of the British Isles, London. KOVACH, W. L. (1998). MVSP- A multi-variate statistical package for Windows, ver. 3.0. Kovach Computing Services, Pentraeth, Wales, U.K. MACKENZIE, K. K. (1935). Cyperaceae-Cariceae. North American Flora 18: 1-478. NACZI, R. F. C., REZNICEK, A. A. & FORD, B. A. (1998). Morphological, geographical, and ecological differentiation in the Carex willdenowii complex (Cyperaceae). American journal of botany 85: 434-447. NELMES, E. (1945). Two critical groups of British sedges. Bot. Soc. Exch. Club Br. Isles 13: 95-105. PALMGREN, A. (1926). C. lepidocarpa subsp. jemtlandica, in LINDMAN, C. A. M., Svensk fanerogamflora, 2nd ed., p. 153. Stockholm, Nordstedt & Soner. PALMGREN, A. (1938). Carex jemtlandica, en ny Carex-art. Memoranda Societatis pro Fauna et Flora Fennica 13: 126-128. PALMGREN, A. (1958). Nagra nya former inom Carex-gruppen Fulvellae Fr. Comment. Biol. 20(3): 3-15. PALMGREN, A. (1959). Carex-gruppen Fulvellae Fr. i. Fennoskandien I. Societas pro Fauna et Flora Fennica 2: 1-165. PERRING, F. H. (1970). Vascular plants new to Ireland. Watsonia 8: 91. PYKALA, J. & TOIVONEN, H. (1994). Taxonomy of the Carex flava complex (Cyperaceae) in Finland. Nordic journal of botany 14: 173-191. ROBERTSON, A. (1984). Carex of Newfoundland. Newfoundland Forest Research Centre, St Johns. ROSE, F. (1989). Colour identification guide to the grasses, sedges, rushes and ferns of the British Isles and north-western Europe. Viking, London. ROTHROCK, P. E., REZNICEK, A. A. & GANION, L. R. (1997). Taxonomy of the Carex straminea complex (Cyperaceae). Canadian journal of botany 75: 2177-2195. SCHMID, B. (1980). Carex flava L. s. 1. 1m Lichte der r-selektion. Mitteilungen aus dem Botanischen Garten und Museum der Universitat Ziirich 322: 1-360. SCHMID, B. (1982). Karyology and hybridization in the Carex flava complex in Switzerland. Feddes Repertorium 93: 23-59. SCHMID, B. (1983). Notes on the nomenclature and taxonomy of the Carex flava group in Europe. Watsonia 14: 309-319. SELL, P. & MURRELL, G. (1996). Flora of Great Britain. Butomaceae-Orchidaceae. Vol. 5. Cambridge University Press, Cambridge. SHAW, G. A. (1946) Two plants new to Yorkshire: Oenanthe fluviatilis Coleman, and Carex flava L. Naturalist 71: 138. SHEPHERD, G. J. (1975). Experimental taxonomy in the genus Carex section Vesicariae. Unpublished Ph.D. Thesis, University of Edinburgh. SPSS (1998). S.P.S.S.® Statistical Algorithms. S.P.S.S. Inc., Chicago. STACE, C. A. (1997). New Flora of the British Isles, Cambridge University Press, Cambridge. TAYLOR, I. (1999). Carex flava L., in WIGGINTON, M. J. (1999). British Red Data Books: 1 Vascular Plants. Joint Nature Conservation Committee, Peterborough. (Accepted February 2001) Watsonia 23: 517-523 (2001) DLT Conservation of Britain’s biodiversity: Hieracium linguans (Zahn) Roffey (Asteraceae), Tongue Hawkweed T. C. G. RICH Dept. Biodiversity and Systematic Biology, National Museum & Gallery, Cardiff CF 10 3NP and G. S. MOTLEY Countryside Council for Wales, Cantref Court, Brecon Road, Abergavenny, Mons. NP7 7AX ABSTRACT Hieracium linguans, Tongue Hawkweed, is a very rare endemic Welsh species. It was first found in 1898 by A. Ley in the Brecon Beacons (v.c. 42), Wales. It was initially named as a variety and later raised to species level, its current status. It is a distinct member of Section Tridentata. The historical herbarium and literature records were somewhat confusing, but indicate three localities in the Tawe Valley. In 1999 and 2000, only one population of about 80 plants was found in Cwm Haffes (the type locality); it was not refound in the other two localities. Records from Upper Nedd Glen, Llyn y Fan Fach and the Upper Tywi are probably errors. It is a polycarpic perennial which flowers in July-August, and regenerates readily from wind-dispersed seeds. Although it is very rare, it does not seem to be significantly at risk. Seed has been deposited in the Millennium Seed Bank and it is being cultivated at the National Botanic Garden of Wales. KEYWORDS: Compositae, Endemic, Lectotype, rare species, Wales. INTRODUCTION Hieracium linguans (Zahn) Roffey (Asteraceae), Tongue Hawkweed (from the Latin lingua, meaning tongue), is a very rare, endemic species. Along with 79 other rare hawkweeds, it has been included in the 3rd edition of the Vascular Plant Red Data Book (Wigginton 1999). In 1999 David McCosh, who provided the original list of species for the Red Data Book, suggested that this species would be worth re-surveying as there were no recent records. A joint project was therefore set up between the National Museum & Gallery, Cardiff and the Countryside Council for Wales to establish its current status and determine its need for conservation. In this paper the information about H. linguans is summarised; full details can be found in Rich (1999), updated here with additional field work in July 2000. TAXONOMY Plants that are now called Hieracium linguans were first found by Augustin Ley in the Brecon Beacons on 2-3 August 1898. He subsequently distributed specimens through the Botanical Exchange Club (B. E. C.) as ‘H. gothicum?’ (Ley 1900). When it became apparent that there was no name for the plants, Ley described them as var. lingua of H. sparsifolium Lindeb. in Linton (1905), and subsequently distributed many more cultivated specimens through the B. E. C. (Ley 1907, 1908). Zahn (1922) raised Ley’s variety to subspecies level, but transferred it to H. laevigatum Willd. and provided it with the new subspecific name linguans Zahn in the process (probably to avoid confusion with another taxon of uncertain status, subsp. lingua Dahlst.). Roffey (1925) raised it to a species, its current status. The two synonyms of H. linguans (Zahn) Roffey are thus H. laevigatum Willd. subsp. linguans Zahn (basionym) and H. sparsifolium Lindeb. var. lingua Ley ex W. R. Linton. 518 T. C. G. RICH AND'G. S. MOTLEY A type specimen (probably selected by H. W. Pugsley and designated here as the lectotype), labelled ‘Haffes Glen, West Brecon, 2 August 1898, A. Ley’ is in the Natural History Museum (BM). Hieracium linguans belongs to sect. Tridentata (Fr.) Arv.-Touv., of which there are at least twelve species in South Wales. Pugsley (1948) gave its distinguishing features as the oblong- elliptic to linear-oblong, sinuate-dentate medium stem leaves, the racemose-corymbose inflorescence with 4-8 large capitula on long, suberect branches, the greyish-green broad, densely floccose phyllaries with numerous unequal, short, dark-glandular and fewer longer dark-based pilose hairs, and dark styles. However, Pugsley’s description includes cultivated material, which is more robust than the wild material. A comparison of plants in the field with wild-collected specimens in herbaria suggests that the latter were also selected as the larger plants. Most plants observed in the wild in 1999 (see below) were less than 20 cm tall, had only one stem (rarely two stems) and up to four capitula (rarely up to 12 capitula). DISTRIBUTION HISTORICAL INFORMATION Locality and habitat information was extracted from herbarium sheets at BIRM, BM, CGE, LIV, OXF and NMW and from the literature (e.g. Perring 1968, Ellis 1983). Records from the Hieracium database were provided by D. McCosh. No records are held in the Rare Plants Database used to compile the Red Data Book (Wigginton 1999). There has been significant confusion about the records, possibly partly due to a note from a B.E. C. report probably referring to H. scoticum F. J. Hanb. being attached to a H. linguans specimen in BM, and partly to some specimens which have subsequently been identified as different taxa. The B.E.C. note (Ley 1900) indicated that ‘H. gothicum Backh?’ had been found in Cwm Haffes (Capel Cellwen, now Callwen), Upper Nedd Glen and above Llyn y Fan Fach, but the style colour was noted as pure yellow which is not a character of H. linguans. Ley in Linton (1905) noted H. linguans from Cwm Haffes and the Upper Nedd Glen when he described it as a new variety and later noted ‘this remarkable plant has not yet been detected except in West Brecon’ (Ley 1909). Pugsley (1948) only cited material from Cwm Haffes. Sell & West (1968) stated H. linguans was only recorded from rocky ledges in three localities in Brecon, and records were plotted in SN81, SN82 and SN84 on the map. The Biological Records Centre, Monks Wood holds details of the records in the Critical Supplement, but the cards for H. linguans could not be traced in 1999 (J. Croft, pers. comm., 1999). The SN81 record appears to be based on the well-known specimens from Cwm Haffes. The SN82 record may be based on 1906 material cultivated by A. Ley from Fan Fechan (CGE) originally determined as H. linguans in 1958 by P. D. Sell and C. West, but later redetermined as H. scoticum by P. D. Sell in 1983. The SN84 record is probably based on a specimen collected in 1955 by J. E. Raven in Llanwrtyd (CGE), later redetermined as ‘H. ?subcrocatum by P. D. Sell in 1983. Ellis (1983) cited a record from the Upper Tywi based on a specimen collected by A. Ley in 1906 in CGE; the only such specimen is from the Upper Tawe Glen, and thus probably a confusion with the two adjacent river valleys with similar names. Records for the Nedd Glen (which would probably have been plotted in either SN81 or SN91) do not appear to have been accepted by recent authors and no material has been traced. The historical records traced are summarised in Table 1. Some of the records may be duplicates labelled slightly differently (e.g. ‘streamside near Capel Cellwen’ and ‘Haffes Glen’ on 2 August 1898; Ley was not consistent in his labelling of specimens), and some dates are suspiciously one month or one year apart. There appear to be three distinct sites: 1. waterfall at the head of Cwm Haffes; 2. stony delta of the Haffes (“streamside near Capel Cellwen’ assumed to be this too); 3. riverside rocks, Upper Tawe Glen. CONSERVATION OF HIERACIUM LINGUANS 519 TABLE 1. RECORDS OF HIERACIUM LINGUANS Date Collector Site Source and notes 2/8/1898 A. Ley Capel Cellwen, streamside near BIRM, BM, LIV; det. P. D. Sell & C. West. 2/8/1898 A. Ley Haffes Glen, West Brecon BM; type of var. lingua Ley. 2/8/1898 A. Ley Haffys Glen, near Callwen BM; det. D. McCosh. 27/7/1899 A. Ley Haffys Glen CGE; det. P. D. Sell. 27/8/1899 A. Ley Waterfall in the Haffes Glen BM; det. P. D. Sell & C. West. 4/7/1906 - Haffes Glen, Callwen LIV; det. J. N. Mills. 5/7/1906 A. Ley riverside rocks, Upper Tawe Glen* CGE; det. P. D. Sell. 10/7/1906 A. Ley stony delta of the Haffes* CGE; det. P. D. Sell. 12/7/1906 H. J. Riddelsdell R. Haffes BM: det. D. McCosh 25/7/1955 J. E. Raven Cwm Haffes CGE; det. P. D. Sell & C. West. 25/7/1955 C.E. Andrews head of Cwm Haffes, rocks by water-BIRM; conf. T. Rich. fall 28/7/1956 J. E. Raven (unlabelled, assumed Cwm Haffes) herb. D. McCosh. 26/7/1957 B.A. Miles side of large waterfall, Cwm Haffes CGE; det. P. D. Sell & C. West. Cultivated material 23/6/1906 A. Ley West Brecon, cultivated at Sellack BM, CGE, LIV, NUW, OXF (Ley 1907). muIso7 =A. Ley : West Brecon, cultivated BM (Ley 1908). 26/6/1908 A. Ley Brecon, cultivated BM (labelled incorrectly as syntype), CGE. 17/7/1909 E. F. Linton cultivated at Edmondsham NMW: det. D. McCosh. Errors 2/8/1898 A. Ley Cliffs of Llyn-y-fan-fechan Ley (1900); confusion of B.E.C. labels. 3/8/1898 branches pedicels length, length, anther cm mm their > their and mm mm _ length, mm spikelets spikelets branches Tiverton, wild (n=1) 15 14-1 8-0 3-0 11-0 19-0 9-1 3-4 Tiverton, cultivated (n=13) 141 14-7 7-1 2:3 9-4 21-0 9.2 3-6 Exeter, wild (n=1) 110 16-0 5-0 2-0 7-0 18-0 9-8 2:8 Exeter, cultivated (n=12) PF 15-8 6-2 1-1 73 19-3 10-2 2-9 Kennford, wild (n=1) 140 16-3 5-0 2-0 7-0 18-0 9-5 3-2 Kennford, cultivated (n=11) 147 15-9 4.4 3-1 7:5 19-9 9.2 2-9 Dartington, wild (n=1) 129 13-5 6-0 - 6-0 18-0 9-0 3-1 Dartington, cultivated (n=6) | 13-1 Dei - 5-7 17-6 9-3 2:9 Bicton, wild (n=1) 135 18-0 6-0 2-0 8-0 20-0 10-5 3-5 Bicton, cultivated (n=10) 122 16-4 4-] 2:3 6-4 19-1 11-0 3-3 Hams Barton, wild (n=1) 172 15-0 3-0 4-0 7-0 14-1 8-8 - Hams Barton, cultivated (n=11) 161 14-3 3-8 4-] 7-9 14-3 9-0 3-5 Figures presented are the measurements from the single wild parents and mean of the cultivated progeny. Parent plants were collected as follows: Tiverton, road lay-by $X996143, 5 May 1999. Exeter, bank above road SX962916, 9 May 1999. Kennford, road verge, SX918875, 15 May 1999. Dartington, disturbed ground, SX787621, 21 May 1999. Bicton, disturbed ground SX072868, 22 May 1999. Hams Barton, edge of wheat field, SX882801, 26 May 1999. DISCUSSION The ‘longipedicellatus’ taxon was thus found to differ from the other taxa investigated. It is most closely related to B. hordeaceus subsp. hordeaceus, but differs in its long culms, long panicles with long pedicels and/or long branches and long chasmogamous undehisced anthers. It superficially resembles B. commutatus var. pubens in being 80-175 cm tall with panicles 10-20 cm long but its lemmas are papery and like B. racemosus and B. arvensis it has long chasmogamous anthers. As the grass is clearly a distinct taxon it is it is named Bromus hordeaceus subsp. longipedicellatus. Bromus hordeaceus L. subsp. longipedicellatus L. M. Spalton, subsp. nov. Varietas robusta. Caules 80-175 cm. Paniculae 10-20 cm, quattuor minimum pedicellis aut ramis longitudinem spicularum suarum excedentibus (ramis ad basem spiculae terminalis mensis). Lemmae 7-5—10-5 mm, pubescentes, texturae papyraceae, venis in siccitate extantibus. Antherae adhuc intactae infimorum florum 2-2—3-8 mm, ceterae breviores. A robust grass with culms 80-175 cm long. Panicles 10-20 cm long with at least 4 pedicels or branches exceeding the length of their spikelets (branches measured to the base of the terminal spikelet). Lemmas 7-5—-10-5 mm long, of papery texture, with protruding veins when dried, pubescent or rarely glabrous. Undehisced anthers from the lowest florets 2-2-3-8 mm long, other anthers smaller. j HOLOTYPUS: South Devon, v.c. 3, Aveton Gifford (SX691474), grassy bank on thin soil above road, 23 May 1999, L. M. Spalton s.n. (BM). Subsp. longipedicellatus is found on road verges and banks, on waysides, on the edges of arable fields (especially wheat), on waste ground and less frequently in pasture. On road verges it is in flower from the beginning of May and is very soon cut during roadside trimming. When in fruit (and when dried) the spikelets lie close to the culm so the panicle no longer appears to be lax. BROMUS HORDEACEUS SUBSP. LONGIPEDICELLATUS 531 Subsp. longipedicellatus has been confirmed in 46 vice-counties of England, eight of Wales, six of Scotland and one of Ireland. Out of 1789 herbarium specimens that were examined only 28 were subsp. longipedicellatus, the earliest of which was collected by G. H. Douglas in Torbay, Devon in 1926 (TOR). This suggests that, whatever the origin of subsp. longipedicellatus, it may have spread rapidly in England and Wales. It is probable that the grass may have been disseminated as a contaminant of Lolium perenne seed in agricultural and amenity grass seed mixtures or of wheat seed. Subsp. longipedicellatus might have resulted from a gene transfer from B. hordeaceus into B. racemosus or B. arvensis with hordeaceus genes becoming dominant. B. commutatus var. pubens has pubescent, less leathery (though still papery) lemmas, a variable indumentum of the lower leaf-sheath and an earlier flowering time than B. commutatus and it might contain genes from hordeaceus. On the rare occasions when B. racemosus and B. commutatus share the same habitat, intermediates are produced. There are many problems in the genus and DNA or isoenzyme studies might resolve at least some of them. A new key to the genus Bromus in Britain will be produced when additional research has been completed. An important consequence of the recognition of subsp. Jongipedicellatus is that the descriptions of B. hordeaceus and subsp. hordeaceus now require revision because B. hordeaceus cannot be reliably separated from B. racemosus and B. commutatus by panicles that are dense or lax. The following descriptions are proposed: Bromus hordeaceus L.: Culms usually erect 2-175 cm. Panicle dense or lax, 1-20 cm, sometimes only 1 spikelet. Lemmas 6-5—10-5 mm, papery with protruding veins when dried, pubescent or glabrous. Undehisced anthers from the lowest florets 1-3-8 mm. Bromus hordeaceus subsp. hordeaceus: culms erect or ascending to 80 cm. Panicle 1-10 cm, dense or somewhat lax with not more than 3 pedicels or branches exceeding the length of their spikelets. Lemmas 7—10-5 mm, papery with protruding veins when dried, pubescent or glabrous. Undehisced anthers from the lowest florets 1-2-6 mm. ACKNOWLEDGMENTS I thank C. J. Bruxner, S. P. Chambers, A. Copping, R. J. Hutchinson and L. J. Margetts for their collaboration in cultivation experiments in their localities, the Principal of Bicton College of Agriculture for providing me with experimental cultivation facilities, the grass specialists for comments on the specimens sent to them, the Keepers of herbaria for the generous loan of specimens and W. H. Tucker for the Latin diagnosis. Special thanks are due to L. J. Margetts for first detecting this grass and for constructive comments on an earlier draft, to T. Rich for additional comments and to the many botanists who have sent Bromus specimens to me. I hope that they will continue to do so. REFERENCES CLAPHAM, A. R.,TUTIN, T. G. & WARBURG, E. F. (1962). Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. CLAPHAM, A. R.,TUTIN, T. G. & Moore, D. M. (1987). Flora of the British Isles, 3rd ed. Cambridge University Press, Cambridge. HUBBARD, C. E. (1968). Grasses, 2nd ed. Penguin Books, London. KERGUELEN, M. (1995). In PORTAL, R. Bromus de France. R. Portal, 43750, Vals-pres-Le Puy, France. SMITH, P. M. (1972). Serology and species relationships in annual bromes. Annals of botany 36: 1-30. SMITH, P. M. (1973). Observations on some critical Brome grasses. Watsonia 9: 319-332. SMITH, P. M. (1980). Bromus L. in TUTIN, T. G. et al. eds. Flora Europeae 5. Cambridge University Press, Cambridge. SMITH, P. M. (1981). Ecotypes and subspecies in annual brome-grasses. Botanische Jahrbucher fur Systematik 102: 497-509. SMITH, P. M. & SALES, F. (1993). Bromus L. Sect. Bromus: Taxonomy and relationships of some species with small spikelets. Edinburgh journal of botany 50: 149-171. STACE, C. A. (1997). New Flora of the British Isles, 2nd ed. Cambridge University Press, Cambridge. (Accepted February 2001) -t- th Watsonia 23: 533-538 (2001) 553 Festuca longifolia Thuill. (Poaceae) in Devon R. E. N. SMITH Institute of Grassland & Environmental Research, North Wyke, Okehampton, Devon EX20 2SP* and L. J. MARGETTS 16 Ashleigh Road, Honiton, Devon EX14 ITD ABSTRACT Festuca longifolia Thuill., Blue Fescue, was first recorded in Devon on limestone at Berry Head in 1936. More recently it has been found on sea cliffs in South Devon between Start Point and Bolt Tail on both mica- schist and hornblende-chlorite schist rocks and another site on limestone near Torquay. F. longifolia is not an obligate calcifuge but occurs on dry soils in the absence of competition. It may be more widely distributed in the south-west peninsula of England. KEYWORDS: fescue, distribution, habitat, geology, soil type. INTRODUCTION There is some confusion over the name Festuca longifolia Thuill., which, especially in post-1945 Floras such as the popular Concise British Flora in Colour (Martin 1965), is used to describe an introduced species, the Hard Fescue. However, recent taxonomic research (Wilkinson & Stace 1989), has shown that the correct name for Hard Fescue is F. brevipila Tracey. Wilkinson & Stace (1991) have further shown that F. longifolia is not conspecific with the garden Blue Fescue, to which the name F. glauca Vill. correctly belongs. The British Red Data Book (Wigginton 1999) shows the distribution of Festuca longifolia Thuill. in Great Britain and the Isle of Man to be restricted to a few sites in eastern England where it occurs on acid soils in heathland communities in Suffolk, Lincolnshire and Nottinghamshire. In fact, this species was recorded in Devon as long ago as 1936 when it was collected by G. T. Fraser at Berry Head and determined by W. O. Howarth, the acknowledged Festuca expert at that time. Two years later, Fraser returned to Berry Head to show the plants to three leading botanists - G. M. Ash, J. F. G. Chapple and J. P. M. Brenan. These records were included in the Flora of Devon (Martin & Fraser 1939) and vouchers deposited in TOR, RAMM and Herb. Druce (OXF). In 1992 R. Arigho, a local botanist with a particular interest in grasses, collected a small fescue from a rock outcrop near Prawle Point, the southernmost point of Devon. This was sent to T. A. Cope who confirmed that it was F. huonii Aug. This new record for the mainland of Great Britain was published locally (Margetts 1993) and nationally (Plant Records 1995). The following year, R. Arigho and L.J.M. visited the South Hams coast in search for further colonies of F. huonii. Near Gara Rock they found a small bluish fescue on rocks by the coastal footpath. They each collected rooted offsets to grow on in pots. When these flowered, it was realised they were not F. huonii, but almost certainly F. longifolia. The specimens were sent to T. A. Cope and P. J. O. Trist, both of whom confirmed this identification. This record was also published (Margetts 1994; Plant Records 1995). Since then, further exploration of this part of the south Devon coast has led to the discovery of F. longifolia at a number of sites between Start Point and Bolt Tail, and at a new station on the “Address for correspondence: Email: roger.smith @bbsrc.ac.uk 534 R. E. N. SMITH AND L. J. MARGETTS \ : 2 a i a acre Prawle Point FIGURE 1. Distribution, by 1 km square, of Festuca longifolia (Thuill.) in Devon. The open circle represents the 1936 record from Berry Head, the closed circles the 1993 and subsequent records. Torbay limestone (Fig. 1; Appendix 1). It still exists on Berry Head (R. Arigho, pers. comm., 1995). Festuca huonii has not been relocated or found elsewhere. It is possible the original specimen may have been young F. longifolia. GEOLOGY AND SOILS Plants growing in eastern England and the Channel Islands have all been described as growing on acid soils (Wilkinson & Stace 1991). In France it has generally been regarded as a calcifuge (Portal 1999). However, the two populations found in the vicinity of Torquay occur on limestone headlands. In addition, the majority of records come from the area between Start Point and Bolt Tail where the rocks fall into two main groups, the mica-schists and hornblende-chlorite schists, which were formed by the metamorphism of mafic lavas or sills (Durrance & Laming 1982). A pulverised sample of the hornblende schist from near Prawle Point had a pH of 8-2. Consequently, it was thought desirable to collect soil samples from below plants and to determine pH in the laboratory. Testing was completed on fresh soil within 24 hours of collecting the samples from the mica-schist and hornblende-chlorite schist sites and 48 hours for the limestone site (Table 1). Standard laboratory procedures were followed. HABITAT On the south Devon coast F. longifolia occurs in sunny exposed situations, close to the edge of the sea cliffs, mostly on flat or gently sloping surfaces of large outcrops of both rock types, usually where there is some soil. It can also be found occasionally in vertical crevices. Soils on both the mica and hornblende are immature, rich in raw humus but prone to dry out. There is little, if any, root penetration of the rock surface. Where plants are established in very thin soil or growing in fissured rock, they will be small and scattered but may be the dominant species in an open community. Where the soil is a little deeper F. longifolia forms semi-continuous cover, with the associated species growing amongst it. In these circumstances plants can be moderately robust. and form tussocks 15 cm or more in diameter. However, in general, where the soil depth exceeds about 10 cm F. longifolia is excluded by more competitive species such as Festuca rubra. In these situations the occurrence of F. longifolia appears to be limited only by the size of the outcrop. Between Start Point and Bolt Tail there are many such sites. The total population almost certainly consists of thousands of individual plants. FESTUCA LONGIFOLIA IN DEVON 535 TABLE 1. THE PH OF SOIL TAKEN FROM BELOW ESTABLISHED FESTUCA LONGIFOLIA (THUILL.) PLANTS IN DEVON AT THE END OF APRIL 2000 Rock type Locality pH Mica schist Bolberry Down 4.7 Cathole Cliff all Hugh’s Hole 5-0 Whitechurch 5-3 Hornblende-chlorite schist Elanders Cove 6-1 Maceley’s Cove 5-9 Gammon Head 5:4 Hamstone Cove 5:3 Hamstone Cove 7), Limestone Fishcombe Point 7:3 Fishcombe Point 7-6 Fishcombe Point 7:7 Fishcombe Point 8-5 In addition to the cliff edge populations reviewed here, F. /ongifolia has also been found a short distance inland (R. Arigho, pers. comm., 1995) where rocky outcrops of both the mica and hornblende schists occur a field or so behind the shore. It is known to occur here in somewhat similar circumstances but the more maritime associates are, presumably, absent. In the Torbay area F. longifolia grows at Fishcombe Point on sheltered south facing slopes in thin soil amongst small limestone outcrops where small groups of plants occur in open communities. The soils here are less organic and very freely draining. It probably occurs in similar situations on Berry Head. In Devon, 47 species were found growing in association with F. longifolia (Appendix 2). Only Dactylis glomerata was more-or-less constant, as a dwarfed form, whilst Sedum anglicum was found on all three rock types. Armeria maritima was constant on the schists but absent on the limestone. The other associated species were either annuals, bulbs or dwarf forms of, presumably, drought-tolerant species. DISCUSSION It is, perhaps, not very surprising that F. /ongifolia has been found on the coast of south Devon. In the Channel Islands, it once occurred in Jersey and may still be found on Guernsey, Sark and Herm where it occurs on coastal cliffs in shallow organic soils over granite or in cracks and crevices in the rock (Trist 1996). Populations here are small, no more than 5-10 individuals at irregular intervals, where the few constant associates include Armeria maritima, Dactylis glomerata, Sedum anglicum, Trifolium campestre and Vulpia bromoides. F. longifolia is also known from the western coast of Normandy, France (Portal 1999) where it occurs between Avranches and Flamanville. These populations, described by Huon (1970) using the name Festuca glauca Lam., occur on coastal schistose cliffs, fixed dunes and, in one locality, on carboniferous limestone a short distance inland. The pH of these three types of substrate was 5-4, 8-0 and 7-6 respectively. Huon came to the conclusion, that in Normandy, F. longifolia was indifferent to the nature of the substrate and simply occupied xerophyllous habitats. For many years, the Berry Head record has been ignored and the plant assumed to be restricted to acid soils. However, this record, together with that from Fishcombe Point, both on limestone, indicate that F. longifolia is not an obligate calcifuge but occurs on soils with a very wide pH range where taller growing species are excluded by environmental stress. This view is supported by the evidence from Normandy. Given this tolerance of substrate, there is a great deal of apparently suitable habitat on the coast of the south-west peninsula where this species could occur and it seems likely that it may be more widely distributed. 536 R. E. N. SMITH AND L. J. MARGETTS ACKNOWLEDGMENTS We wish to thank R. Arigho for stimulating our interest in this species, Derek Wells for making the manuscript and notes of the late P. J. O. Trist available to us, and Julia Tallowin for measuring the pH of the soil samples. The distribution map was prepared using DMAP software produced by Dr. Alan Morton. REFERENCES DURRANCE, E. M. & LAMING, D. J. C. (1982). The geology of Devon. University of Exeter. Huon, A. (1970). Les Fetuques de l’ouest de la France - recherches de biosystematique et de biogeographie. Thése Rennes. MARGETTS, L. J. (1993). 85th report on botany, vascular plants. Report and transactions of the Devonshire Association for the advancement of science, literature and the arts 125: 227-235. MARGETTS, L. J. (1994). 86th report on botany, vascular plants. Report and transactions of the Devonshire Association for the advancement of science, literature and the arts 126: 251-257. MARTIN, W. K. (1965). The concise British Flora in colour. Ebury Press & Michael Joseph, London. MARTIN, W. K. & FRASER, G. T. (1939). Flora of Devon, T. Buncle & Co., Arbroath. PLANT RECORDS (1995). Watsonia 20: 419-437. PORTAL, R. (1999). Festuca de France, R. Portal, Vals-prés-le Puy. STACE, C. A. (1997). New Flora of the British Isles. 2nd ed. Cambridge University Press, Cambridge. TRIST, P. J. O. (1973). Festuca glauca Lam. and its variety caesia (Sm.) Richt. Watsonia 9: 257-262. TRIST, P. J. O. (1996). Unpublished manuscript on Festuca longifolia Thuill. WIGGINTON, M. J., ed. (1999). British Red Data Books, 1 Vascular Plants, 3rd ed. Joint Nature Conservation Committee, Peterborough. WILKINSON, M. J. & STACE, C. A. (1989). The taxonomic relationship and typification of Festuca brevipila Tracy and F. lemanii Bastard (Poaceae) Watsonia 17: 289-299. WILKINSON, M. J. & STACE, C. A. (1991). A new taxonomic treatment of the Festuca ovina L. aggregate (Poaceae) in the British Isles. Botanical journal of the Linnaean Society 106: 347-397. (Accepted February 2001) FESTUCA LONGIFOLIA IN DEVON 537 APPENDIX 1 THE DISTRIBUTION OF FESTUCA LONGIFOLIA THUILL. IN DEVON 10 km square Location Grid Year Recorder Det. SX63 Bolberry Down SX687383 1994 LJM & LMS ci Cathole Cliff SX693379 2000 RENS & CJS 2 Hugh’s Hole SX679388 2000 RENS & CJS £ nr Bolt Tail SX672393 2000 RENS & CJS ay SX73 nr Sharp Tor, Salcombe SX729368 1994 LJM & LMS nr Gara Rock SX742371 1993 LIM & RA ss on Gara Rock SX751370 1994 LJM & LMS Portlemouth Down SX747369 1994 LJM & LMS Pigs Nose Valley SX762363 1994 DAB < Gammon Head SX764357 1994 DAB 7 SX83 above Lannacombe Beach SX802372 1995 RA Start Point SX826372 1993 RA SX85 Fishcombe Point SX918570 1997 RENS & CJS £ SX95 Berry Head SX9456 1936 GTF Ist record | GiE— G. T. Fraser LJM = L. J. Margetts LMS = L. M. Spalton RENS é& CJS =R.E.N. & C. J. Smith RA =R. Arigho DAB = field meeting, Botany Section, Devonshire Association o) =} feel: Oni .J -O. Trist onf. L. J. Margetts et W. O. Howarth Hh MW Ho @ @ MH —-+ xX In addition, a further seven sites in SX73, between Prawle Point and Evator Cove and based on non-flowering material, were marked as provisional by RA. These are unlikely to be any other species since the only other blue-leaved fescue in the F. ovina group is F. glauca Vill., the garden plant, which is inconceivable here. R. E.N. SMITH AND L. J. MARGETTS 538 APPENDIX 2. PLANT SPECIES ASSOCIATED WITH FESTUCA LONGIFOLIA IN DEVON ae a ene (4 ql ey c I I I I I Cc I I Cc I JUIOT IQUIODY ST] JAOD ouO]sSWUIe}]] proH dA0Z dao | YonyooyyAA = OFOH] Sy Y JUNO umog ALlaqiog uouued s Agjooeyy ssopuryly goyyea JUO}SOWI'T JSIYOS IOTYS-opus[ quiopy] ISIYS BIT pavdosna xa] DIUOPOAOIS WUNIAINA pjonidns pispnssads SNIIDAIJO SNYIUOS DAOLIUN auapls wn sUuY WNpas AOU DGAOSINSUDS DSOJAID XOWNY DIDJOAIUD] OSDIUD] Sndouos09 OSYIUP] gq AOJOISIP SIOSOKP DJIAIA DIYIUION SNAOLJIGGNS SnJOT SNIDINAUAOI SNOT S1]1JDXDS UOpoJuoaT pDUuDJUOU aUOISPL avzAKuod pINU] DIDIIPDA StAaDYIOdA TT WNADUIIYJO DJJ ASO] XNAY DAIPAH poipuppisod piqsoydny DadIUII DIIATD DJOADI SNONDG wnsnffip WNISD1ID smuuadsad syjag DUJIADUL DIAQUIAY 8 sisuadsv saunydy DIADAIUINA SIJAYUy wNIOfa]]IU DALPIYOV DUAAA DIPIIS DIdIA4IS-UOU SaployuIgvA HY DIIVD] XAADJ SNJDUD] SNIIOH DAQNA DINIS AJ DUIAO DINISAJ DIDAIUO]S SIAJIVG wnulipu wnipodvyp) Snarpapsloy SNUOAG xOJavAd DALY DajJAydoKivod DALY SIAD][IADI SISOASY Watsonia 23: 539-541 (2001) 539 Fallopia x conollyana The Railway-yard Knotweed J. P. BAILEY Biology Department, University of Leicester, United Kingdom, LEI 7RH ABSTRACT The hybrid between Fallopia japonica and F. baldschuanica has been known to exist in Britain as seed since 1983, and as an established plant since 1986. Over the last 15 years there have been a number of reports of this hybrid from continental Europe, both as open-pollinated seed on F. japonica plants and as established plants. In the light of the wide occurrence of the hybrid and the absence of a hybrid binomial, it is accordingly named Fallopia x conollyana. The hybrid and its current distribution are described and its significance and potential for further distribution are discussed briefly. KEYWORDS: Fallopia japonica, Fallopia baldschuanica, wide-hybridisation, invasive plants, Japanese Knotweed, Russian Vine. INTRODUCTION The hybrid between F. japonica (Houtt.) Ronse Decraene (Japanese Knotweed) and F. baldschuanica (Regel) Holub (Russian Vine) was first discovered in Wales in 1983, as open pollinated seed from F. japonica, and subsequently confirmed by artificial hybridisation the same year (Bailey & Conolly 1984). Despite the almost ubiquitous occurrence of seed pollinated by F. baldschuanica on F. japonica plants, it was not until 1987 (Bailey 1988) that a plant of some considerable size and age was discovered by David Bevan in Haringey, Middlesex v.c. 21 (Bailey 1992). Since that date there have not been any confirmed sightings of further hybrids of this constitution as established plants in the British Isles. However, increasing reports of this hybrid in continental Europe convince me that this taxon should be given a hybrid binomial name. Accordingly I am naming it Fallopia x conollyana in recognition of the decades of research into alien Polygonaceae carried out by my close colleague Ann Conolly at Leicester. Fallopia x conollyana J.P. Bailey, hybrida nova Hybrida sterilis inter Fallopia japonica (Houtt.) Ronse Decr. et F. baldschuanica (Regel) Holub. A Fallopia japonica caulibus basi lignosis minus quam 1 cm diametris tepalis per anthesin validius carinatis differt. A F. baldschuanica planta rhizomatosa, caulibus cavis, foliis basi subcordatis vel truncatis, stigmatibus fimbriatis differt. HOLOTYPUS: Overgrown railway marshalling yard at Haringey (Railway Fields), Middlesex, v.c. 21, map reference TQ317882. 2 October 1987 J. P. Bailey & D. Bevan. (LTR) Herbaceous perennial, younger plants weakly rhizomatous, but stout woody rhizomes in long- established plants; stems erect but bowing over, hollow, up to 2 m long, slender with red blotches, less than lcm diameter, becoming woody at the base; leaves glabrous with slender petioles 1-5—0-5 cm; lamina up to 13 x 6-5 cm, triangular ovate, acuminate to acuminate cuspidate at apex, sub- cordate to truncate at base; inflorescence axillary and terminal panicles; flowers superficially similar to, but larger (up to 3 mm diameter) and more conspicuous than those of F. japonica, the 3 keeled petals much more conspicuously winged than those of F. japonica; style trifid with 3 club shaped stigmata with short fimbriae (resembling more closely the stigmas of the F. japonica parent). Flowering occurs from September to early October. In spite of the woody lignified stem with buds, the plant is herbaceous, overground growth dying back to ground level or near ground level in winter, to be replaced by new canes in the spring. 2n=54 540 J; PALEY Identification of this hybrid is problematic, particularly if only dried material is available. Neither of its parents possesses distinctive epidermal trichomes and unfortunately the leaf shape and overall growth-form are very similar to post-herbicide treatment regrowth of F. japonica. The flowers are diagnostic, but even in well-established plants flowers are not often produced. It is possible to distinguish young plants of the hybrid from weed-killed F. japonica, if the whole plant complete with root system is collected, since young hybrid plants will not yet have developed rhizomes. Chromosome counts are the most reliable means of identification, but are obviously of limited utility. F. x conollyana can be raised readily from seed, which germinates freely under greenhouse conditions. However as the plants do not compete well with surrounding vegetation, and tend to get blown down if not staked, flowering is very rare. Plants have occasionally flowered at Leicester, but most of the information on flower structure comes from the plant at Haringey. Meiotic behaviour is reported in Bailey & Stace (1992). DISTRIBUTION In 1994, Petr Horn informed me that he suspected that the plants that he had grown from seed collected from open-pollinated F. japonica var. japonica plants in Ceske Budejovice (Czech Republic) were hybrids with F. baldschuanica. Subsequent chromosome counts confirmed this as the first occurrence of the hybrid outside Britain. Alberternst (1995) reported that seed collected from a Japanese Knotweed plant growing near a plant of F. baldschuanica in Germany produced plants as a result of pollination by F. baldschuanica. In 1995 I was sent a herbarium specimen of a whole plant, including the root system, of F. x conollyana collected from an old railway yard in Oldenberg, Germany by Bernd Hayen. He reported that he thought the plants were about six to eight years old, certainly not much older since the railway yard had only been closed for 10 years at that time. In 2000 Lajos Balogh sent me material from plants growing at the MTA Botanical garden at Vacratét, Hungary. It is not known whether the plants originated spontaneously or were deliberately planted, but they were quite well established. Fremstad & Elven (1997) reported the hybrid growing in Stavanger, Norway. TAXONOMY Holub (1992) published the hybrid genus name x Reyllopia (Reynoutria x Fallopia) based on my earlier report of the hybrid (Bailey 1988). Ironically, the existence of this hybrid was the final piece of evidence needed to convince me that the genera Fallopia and Reynoutria should be amalgamated; it follows that a hybrid generic name is no longer required. DISCUSSION F. x conollyana is a most unlikely hybrid for a number of reasons. Firstly, it is a cross between an octoploid with a base number of 11 and a diploid with a base number of 10. Secondly its two parents have contrasting growth-forms and over-wintering strategies, F. japonica is a herbaceous perennial which stores reserves in underground stems or rhizomes, whilst F. baldschuanica is a deciduous woody perennial that stores reserves in woody, overground stems. Hence F. x conollyana may fall between two stools in terms of resource partitioning for over-wintering, and this may contribute to its failure to establish more generally in the British Isles. Finally, the parental species are not generally sympatric in their native distribution. Based on the somewhat crude measure of herbarium specimen labels, there does appear to be some slight geographical overlap of the two species in northern China. Although vast amounts of seed are produced every year by F. japonica throughout Europe bearing F. x conollyana embryos, only a minute proportion of them seem ever to have germinated or become established in nature. I suspect that for this to occur, a long warm autumn (for maximum seed ripening), followed by a cold winter (to stop the seed from rotting), followed by a warm winter (for seedlings to survive the all-important first winter) are required - a rare but admittedly not impossible combination! FALLOPIA x CONOLLYANA 541 ACKNOWLEDGMENTS I am very grateful to Clive Stace for preparing the Latin diagnosis and for reading through the paper, and to Lajos Balogh and Bernd Hayen for sending me material of F. x conollyana. REFERENCES ALBERTERNST, B. (1995). Handbuch Wasser 2: Kontrolle des Japan-Knéterichs an Fliessgewdssern: II. Untersuchungen zur Biologie und Okologie der neophytischen Knoteriche-Arten. Landesanstalt fiir Umweltschutz Baden-Wiirttemberg. BAILEY J. P. (1988). Putative Reynoutria japonica Houtt. x Fallopia baldschuanica (Regel) Holub hybrids discovered in Britain. Watsonia 17: 163-164. BAILEY, J. P. (1992). The Haringey Knotweed. Urban Nature Magazine 1 : 50-51. BAILEY, J. P. & CONOLLY, A. P. (1984). A putative Reynoutria x Fallopia hybrid from Wales. Watsonia 15: 162-163. BAILEY, J. P. & STACE, C. A. (1992). Chromosome number, morphology, pairing, and DNA values of species and hybrids in the genus Fallopia (Polygonaceae). Plant Systematics and Evolution 180: 29-52. FREMSTAD, E. & ELVEN, R. (1997). Fremmede planter 1 Norge. De store Fallopia-artene. Blyttia 55: 3-14. HOLUvB, J. (1998). Reclassifications and new names in vascular plants 1. Preslia 70: 97-122. (Accepted May 2001) - Watsonia 23: 543-551 (2001) 543 Notes THE INTERSPECIFIC HYBRIDS OF THE RHIZOMATOUS COUCHES, ELYTRIGIA DESV. (POACEAE) Until about 1980 the rhizomatous couches were usually placed in the genus Agropyron Gaertn. There are traditionally three species recognised in Britain: A. repens (L.) P. Beauv.; A. pycnanthum (Godr.) Godr. (A. pungens auct., non (Pers.) Roem. & Schult.); A. junceum (L.) P. Beauv. (of which the British representative is ssp. boreoatlanticum Simonet & Guin., sometimes recognised as a separate species, A. junceiforme (A. & D. Love) AD: Love). Following Melderis (1978) and Melderis & McClintock (1983) all these taxa were segregated into the genus Elymus L. and the necessary new combinations were made: E. repens (L.) Gould; E. pycnanthus (Godr.) Melderis (FE. pungens auct., non (Pers.) Melderis) (later corrected to E. athericus (Link) Kerguélen by Kerguélen (1983)); E. farctus (Viv.) Runemark ex Melderis ssp. boreoatlanticus (Simonet & Guin.) Melderis. Plant breeders and other students of wheat and other Triticeae (e.g. Love 1984) have for long further subdivided Elymus, particularly into the non-rhizomatous species (e.g. E. caninus (L.) L.) and the rhizomatous species (e.g. the above three). This view is now the most usually accepted one, and for this reason it was adopted by Stace (1991) and Kent (1992). The rhizomatous species were separated into the genus Elytrigia Desv.: E. repens (L.) Desv. ex Nevsk1; E. atherica (Link) Kerguélen; E. juncea (L.) Nevski ssp. boreoatlantica (Simonet & Guin.) Hyl. Kerguélen (1987) actually cited the second species as E. atherica (Link) Kerguélen ex Carreras Mart. The publication by Carreras Martinez (1986) quoted by Kerguélen was a thesis at the University of Murcia, Spain. I have not seen a copy of this, and I have not traced one in this country (not in BM or K), but I have been kindly informed by Valéry Malecot of Paris (who has also not seen a copy) that his enquiries in Spain have shown that the thesis was not properly published and was not available for sale or on request, so cannot have been a vehicle for the valid publication of new names. It is quite likely that Kerguélen (deceased 1999) realised this, because in his standard French checklist (Kerguélen 1993) he dropped the ““ex Carreras Mart.” part of the author citation. Since Kerguélen (1987) quoted the full place of publication of the basionym, the combination E. atherica (Link) Kerguélen can be dated from his 1987 publication. The same applies to one other species (E. campestris (Godr. & Gren.) Kerguélen) and one subspecies (E. elongata ssp. scirpea (C. Presl) Kerguélen), which Kerguélen (1987) had similarly ascribed to Carreras Martinez. Kerguélen’s checklist is now available on-line at www.dijon.inra.fr/malherbo/ fdf/accueill.htm, providing a valuable resource (containing over 74000 names) which is being maintained by Valéry Malecot. Strangely, although this was not made available until 1998 (fide Malecot), it reinstates the Carreras Martinez ascriptions (possibly Kerguélen neglected to update the electronic version of his work). THE HYBRIDS Hybrids between the three British species also occur in all three possible combinations. The nomenclatural situation regarding these is best discussed separately. Elytrigia juncea x E. repens The epithet usually used for this hybrid is /axa, originally as Triticum laxum Fr., and the correct combinations under the three genera are given below in the list of citations. Under Elytrigia Kerguélen (1987) correctly gave the citation E. x laxa (Fr.) Kerguélen. There appear to be no 544 NOTES Watsonia 23 (2001) nothosubspecific names to differentiate the hybrids involving the two subspecies of E. juncea. The epithet /axa (type from Sweden) actually applies to our (the North Atlantic) subspecies boreoatlantica, so a new nothosubspecific epithet would be needed for the hybrid involving the Mediterranean subspecies juncea. There is also another (later) epithet used for this species in the earlier British literature. Druce (1907, p. 252) coined the name Agropyron x hackelii Druce; in that publication there is no diagnosis, but the name was validated in Druce (1929b, p. 877) by reference to an even earlier note (Druce 1905, p. 139-140), based on specimens labelled “A. junceum x repens” from Port Talbot Docks, Glamorgan, collected by H. J. Riddelsdell in July 1904, and from Arbroath, Angus, collected by W. A. Shoolbred in the same month. I choose the former as lectotype, represented by a specimen in OXF, with the entry from the Report of the Distributor pasted on the sheet. Riddelsdell commented: “I took the plant for this hybrid at first sight, and Mr Druce kindly removed any doubts I had.” There follows a note to say that Hackel confirmed the determination. I fully agree with this, despite the fact that in 1967 A. Melderis annotated the sheet “Agropyron pungens x repens’. The lower spikelets are very remote and the leaves have many prickle-hairs on the ribs on their upper surface, clearly showing the A. junceum influence. Despite the fact that Melderis had identified this specimen (wrongly) as A. pungens x A. repens, Melderis & McClintock (1983) listed A. x hackelii (wrongly) as a synonym of Elymus x obtusiusculus. However, there is, strangely, another specimen in OXF collected from Port Talbot, Glamorgan, again in July 1904, in this case by Druce. It is labelled on one of Druce’s own “Herbarium Britannicum” labels: “Agropyron hackelti Druce Port Talbot Glamorgan July 1904 G. C. Druce”. There is, however, not the slightest sign of any Elytrigia juncea characters in the specimen: the spike bears closely overlapping spikelets and the leaves have almost smooth ribs on their upper surface. The specimen was labelled “Agropyron pungens x repens” by A. Melderis in 1967, and I fully agree: the anthers are indehiscent and there are a few small marginal cilia on the leaf-sheath margins. Both hybrids were collected at Port Talbot in the same month, but the fact that the only reference cited by Druce (1929b) that contained a diagnosis was the one relating to the Riddelsdell plant fixes the name hackelii to the latter specimen, and places A. x hackelii within the synonymy of E. x laxa. There are two other Druce specimens in OXF labelled A. x hackelii. One, from Sker, Glamorgan (July 1904) is labelled A. pungens x A. repens by A. Melderis, and I agree with that. The other, from Yarmouth, E. Norfolk (July 1904) is labelled A. pungens by A. Melderis; it has dehisced anthers and again I agree with Melderis. Elytrigia atherica x E. juncea The relevant epithet is acuta, originally as Triticum acutum DC., and again the three correct combinations are listed below. The type of Triticum acutum is from the French Mediterranean. Kerguélen (1987) cited this taxon as E. x acuta (DC.) Kerguélen ex Carreras Mart, but this is incorrect on two counts: the attribution to Carreras Mart. is erroneous as noted above; and the combination E. x acuta had been made earlier by Tzvelev. In this case the nothosubspecies involving E. juncea ssp. boreoatlantica has the epithet obtusiuscula, originally as Agropyron obtusiusculum Lange, and as a subspecies under Elytrigia x acuta as nothossp. obtusiuscula (Lange) Kerguélen. Elytrigia atherica x E. repens There has been extraordinary confusion regarding the name of this hybrid. The epithet used almost universally over the past 25 years for this combination is oliveri, originally as Agropyron oliveri Druce (Kerguélen 1975; Melderis 1975, 1980; Melderis & McClintock 1983; Stace 1991; Kent 1992). However, in his original diagnosis of “X Agropyron oliverii’, Druce (1912, p. 38) gave the parentage of the plant (from Blakeney, E. Norfolk, August 1911) as A. pungens var. littorale x A. junceum, and the diagnosis reiterated that parentage: “It differs from pungens by the more simple [presumably meaning less congested] inflorescence and from junceum by the smaller spikelets and less brittle stem”. But soon after this (Druce 1914, p. 514) he was using the name A. x oliveri for the hybrid A. pungens x A. repens, and he remained of this opinion right up to the second edition of his British Plant List (Druce 1919, p. 410; 1927, p. 143; 1928, p. 134; 1929a, p. 768; 1929b, p. 877) and presumably until his death in 1932. Kerguélen (1987), despite his earlier (Kerguélen 1975) view on the.parentage of A. x oliveri, stated “Melderis & McClintock (1983) rangent les épithetes ‘laxa’ et ‘oliveri’ dans des taxons séparés, mais en leur donnant la méme formule hybride NOTES Watsonia 23 (2001) 545 ‘Elymus pycnanthus x E. repens’”’. This is not true! Melderis & McClintock clearly cited Elymus x laxus as E. farctus x E. repens and Elymus x oliveri as E. pycnanthus x E. repens. Kerguélen nevertheless cited Elytrigia x oliveri as a synonym of E. x laxa, under the parentage E. atherica x E. repens. Hence Kerguélen presented a novel but erroneous interpretation of the parentage of E. x laxa, while retaining Druce’s later interpretation of E. x oliveri. Kerguélen (1987) cited Druce’s plant as Elytrigia x oliveri (Druce) Kerguélen ex Carreras Mart. In this case it is not possible to correct the authority simply by dropping the “ex Carreras Mart.” part because Kerguélen cited the name in synonymy and therefore it is invalid. It was most unusual for Kerguélen to make such an error, and he corrected it in the electronic version of his Index Synonymique, where laxa is given as the epithet for E. juncea x E. repens, and oliveri for E. atherica x E. repens. (The printed version of his work (1993) did not cover these hybrids). The true parentage of A. x oliveri can of course be cleared up only by examination of the type specimen, and in Druce’s herbarium at OXF is a specimen which I deduce to be this. It is labelled “Agropyron pungens x repens Blakeney Norfolk Aug. 1911 G.C. Druce Phyt.-Geog. Excurs.” on one of Druce’s own “Herbarium Britannicum” labels. Also on the sheet is an undated label in the handwriting of the eminent grass expert W. Hackel, to whom Druce sent much material for naming: “Agropyron repens x junceum [signed] Hackel’. Thirdly, the sheet has attached an entry cut from a Report of the Botanical Exchange Club and Society of the British Isles (the date and exact origin of which I have been unable to trace, despite repeated searching through the printed Reports, and which in any case might not be relevant to this specimen), reading: “X Agropyron Hackelii Druce (= A. junceum x repens). This hybrid grass was found near Blakeney 28 in small quantity, at Southport 59, but these much nearer A. junceum, and at North Bull, Co. Dublin. At Blakeney a possible hybrid A. pungens x repens also occurred.” Taken together, these three labels tell a confusing story. It seems that Druce collected his material from Blakeney in August 1911, named it as A. pungens x A. repens, and so labelled it on his sheet. He also sent some (or the same) material to Hackel, who named it A. repens x A. junceum, But why did Druce (1912) attribute a third parentage (A. pungens x A. junceum) to his plant? He was clearly hedging his bets in the undated Botanical Exchange Club Report quoted above, claiming that although the main material at Blakeney was A. junceum x A. repens there was some A. pungens x A. repens present as well, but there is no mention on that label of the original published parentage. Subsequently (1914 onwards) he used oliveri exclusively for A. pungens x A. repens. There seem to be no sheets at OXF with the epithet ‘oliveri’ on the original label. The type specimen of A. x oliveri is very clearly a hybrid of Elytrigia juncea. There can be little doubt that, as Hackel stated, it is E. juncea x E. repens, and therefore a synonym of E. x laxa, as was listed by Kerguélen (1987) albeit for the wrong parentage! Since it is merely a synonym I have not bothered to validate the combination under Elytrigia here. As there seems to be no name available for the hybrid E. atherica (pungens) x E. repens | have decided to name it E. x drucei, based on Druce’s Port Talbot specimen, which he wrongly labelled Agropyron hackelii. POSSIBLE FUTURE COMPLICATIONS A further taxon in this group is Elytrigia repens ssp. arenosa (Spenn.) A. Léve, which was carefully described by Trist (1995). Recently Scholz (1998) has pointed out that the epithet arenosa does not apply to the Atlantic coastal plant, but to a Continental plant earlier typified by Scholz (1993) on a specimen from Mainz, Germany. The epithet applicable to our coastal plant is maritima Tzvelev. Moreover Scholz claimed that this taxon belongs better in E. campestris than in E. repens, based mainly on the prominent, close veins on the leaf adaxial surface, and he made the new combination E. campestris ssp. maritima (Tzvelev) H. Scholz. This effectively adds a new British native species. Should it be found to form hybrids with the other three British species, new hybrid names will be needed. A further complication might loom. Many Triticeae specialists, e.g. Love (1984), further split from Elytrigia the genus Thinopyrum A. Love for the species with a disarticulating rhachis, e.g. E. juncea. as T. junceum (L.) A. Love. The evidence for doing this is not unequivocal, but, should it become desirable, new combinations for the two hybrids concerned under a new nothogeneric name for Elytrigia x Thinopyrum will be required. 546 NOTES Watsonia 23 (2001) THE CORRECT HYBRID NAMES Elytrigia x laxa (Fr.) Kerguélen, Lejeunia, n.s., 120: 88 (1987) Basionym: Triticum x laxum Fr., Novit. Fl. Suec. Mant. 3: 13 (1842) Agropyron x laxum (Fr.) Tutin in A.R. Clapham, Tutin & E.F. Warb., Fl. Brit. Isl. 1463 (1952) Elymus x laxus (Fr.) Melderis & D.C. McClint., Watsonia 14: 394 (1983) Agropyron x oliveri Druce, Bot. Exch. Club Soc. Brit. Isles 3: 38 (1912) Elymus x oliveri (Druce) Melderis & D.C. McClint., Watsonia 14: 393 (1983) Elytrigia x oliveri (Druce) Kerguélen, nom. inval., Lejeunia, n.s., 120: 88 (1987) Agropyron x hackelii Druce, Bot. Soc. Exch. Club Brit. Isles 8: 877 (1929) (E. juncea x E. repens) Elytrigia x drucei Stace, hybr. nov. Hybrida inter Elytrigiam repentem et E. athericam; ab E. repenti foliorum costis adaxialibus approximatioribus et prominentioribus differt; ab E. atherica foliorum vaginis ciliis marginalibus brevioribus et sparsioribus differt; antheris indehiscentibus, polline pro parte majore sterili. Holotypus: Port Talbot, Glamorgan, v.c. 41, July 1904, G.C. Druce 28945 (OXF). (E. atherica x E. repens) Elytrigia x acuta (DC.) Tzvelev, Novosti Sist. Vyssh. Rast. 10: 32 (1973) Basionym: Triticum x acutum DC., Cat. Pl. Horti Monsp. 153 (1813) Agropyron x acutum (DC.) Roem. & Schult., Syst. Veg. 2: 751 (1817) (E. atherica x E. juncea) nothosubsp. obtusiuscula (Lange) Kerguélen, Lejeunia, n.s., 120: 86 (1987) Basionym: Agropyron x obtusiusculum Lange, Haandb. Danske FI., ed. 2, 48 (1857) Elytrigia x obtusiuscula (Lange) Hyl., Nordkdrlvaxt. Fl. 1: 369 (1953) Elymus x obtusiusculus (Lange) Melderis & D.C. McClint., Watsonia 14: 394 (1982) (E. atherica x E. juncea subsp. boreoatlantica) ACKNOWLEDGMENTS I am grateful to Dick Brummitt (K) for helping to clarify a point in the ICBN, to Valéry Malecot (P) for investigating the thesis from Murcia and for alerting me to the web site for the Index Synonymique, and to Stephen Harris (OXF) for arranging the loan of the Druce material. REFERENCES CARRERAS MARTINEZ, M. A. (1986). Estudio nomenclatural y corologico de las Gramineas del sureste de Espana. Unpublished thesis, University of Murcia. DRUCE, G. C. (1905). Report of the Distributor for 1904. Botanical Exchange Club and Society of the British Isles 2: 108-140. DRUCE, G. C. (1907). Report of the Distributor for 1906. Botanical Exchange Club and Society of the British Isles 2: 206-252. DRUCE, G. C. (1912). Plant Notes for 1911, etc. Botanical Exchange Club and Society of the British Isles 3: 7-6. DRUCE, G. C. (1914). Report of the distributor for 1913. Botanical Exchange Club and Society of the British Isles 3: 441-516. : DRUCE, G. C. (1919). New county and other records [for 1918]. Botanical Society and Exchange Club of the British Isles 5: 365-412. DRUCE, G. C. (1927). New county and other records, 1926. Botanical Society and Exchange Club of the British Isles 8: 102-144. DRUCE, G. C. (1928). British Plant List, 2nd ed. T. Buncle & Co., Arbroath. DRUCE, G. C. (1929a). New county and other records, 1928. Botanical Society and Exchange Club of the British Isles 8: 722-769. DRUCE, G. C. (1929b). Notes on the second edition of the “British Plant List’. Botanical Society and Exchange Club of the British Isles 8: 867-877. KENT, D. H. (1992). List of Vascular Plants of the British Isles. Botanical Society of the British Isles, London. NOTES Watsonia 23 (2001) 547 KERGUELEN, M. (1975). Les Gramineae (Poaceae) de la flore francaise. Essai de mise au point taxonomique et nomenclaturale. Lejeunia, n. s. 75. KERGUELEN, M. (1983). Les Graminées de France au travers de “Flora Europaea” et de la “Flore” du C.N.R.S. Lejeunia, n. s. 110. KERGUELEN, M. (1987). Données taxonomiques, nomenclaturales et chorologiques pour une révision de la flore de France. Lejeunia n. s. 120. KERGUELEN, M. (1993). Index synonymique de la flore de France. Muséum Nationale d’ Histoire Naturelle, Paris. Love, A. (1984). Conspectus of the Triticeae. Feddes Repertorium 95: 425-521. MELDERIS, A. (1975). Agropyron Gaertn., in STACE, C. A., ed. Hybridization and the flora of the British Isles, pp. 568-570. Academic Press, London. MELDERIS, A. (1978). Taxonomic notes on the tribe Triticeae (Gramineae), with special reference to the genera Elymus L. sensu lato, and Agropyron Gaertner sensu lato. Botanical Journal of the Linnean Society 76: 369-384. MELDERIS, A. (1980). Elymus L., in TUTIN, T. G. et al., eds. Flora Europaea 5: 192-198. Cambridge University Press, Cambridge. MELDERIS, A. & MCCLINTOCK, D. (1983). The genera Elymus L. and Leymus Hochst. in Britain. Watsonia 14: 391-395. SCHOLZ, H. (1993). Elytrigia arenosa (Gramineae) - ein mitteleuropdischer Relikt-Endemit. Botanische Jahrbiicher fiir Systematik, Pflanzengeschichte und Pflanzengeographie 115: 351-366. SCHOLZ, H. (1998). Taxonomy and nomenclature of “Elytrigia maritima” Tzvelev. Bulletin. Société pour l’échange des plantes vasculaires de |’ Europe et du bassin Méditerranéen 27: 102-104. STACE, C. A. (1991). New flora of the British Isles. Cambridge University Press, Cambridge. TRIST, P. J. O. (1995). Elytrigia repens (L.) Desv. ex Nevski subsp. arenosa (Spenner) A. Love (Poaceae) in north-western Europe. Watsonia 20: 385-390. C. A. STACE Department of Biology, University of Leicester, Leicester LEI 7RH SALIX CAPREA L. x S. MYRSINIFOLIA SALISB. x S. PHYLICIFOLIA L. IN PERTHSHIRE During a programme involving field surveys and a general reassessment of Scottish montane willow hybrids, a number of apparently new hybrid combinations have been discovered. One of these which was collected on 7 August 1997 by the author near Milton Roro, Glen Lyon in mid- Perths., v.c. 88, was suspected to be the triple hybrid, Salix caprea L. x S. myrsinifolia Salisb. x S. phylicifolia L., due to its appearance and also because both S. caprea and S. myrsinifolia x S. phylicifolia (S. x tetrapla Walker) were growing nearby. Further specimens were collected in July 2000 and sent to R. D. Meikle who confirmed the tentative determination. There appeared to be no previous confirmed records for this hybrid in Britain according to Meikle (1975), although E. F. Linton (1913) had earlier suggested that specimens collected by him near Clova, v.c. 90 (E.F.L. reference numbers 36 and 78) might possibly be referred to this cross. The hybrid is not shown in the British list (Kent 1992, 1997; Kent & Stace 2000). However, subsequent information has been provided by R.D.M. which shows that he determined the same cross very recently from specimens collected late in 2000 in Yorkshire, v.c. 64, by M. Wilcox (R.D.M. and M. Wilcox, pers. comm.). As there were no certain records for this hybrid combination prior to these confirmations and the fact that triple hybrids are always extremely difficult to determine, the alternative possibilities were considered, by comparing the published descriptions in Linton (1913), and those of Meikle (1975) with the specimens collected by the author in Perthshire, reference number D/T CR9/00, although the latter lacked catkins, so a full comparison was not possible. There is very little doubt that S. phylicifolia is one parent of the hybrid D/T CR9/00, therefore likely alternatives seemed to be S. cinerea L. x S. phylicifolia (S. x laurina Sm.), or S. caprea x S. phylicifolia. Although Linton (1913) suggested that some material determined as S. x /aurina might possibly refer to the triple hybrid S. cinerea x S. myrsinifolia x S. phylicifolia (which is not only possible but very probable, as S. phylicifolia may not occur in the pure state in Breadalbane) most if not all of its records have been redetermined as S. x tetrapla Walker (R.D.M., pers. comm.), although it is possible that S. x laurina could have been introduced in some areas. Similarly, it is also suggested that at least some of the records, and accordingly the descriptions, of S. caprea x S. phylicifolia might refer to the 548 NOTES Watsonia 23 (2001) triple hybrid S. caprea x S. myrsinifolia x S. phylicifolia, the same cross determined by R.D.M. for the Perthshire hybrid, D/JT CR9/00. However, Linton (1913) suggested that S. caprea x S. phylicifolia was extremely rare in Britain and had been much confused with the widely distributed S. x laurina, so it cannot be assumed that the old records for the former refer to the triple cross determined for the Perthshire hybrid D/T CR 9/00. Although a comparison of the descriptions in Linton (1913) and by Meikle (1975) for hybrids involving S. phylicifolia which are recorded in Perthshire shows that the descriptions which fit the hybrid D/T CR 9/00 best are those which were published as S. caprea x S. phylicifolia, and therefore suggest that these do not refer to S. x laurina or to the similar triple cross S. cinerea x S. myrsinifolia x S. phylicifolia, but for the reasons given above most probably refer to S. caprea x S. myrsinifolia x S. phylicifolia, and therefore provide additional evidence that the hybrid DJT CR 9/00 has been correctly determined as the latter. A description of the main characters of the hybrid DJT CR 9/00 is as follows: A bush approximately 2-5 m high. Mature twigs dark reddish-brown, glossy and becoming glabrous; shoots moderately pubescent. Leaves coriaceous, dark, bright green and glossy on the upper surface, and rather thinly pubescent with the hairs in the majority confined to the midrib and near to the apex, or subglabrous, the lower surface pale green, glaucous and glabrous with prominent nervation, the majority with the lamina obovate or broadly obovate to subrotund, the largest typically 5 cm long, 3:5 cm wide, shortly mucronate to cuspidate at apex, the margins narrowly recurved, shallowly serrate to remotely serrulate, the smaller leaves narrowly rounded, the larger broadly rounded at base; petioles moderately pubescent. Foliage not turning black when dried. CONCLUSIONS Based on the published evidence the description of the Perthshire hybrid DJT CR 9/00 best fits the descriptions given for S. caprea x S. phylicifolia, other than a slight variance in the leaf indumentum, which may not be significant. However, the records of S. phylicifolia in the area in question have been shown to refer to S. myrsinifolia x S. phylicifolia, so the descriptions of S. caprea x S. phylicifolia very probably refer to the triple hybrid S. caprea x S. myrsinifolia x S. phylicifolia, and provide additional support for the determination of the Perthshire hybrid D/JT CR 9/00 as this cross and to confirm its presence in Britain. The recent record from Yorkshire gives further confidence in its existence. It is, however, possible that this hybrid has been overlooked in the past. In the absence of S. phylicifolia, at least some of the older records for S. caprea x S. phylicifolia might have more correctly referred to the same triple cross. Ideally, it would have been advisable to locate the herbarium material which represents the old records on which the descriptions were based in order to confirm the status of the Perthshire record, however, in the absence of a prolonged and perhaps unsuccessful search it is necessary to rely on the present specimens and the original descriptions, and assume that they refer to the hybrid combinations as stated. The Perthshire hybrid has now been propagated from cuttings so further information may be obtained if catkins are eventually produced. ACKNOWLEDGMENTS I should like to thank Desmond Meikle for his confirmation of identity and other helpful comments, David Mardon for local field information, Mr. M. Wilcox for additional details of the Yorkshire record, and finally, the Science and Research Committee of the B.S.B.I. for their support and a grant towards the costs of part of this project. REFERENCES KENT, D. H. (1992). List of vascular plants of the British Isles. Botanical Society of the British Isles, London. KENT, D. H. (1997). List of vascular plants of the British Isles, Suppl. 1. Botanical Society of the British Isles, London. KENT, D. H. & STAGE, C. A. (2000). List of vascular plants of the British Isles, Suppl. 2. Botanical Society of the British Isles, London. LINTON, E. F. (1913). A monograph of the British Willows. Journal of Botany, London 51 (Suppl.): 1-92. NOTES Watsonia 23 (2001) 549 MEIKLE, R. D. (1975). Salix L., in STACE, C. A., ed. Hybridization and the flora of the British Isles, pp. 304— 338. Academic Press, London. D. J. TENNANT Low Missise Farm, Laverton, Ripon, North Yorkshire, HG4 3SY A NEW SPECIES OF RUBUS SECT. MICANTES (ROSACEAE) FORMERLY KNOWN AS A VARIANT OF R. CRINIGER (E. F. LINTON) ROGERS A bramble which is locally abundant in the Colchester area of Essex (v.c. 19) and along the Stour valley on the Suffolk side (v.c.c. 25 and 26) and is scattered in North West Suffolk (v.c. 26), West Norfolk (v.c. 28) and South Essex (v.c. 18) with one known station in East Kent (v.c. 15) has been dismissed as “var. trifolius” of Rubus criniger in recent years. Previously to this, E. S. Edees had considered it to be a form of Rubus conspersus W. C. R. Watson, though Watson himself had appended the name Rubus egregius var. effeminatus Focke, to a sheet sent him in 1923 through the Botanical Exchange Club by G. C. Brown, and collected from Stanway near Colchester. This sheet is now in BM. (D. E. Allen, pers. comm.) The present author has had ample opportunity to study the plant, with annual visits to Essex for nearly twenty years, as well as its stations in the other counties mentioned, and has long held the view that it is distinctly and constantly different from its supposed parent. During the summer of 2000, the plant was collected as widely as possible in Essex, West Suffolk and West Norfolk and comparisons made with R. criniger, also from as many sites as possible, including some from Herts. v.c. 20 and Cambs. v.c. 29, where the present plant has not been recorded, and it was discovered that the differences between the two plants, even when they are growing in close proximity, as they do at Waterhouse Plantation, Tottington, v.c. 28, were even greater than was at first thought. Neither have any plants been found which could be regarded as intermediate between the two. It would be true to say that the only similarities between the two are the overall greyish green coloration, the hairiness of the primocanes and the structure and dimensions of the stem prickles. It is therefore proposed to name the plant with three leaflets as: Rubus trinovantium A. L. Bull sp. nov. A R. crinigero his notis differt. Turiones glandibus brevistipitatis numerosis vel copiosis, in partibus turionis ejusdem diversis quoad numerum variabilibus, vestiti. Aculei aliquot in turionum paginis sicut in angulis interdum reperiantur. Folia ternata vel rarissime quatuor vel quinque foliola ferentia; foliola terminalia elliptica vel obovatocuspidata, cuspide 1-5-2 cm longa praedita. Paniculae laxae, relative pauciflorae, parte superiore subracemosa brevi truncata; pedicelli 2-4 cm longi. Adsunt 2-4 rami inferiores ascendentes ad 15 cm longi. plerumque modo 3-7 floribus instructi. Flores magni, 3-3-5 cm diametro, stellati; petala elliptica, 1-5-1-75 cm _ longa, emarginata; antherae glabrae. Rubus trinovantium differs from R. criniger in the following characters. The stems are clothed with numerous to abundant short-stalked glands, variable in number on different parts of the same stem. A few prickles may occasionally be found on the faces of the stems as well as on their angles. The leaves are ternate or very rarely bear four or five leaflets; the terminal leaflets are elliptic to obovate-cuspidate, with the cusp 1-5 to 2 cm long. The panicles are lax and relatively few-flowered, with a short truncate subracemose upper part; the pedicels are 2-4 cm long. There are 2-4 ascending lower branches up to 15 cm long and usually only 3—7 flowered. The flowers are large, 3-3-5 cm in diameter and starry the petals are elliptic, 1-5—1-75 cm long and notched, and the anthers are glabrous. Rubus criniger belongs to the Series Vestiti, but bearing in mind that the armature of R. trinovantium is somewhat variable, to the extent that shade plants occasionally have consistently longer stalked glands reminiscent of the group Radulae as at Lexden Gathering Ground near Colchester, and some prickles may be found which are not on the angles of the stem. it is felt that the new species should be placed in the Series Micantes. It has been found in 16 hectads to date. HOLOTYPE: Tiptree Heath, Essex. v.c. 19., TL883148 July 12th 2000. BM. Isotypes are in Herb. A. Newton and Herb. A. L. Bull. 550 NOTES Watsonia 23 (2001) Representative Exsiccatae from High Woods, Colchester, TL9926, July 12th 2000; Lexden Gathering Ground, TL9725, July 12th 2000; Layer Breton Heath, TL9118 July 16th 1981; East Donyland, Colchester, TM02 July 23rd 1972, all in Essex; Denstead Wood, East Kent, TROS July 19th 1990; Arger Fen, West Suffolk, TL9335 July 10th 2000; Assington Thicks, West Suffolk TL9337, July 10th 2000; Letch Moor, Icklingham, West Suffolk, TL7971, Aug. 20th 1971 and July 17th 2000; South Runcton, West Norfolk TF6407, July 23rd 1977 and Waterhouse Plantation, Tottington, West Norfolk, TL9094, July 15th 2000 are all in Herb. A. L. Bull. R. trinovantium is a bramble of acid sands and gravels which characterise the formerly extensive heathland that stretched from Tiptree to both north and south of Colchester. The West Suffolk site at Arger Fen is on the edge of the former Leaven Heath, whilst nearby Assington Thicks is ancient woodland with some patches of acid soils. Letch Moor at Icklingham and Waterhouse Plantation, Tottington, are both areas of overgrown wet acid heathland and South Runcton lies on the Norfolk Greensand. The name trinovantium derives from Trinovantes, the ancient British tribe whose territory centred on the Colchester area at the time of the Roman invasion. My thanks are due to Mr Philip Oswald for writing the Latin description, to Dr D. E. Allen for tracing specimens in BM and to Mr A. Newton for advice on the name trinovantium. A. L. BULL “Hillcrest”, East Tuddenham, Dereham, Norfolk, NR20 3JJ DATES OF PUBLICATION OF COUNTY FLORAS It seems to have been generally overlooked that, for various reasons, county Floras are liable to bear an erroneous date on the title page. This is a matter for concern not merely bibliographically, for in some of these publications new taxa have been described or valid new combinations perpetrated unwittingly. The increasing attention being given to establishing the history of rare species at individual sites also makes precision in published dates of records a matter of importance. The usual reason for misdating would seem to have been the tendency for small printers, in the days before computerised typesetting, to treat such major and often typographically complex jobs as ‘fillers’. to be worked on whenever business was slack and to be laid aside for extended periods when more urgent or more profitable commitments intervened. In a publication process so leisurely and protracted, including the dispatch of bound copies maybe some considerable time after the receipt of the final corrected proofs, it could easily happen that an obsolete scheduled date was left unamended. A particularly glaring instance where this is presumed to have been the cause is the first edition of F. Townsend’s Flora of Hampshire, including the Isle of Wight. Though allegedly published in 1883, an addendum was inserted (just before the index) at a clearly very late stage and the dates of some of the records in that show that it cannot have appeared before 1884 - assuming that all the bound copies initially issued included it. There is also reason to suspect that the second edition of that work came out in 1905 instead of, as stated, 1904 (Allen 1986). Two more recent cases have been pointed out by Mitchell (2000). J. P. Brunker’s Flora of the county Wicklow, though bearing the date *1950’, actually appeared in the year following, as mentioned at the time by Praeger (1951) and since confirmed by the publisher’s records. J. Harron’s Flora of Lough Neagh was similarly published a year later that the indicated one. Printing delays are not invariably responsible for this phenomenon, though. The Flora of the Isle of Man (Allen 1986), though ready for issue by the date on the title page, was held back by the publisher for two years in order for its appearance to coincide with ‘Manx Heritage Year’. Publication can also be earlier than the year stated. T. Whilde’s The natural history of Connemara, in which vascular plants are included in a list in the appendix, appeared in reality in 1993, not ‘1994’ (Mitchell 2000). Copies of the section on the botany contributed by T. H. Cooper to the second volume (1835) of T. W. Horsfield’s The history, antiquities and topography of the county of Sussex were distributed by Cooper as a separate pre-print a year before the publication of the book itself, as shown by one that has survived in the W. J. Hooker Letters in the archives of the Royal Botanic Gardens, Kew, accompanied by a dated covering letter. NOTES Watsonia 23 (2001) 551 REFERENCES ALLEN, D. E. (1986)(°1984’). Flora of the Isle of Man. Manx Museum & National Trust, Douglas. ALLEN, D. E. (1996). Some earlier workers on the Hampshire flora, in BREWIS, A., BOWMAN, P. & ROSE, F., eds., The flora of Hampshire. Harley Books, Colchester. MITCHELL, M. E. (2000). The Irish floras: a checklist of non-serial publications. Glasra 4: 47-57. PRAEGER, R. L. (1951). Review of J. P. Brunker, Flora of the county Wicklow. Irish Naturalist’s Journal 10: 192-195. D. E. ALLEN Lesney Cottage, Middle Road, Winchester, Hampshire, SO22 5EJ Watsonia 23: 553-592 (2001) 553 Plant Records Records for publication must be submitted to the appropriate Vice-county Recorder (see BSBI Year Book 2001), and not to the Editors. The records must normally be of species, hybrids or subspecies of native or naturalised plants (listed in Kent (1992) or Stace (1997)), belonging to one or more of the following categories: Ist or 2nd v.c. record; Ist or 2nd post-1930 v.c. record; only extant v.c. locality, or 2nd such locality; a record of an extension of range by more than 100 km. Such records will also be accepted for the major islands in v.cc. S, 102-104 and 110. Only Ist records can normally be accepted for Rubus, Hieracium Taraxacum and hybrids. Records for subdivisions of vice-counties will not be treated separately; they must therefore be records for the vice-county as a whole. Records for Taraxacum must normally be additional to those in Dandelions of Great Britain and Ireland by A. A. Dudman & A. J. Richards (1997). 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. Taxa not in that book follow New Flora of the British Isles Ed. 2 by C. A. Stace (1997). 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. Records from the following vice-counties are included in the text below (superscript figures indicate pmnbemniaac) ta 2 34. 576.8 9°, 10’, 11°, 127.13", 15°, 17 21, 23°; 25°, 26, 28", 28°, 33°, Beeman), 39°, 40°41, 42”, 437, 44° 45°, 461%, 47°, 48°, 49°, 50, 51”, 53’, 54’, 58)°, 59°, 60), 61°, fremG 687,69), 70,717, 77°, 78, 79, 80 , 81, 83°, 89, 91°, 93°, 94", 95, 97’, 99", 100’, 1017, 104°, Mise, 109, 110’, H10°, H16', H21', H27°, H30", H31°, H35°. The following signs are used: * before the record: to indicate a new vice-county record. + before the species number: to indicate that the plant is not a native species of the British Isles. + before the record: to indicate a species which, though native in some parts of the British Isles, is not so in the locality recorded. [] enclosing a previously published record: to indicate that the record should be deleted. 1/3.1. LYCOPODIUM CLAVATUM _ 28, W. Norfolk: One small plant on floor of extensive gravel pit, formerly heath, Leziate RSPB Reserve, TF682195, S. Fisher, 1997, conf. G. Beckett. 1st post- 1930 record. 1/4.1. DIPHASIASTRUM ALPINUM 80, Roxburghs.: NE facing bank, old forestry quarry, Ramsaygrain West Forest, Teviothead, NT332022, R.W.M. Corner, 1996. 2nd extant site. 1/4.2a. DIPHASIASTRUM COMPLANATUM subsp. ISSLERI *108, W. Sutherland: Stony ground, Canisp, NC2216, A.G. Kenneth, 1985, conf. A.C. Jermy. 72/1.2. SELAGINELLA KRAUSSIANA *50, Denbs.: Naturalised in mown grass by stream, Bodnant, SH799721, J.A. Green, 1997. 3/1.1. ISOETES LACUSTRIS 94, Banffs.: Sandy bay, Loch Avon, NJO25032, J. Edelsten, 1989. Not seen at this site since 1867. Pool by Loch Avon, NJO18026, D. Law, 1989. Not seen at this site since 1866. Both BM, det. A.C. Jermy. Ist and 2nd recent records. 3/1.1x2. ISOETES ECHINOSPORA x I. LACUSTRIS (I. x HICKEYI) *46, Cards.: NE corner of Llyn Pendam, SN709839, A.O. Chater & A.C. Jermy, 1997, BM, conf. D. Britton. Ist British record. 4/1.3. EQUISETUM VARIEGATUM 64, Mid-W. Yorks.: Riverside marsh, Dunsop Bridge, SD6551, P. Jepson, 1997. 2nd record. 4/1.4x8. EQUISETUM FLUVIATILE x E. PALUSTRE (E. x DYCEI) *46, Cards.: With both parents on peaty mud of summer-dry leadmine reservoir, Bog Pond, 2km NW of Ponterwyd, SN732824, A.O. Chater, 1994, NMW, conf. A.C. Jermy. Ist record for Wales. 4/1.5x8. EQUISETUM ARVENSE x E. PALUSTRE (E. x ROTHMALERI) *73, Kirkcudbrights.: Shoulder high rushy area, Southness, Gillfoot side, NX977546, O.M. Stewart, 1994, E, det. C.N. Page. Presumed Ist record. *80, Roxburghs.: Flushed bank, Crookedsike Head, Cheviot Burn, Cocklawfoot, NT871183, C. Dixon, 1997, herb C.N.P., det. C.N. Page. 4/1.6x7. EQUISETUM PRATENSE x E. SYLVATICUM (E. x MILDEANUM) *89, E. Perth: Road verge, N side of minor road 1km E of Loch of Lowes, NO064439, C.I. Pogson, 1996, det. C.N. Page. 5/1.1. OPHIOGLOSSUM VULGATUM 81, Berwicks.: Parkland, Newton Don, NT711374, M.E. Braithwaite, 1995. 2nd extant record. 554 PLANT RECORDS 5/2.1. BOTRYCHIUM LUNARIA 28, W. Norfolk: Dune slack, Wells-next-the-Sea, TF909454, E. Banham, 1996. 2nd record since 1973, a new area. 7/1.1. CRYPTOGRAMMA CRISPA = *95, Moray: Only four plants on rocky outcrops on mountain, N facing, Creag an Tarmachain, Grantown-on-Spey, NJ154312, I.P. Green, 1997. 8/1.2. HYMENOPHYLLUM WILSONII *89, E. Perth: North facing mica schist cliff, Kindrogan Crag, NO049628, H.S. McHaffie, 1985. 9/1.1. PILULARIA GLOBULIFERA 47, Monts.: Three populations on lake-edge, shallow (c.30cm) and deep (1-6m) water, Llyn Gwyddior, SH9307, R.A. Jones, E.J. Lomas & T. Teearu, 1997. Ist recent records. 51, Flints.: Mud, edge of lake, Llyn Helyg, SJ115775, W. Hay, 1993. Ist record for 20 years. 10/2.1. TRICHOMANES SPECIOSUM (gametophyte) *44, Carms.: Abundant in dry rock cavity, under overhang on NW facing slope in wood, by track along bottom of Allt Penrhyn wood, Forge- Cych, SN254413, A.O. Chater, 1994. 11/1.1x2. POLYPODIUM VULGARE x P. INTERJECTUM (P. x MANTONIAE) —_*37, Worcs.: Mossy bank, West Malvern, SO7645; North Quarry, Malvern, SO7646; both P.G. Garner, 1989, det. R.H. Roberts. Ist and 2nd records. *58, Cheshire: Wooded bank, Barnston, Wirral, SJ279837, E.F. Greenwood, 1995, LIV, det. R.H. Roberts. Mixed with P. vulgare. 15/2.5a. ASPLENIUM TRICHOMANES subsp. TRICHOMANES 50, Denbs.: Roadside wall, Capel Garmon, SH8157, J.A. Green, 1996. 2nd record. 15/2.5c. ASPLENIUM TRICHOMANES subsp. PACHYRACHIS *42, Brecs.: Limestone rocks near Pontneddfechan, SN9108, M. Porter, 1995, det. J.C. Vogel & A.C. Jermy. *58, Cheshire: Sandstone cliffs, Maiden Castle, Bickerton, SJ4952, J.P. Guest. Sandstone cliffs, Bickerton Hill, SJ5053, G.M. Kay; both 1997, det. J. Vogel. Ist and 2nd records. 15/2.5x8. ASPLENIUM TRICHOMANES x A. SEPTENTRIONALE (A. x ALTERNIFOLIUM), , 89, E. Perth: Stenton, Dunkeld, NO04, J.H. Penson, 1966, herb. J.H.P. Volcanic rocks, Craig Tronach, NO053402, J. Castle-Smith & N.W. Taylor, 1991, E. 1st and 2nd recent records. 15/2.6. ASPLENIUM VIRIDE *47, Monts.: In crevices on vertical rock face in rocky stream gorge, Lake Vyrnwy, SH996205, P. Parker, 1997, NMW, det. P.M. Benoit. +16/1.1. MATTEUCCIA STRUTHIOPTERIS 6, N. Somerset: Over 50 plants under trees in marshy area of wood on stream edge, Vobster, ST710489, P.R. Green, 1997. 17/1.1. POLYSTICHUM SETIFERUM 67, S. Northumb.: Wooded slope above stream bank, Dovecrag Burn, NT945024, G.A. & M. Swan, 1997, herb. G.A.S. 2nd confirmed record. 17/1.1x2. POLYSTICHUM SETIFERUM x P. ACULEATUM (P. x BICKNELLII) +7 Comvwall: With both parents in woodland by the River Tamar, Lamerhooe Ford, W of Latchley, $X398732, M. Atkinson, 1996, conf. A.M. Paul. *37, Worcs.: Park Wood, Malvern Hills, SO7644, R. Maskew, 1991, conf. A.M. Paul. *50, Denbs.: Bank of old railway, Pwllglas, SJ125551, M. Keene, 1995, det. A. Paul. +17/1.1xmun. POLYSTICHUM SETIFERUM x P. MUNITUM *17, Surrey: One plant of several fronds with both parents on bank of B.P., Winkworth Hill, SU996403, A.C. Leslie, 1995, herb. A. C.L. 17/3.2x3. DRYOPTERIS FILIX-MAS x D. AFFINIS (D. x COMPLEXA) *37, Worcs.: Streamside wooded bank, Southstone Rock, Teme Valley, SO7063, P.G. Garner & R. Maskew, 1989, det. A. C. Jermy,. 17/3.3a. DRYOPTERIS AFFINIS subsp. AFFINIS *S, Channel Islands: Frequent in wooded valley near stream, Dixcart Valley, Sark, WV467754, R.M. Veall, 1996, det. M. Gibby. 17/3.3b. DRYOPTERIS AFFINIS subsp. CAMBRENSIS *2, E. Cornwall: Two plants only in unshaded roadside hedge on granite, Kit Hill, near Callington, SX383714, R.J. Murphy & M.G.C. Atkinson, 1999, det. A.C. Jermy. Ist confirmed record. *11, S. Hants.: Beside main ride in wood, Crowdhill Copse, Fair Oak, SU487198, P.J. Selby, 1995, herb. R.P. Bowman, det. H. Corley. 719/1.1. AZOLLA FILICULOIDES 41, Glam.: Settling pond, Aberthaw Cement Works, ST0367, M. Hampton, 1996. Only extant locality. PLANT RECORDS 555 720/1.1. ABIES ALBA *$1, Berwicks.: Maturing saplings established at plantation edge, Woodhead, Lauder, NT531468; Plantation edge, Blinkbonny Hill, NT566395; both M.E. Braithwaite, 1996. Ist & 2nd records as self-seeded. +20/1.nor. ABIES NORDMANNIANA *46, Cards.: About 15 self-sown trees 2-4m tall in estate woodland, Llanerchaeron, SN479601, S.P. Chambers, 1996. +20/2.1. PSEUDOTSUGA MENZIESII *46, Cards.: Shingle by Afon Ystwyth 2km W of Pontrhyd-y-groes, SN720721, A.O. Chater, 1990. Estate woodland, Coed Ffynnon Caradog, Llangorwen, SN617831, A.O. Chater, 1991. Ist and 2nd records. *81, Berwicks.: Bank, A68 near Chapel-on-Leader, NT564412, M.E. Braithwaite, 1996. Ist record as self-seeded. +20/3.1. TSUGA HETEROPHYLLA *47, Monts.: Cwm wood, Aberhafesp, SO0595, IC. Trueman & M. Oliver; Hafren Forest, SN8685, A.J. Morton, both 1988. Ist and 2nd records. Regeneration observed in both sites. +20/4.1. PICEA SITCHENSIS *S, Channel Islands: One small tree (planted) about Im tall in wooded valley, Dixcart Valley, Sark, WV466754, R.M. Veall, 1997, JSY, det. R.A. Waterman. *46, Cards.: Roadside slopes by plantations, Cwm Einion, SN695942, A.O. Chater, 1983. Edge of reservoir by plantations, Llyn Pendam, SN702842, A.O. Chater, 1985. Ist and 2nd records. Widely naturalised in Cards. +20/4.2. PICEA ABIES *46, Cards.: Roadside banks and plantation, 300m NW of Panthirion, Brynafan, SN701729, A.O. Chater, 1993. Roadside slope by plantation, W of Blaenmyherin, SN785795, A.O. Chater, 1993. 1st and 2nd records. Rarely naturalised. +20/5.1x2. LARIX DECIDUA x L. KAEMPFERI (L. x MARSCHLINSID) *12, N. Hants.: Self-sown, Alice Holt Forest, Glenbervie Enclosure, SU8142, A. Brewis, 1996. *81, Berwicks.: Plantation edge, Mellerstain, NT644401, M.E. Braithwaite, 1996. 1st record as self-seeded. 720/5.2. LARIX KAEMPFERI *46, Cards.: Roadside verge and slopes by plantations by A44, W of Nantyrarian, SN709811, A.O. Chater, 1988. Rocky slopes by plantations, Coed Bwlchgwallter, Hafod, SN770717, A.O. Chater, 1993. Ist and 2nd records. Widely naturalised. *47, Monts.: Plentiful regeneration on forestry tracksides, Uwch-y-garreg, SN7793, A.J. Morton & W.A. Thompson, 1990. 720/7.2. PINUS NIGRA *46, Cards.: Edges of mixed plantation, Coed Ty’n-y-garth, Cwm Einion, SN689946, A.O. Chater, 1991. Roadside verges by plantations, 800m ENE of Bryn Llychese, SN832810, A.O. Chater, 1993. 1st and 2nd records. Widely naturalised. 720/7.5. PINUS RADIATA *35, Mons.: A line of 10 trees at field edge on B4293 for at least the last 40 years, Devauden, ST4898(Z), T.G. Evans, 1995, herb. T.G.E. +20/7.6. PINUS STROBUS *46, Cards.: In middle of pasture, below the Rookery Wood NR, Nanteos, SN616781, A.O. Chater, 1991, NMW. *50, Denbs.: Probably planted near farm, Nantglyn, SJO061, P. Williams, 1996. 720/7.mug. PINUS MUGO-_ *46, Cards.: S facing sheepwalk by side of old estate lodge, Angler’s Retreat (Plas y Mynydd), SN746923, A.O. Chater, 1992, NMW. = *81, Berwicks.: Heather bank, A68 at Soutra, NT476574, M.E. Braithwaite, 1997. lst record as self-seeded. +20/7.mur. PINUS MURICATA _ S, Channel Islands: A mature tree, edge of maritime scrub, near hotel, near Jaspellarie, overlooking Dixicart Bay, Sark, WV467252, R.M. Veall, 1983, det. R.A. Waterman. Confirmed R.A. Waterman (1997) as obviously planted but no longer present. Noted by M.H.Marsden in 1994. *20/7.wal. PINUS WALLICHIANA *35, Mons.: One old tree producing cones at woodland edge at top of cliffs, Piercefield, ST5295, T.G. Evans, 1995. t20A/CRY jap. CRYPTOMERIA JAPONICA *46, Cards.: Open mixed plantation, Cockshead Wood, 3km NE of Llangybi, SN630554, A.O. Chater, 1996. *49, Caerns.: Probably planted on roadside, Bettws-y-Coed, SH7853, J.A. Green, 1997. +20A/SEQ.sem. SEQUOIA SEMPERVIRENS *S, Channel Islands: In wooded valley bordering hotel garden, Dixcart Valley, Sark, WV465754, R.M. Veall, 1984, det. R.A. Waterman. Two trees obviously planted; one on edge of hotel garden subsequently destroyed by gale, survivor is inside hotel garden. *47, Monts.: A single seedling, about 3 yrs old, under the largest trees in the Redwood Grove, Leighton near Welshpool, $J2403, A. Markwick, 1987. The first seedling to have occurred naturally here and still flourishing in 1997. 556 PLANT RECORDS 421/1.1x2.1. CUPRESSUS MACROCARPA x CHAMAECYPARIS NOOTKATENSIS (x CUPRESSOCYPARIS LEYLANDII) *S, Channel Islands: Established plantation of mixed, mature trees, near L’ Ecluse, Sark, WV461766, R.M. Veall, 1997, JSY, det. R.A. Waterman, Mature trees (also at other sites on Sark), all apparently planted, up to 9m or more. +021/2.1. CHAMAECYPARIS LAWSONIANA ~— *81, Berwicks.: Plantation edge, Blinkbonny Hill, NT566395, M.E. Braithwaite, 1996. Ist record as self-seeded. *21/3.1. THUJA PLICATA *46, Cards.: Estate woodland by stream, Tyglyn, Ciliau Aeron, SN498598, A.O. Chater, 1992. Slope above railway and in adjacent woodland, Glandyfi, SN694968, A.O. Chater, 1995. Ist and 2nd records. Widely naturalised. +022/1.1. ARAUCARIA ARAUCANA *17, Surrey: Margin of raised area, sandy soil, Dorking, TQ174494, J.F. Leslie, K. Page & J.E. Smith, 1997. 23/1.1. TAXUS BACCATA _+*81, Berwicks.: Cliffs, Newton Don, NT714374, M.E. Braithwaite, 1995. Ist record as established. +26/2.3. NUPHAR ADVENA _ *81, Berwicks.: Established in artificial loch, Spottiswoode Loch, NT6049, A.G. & D.G. Long, 1964, HAMU. Originally det. as N. /utea. Confirmed by D.G. Long, 1996, E. Ist record. 28/12.1. CLEMATIS VITALBA 73, Kirkcudbrights.: Growing up 12m around a 2m diameter dead tree, near Grove Hospital, NX9278, O.M. Stewart, 1997. 2nd extant record. 28/13/7. RANUNCULUS PARVIFLORUS 12, N. Hants.: Large patch near Range Gate No. 2, Woolmer Forest, SU788327, J.S. Denton, 1994. Ist recent record, not seen since. +28/13.16. RANUNCULUS ACONITIFOLIUS flore pleno *68, Cheviot: Naturalised in wet grass by stream, Shawdon, NZ083151, G.A. Swan, 1997, herb. G.A.S. 28/13.24. RANUNCULUS PELTATUS 94, Banffs. : Ditch, Nether Dallachy, NJ352644, J. Edelsten, 1994, E, det N.T.H. Holmes. 2nd record. 28/13.25a. RANUNCULUS PENICILLATUS subsp. PENICILLATUS 50, Denbs.: River Conwy, Llanrwst, SH7962, J.A. Green, 1996. 2nd record. 28/17.4. THALICTRUM MINUS _+*46, Cards.: Roadside verge and scrub 250m W of Post Office, Penbontrhydybeddau, SN676836, A.O. Chater, 1992. Roadside verge on river bank, 50m E of Pont Tanycastell, Rhydyfelin, SN589788, A.O. Chater, 1992. 1st and 2nd records. 28/17.5. THALICTRUM ALPINUM *77, Lanarks.: Rocky outcrop and bank above Gana Burn, Gana Hill, NS9501, K. Watson, 1997. +28/4.1. ERANTHIS HYEMALIS *12, N. Hants.: Hedgebank, South Warnborough, SU720453, C.R. Hall, 1989. Planted in woodland, dominant, spreading to adjoining fields, Wyck Place, near Alton, SU759392, A. Brewis, 1996. Ist and 2nd records. +28/5.1. NIGELLA DAMASCENA —_*49, Caerns.: Waste ground near railway station, Llandudno, SH7881, W. McCarthy, 1997. +28/6.1xvar. ACONITUM NAPELLUS x A. VARIEGATUM (A. x CAMMARUM) *12,.N. Hants:: One slowly spreading clump with violet flowers, Odiham Common, SU750524, J. Moon, 1996. +28/7.1. CONSOLIDA AJACIS *46, Cards.: Disturbed slopes of grassed-over rubbish tip, Pendinas, Aberystwyth, SN584798, A.O. Chater, 1991, NMW. +28/9.bla. ANEMONE BLANDA *49 Caerns.: Naturalised on roadside, Bodafon Lane, Llandudno, SH8081, W. McCarthy, 1997. *29/1.2. BERBERIS THUNBERGII *73, Kirkcudbrights.: Wood, Loch Ken side of Kenmuir Castle, NX635764, O.M. Stewart, 1995. 729/1.5. BERBERIS AGGREGATA *95, Moray: Several plants, self sown on wall an edge of wood, W of Elgin, NJ170630, P.R. Green, 1997. +29/1.9. BERBERIS DARWINII *11, S. Hants.: Scrub on heath in brambles, East Boldre, SZ366991, A.E. Bolton, 1997, det. Everton Nurseries. +30/1.1. PAPAVER PSEUDOORIENTALE *81, Berwicks.: Established on road verge, A6105 -Longformacus road end, NT772532. Known here for many years; Established on and under Town walls, Coldstream; NT842395; both M.E. & P.F. Braithwaite, 1997. Ist and 2nd records. *99, Dunbarton: Waste ground, bank near Hillfoot Station, Bearsden, NS5572, A.McG. Stirling & A. Rutherford, 1983. PLANT RECORDS 557 30/1.5b. PAPAVER DUBIUM subsp. LECOQHI *81, Berwicks.: Soil heap, housing estate, Oxton, NT498534, M.E. & P.F. Braithwaite, 1997. Well established around village; recorded here as a casual in 1960. +30/6.1. ESCHSCHOLZIA CALIFORNICA 46, Cards.: Established for several years and abundant on sandy slope above shore, Penyrergyd, Gwbert, SN164485, A.O. Chater, 1996. 2nd record. 731/1.1. DICENTRA FORMOSA *44, Carms.: Escaping along stream from garden, near Blaenwaun, SN219280, G. Hutchinson & J. Bevan, 1995, conf. A.O. Chater. 731/3.2. PSEUDOFUMARIA ALBA *44, Carms.: Gelli Aur Country Park, SN5920, D. Smith, 1988, NMW, det. D. McClintock. Garden escape. 31/5.1b. FUMARIA CAPREOLATA subsp. BABINGTONII *37, Worcs.: Old wall, Witley Court, SO7764, R. Maskew & P. Whitehead, 1991, conf. M.G. Daker. 39, Staffs.: One plant in flower and fruit in hedgebank, Workhouse Lane, Keele, SJ805452, IJ. Hopkins, 1995. Ist record since 1954. 31/5.3. FUMARIA BASTARDIT —108, W. Sutherland: Garden weed, Clachtoll, NC039280, I.M. Evans, 1997. 2nd record. 31/5.6. FUMARIA PURPUREA 51, Flints.: Hedgebank along roadside between Bodfari and Tremeirchion, SJ085717, G. Wynne, 1992, det. M. Daker. Ist record since 1942. 31/5.10. FUMARIA VAILLANTII 28, W. Norfolk: Garden weed, Fakenham, TF934298, M. Keene, 1997, conf. K. & G. Beckett. Ist record since 1958. +32/1.occxori. PLATANUS OCCIDENTALIS x P. ORIENTALIS (P. x HISPANICA) *35, Mons:.: Field/road edge, Piercefield Park, Chepstow racecourse, ST5294, T.G. Evans, 1995. 33/1.3. ULMUS MINOR = *43, Rads.: In corner of field, near Shaky Bridge, Llandrindod Wells, SO0081614, A. Orange, 1984, NWVW. +33/1.lae. ULMUS LAEVIS _ S, Channel Islands: Clump of trees near well in valley meadow, S of Le Manoir, Sark, WV465757, R.M. Veall, 1975, det. T.R.G. Rountree. One mature tree presumably planted, storm damaged 1987, dead 1990. *46, Cards.: Well established by suckers from two mature trees in mixed woodland, Y Glog, Rhydyfelin, SN593793, A.O. Chater, 1995, NMW. nos-2-L. HICUS CARICA 58, Cheshire: Canal bank, Winsford, SJ703663, J.H. Clarke, 1997. Only extant locality. 739/2.ant. NOTHOFAGUS ANTARCTICA *S, Channel Islands: Planted tree about 2-5m tall, roadside verge, Le Manoir, Sark, WV466758, R.M. Veall, 1997, JSY, det. R.A. Waterman. 39/4.2xsub. QUERCUS CERRIS x Q. SUBER (Q.x CRENATA) *35, Mons.: 2 trees in churchyard at Trevethin, SO284020, C. Titcombe, 1997. 139/4.3.+ QUERCUS ILEX 47, Monts.: One large tree in council car park, Newtown, SO1091, C.A. Small, 1980. Still there in 1997. 2nd record. 740/1.pap. BETULA PAPYRIFERA _*S, Channel Islands: Roadside verge, Aval du Creux, Sark, WV471759, R.M. Veall, 1997, JSY, det. R.A. Waterman. Planted. 40/2.1x2. ALNUS GLUTINOSA x A. INCANA (A. x HYBRIDA) *77, Lanarks.: Self sown on to a coal bing, Hamilton, NS7253, P. Macpherson, 1996, herb. P.M., det. C.S. Crook. 740/2.3. ALNUS CORDATA *46, Cards.: Marshy waste ground NW of Pen-y-bont Bridge, Llanbadarn Fawr, SN594804, A.O. Chater, 1997. 40/3.1. CARPINUS BETULUS *%+81, Berwicks.: Established on bank, Carolside, NT565397, M. E. & P.F. Braithwaite,1995. 1st record as an established alien. 43/1.10. CHENOPODIUM HYBRIDUM 3, S. Devon: Pottles Farm, Exminster, $X924873, L.M. Spalton, 1996. 2nd record. 43/1.6. CHENOPODIUM RUBRUM _ *43, Rads.: On heaps of farm manure on roadside tip, S of Upper Pits Farm, Knighton, SO315710, R.G. Woods, 1990, NMW. 43/3.2x4. ATRIPLEX PROSTRATA x A. LONGIPES (A. x GUSTAFSSONIANA) *35, Mons.: Many plants mixed with A. procumbens on upper shore and rocks of sea wall, Uskmouth, $T3382, T.G. Evans, 1996, NMW. Upper saltmarsh, Blackrock, ST5188, T.G. Evans, 1996, NMW. Ist and 2nd records. 558 PLANT RECORDS +45/2.1. CLAYTONIA PERFOLIATA 46, Cards.: Among gorse on S facing scree W of Cliff Railway, Constitution Hill, Aberystwyth, SN583826, A.D. Hale, 1996. 2nd record. 745/2.2. CLAYTONIA SIBIRICA — 41, Glam.: North bank of River Rhymney, N of Llanrumney school, ST209803, 1995, J.P. Woodman. Only extant locality. 46/5.3. STELLARIA PALLIDA *8, S. Wilts.: Windmill Hill, East Knoyle, ST873310, D. Green, 1997. Not seen subsequently. *39, Staffs.: Locally frequent on thin soil over rock ledge, Ilam, Hall Dale, SK133538, R.E. Groom, 1997, herb. B.R. Fowler. 46/5.4. STELLARIA NEGLECTA *81, Berwicks.: Wood, Coldingham Bay, NT9166, E.P. Beatie, 1956. Shoreline, Coldingham West Loch, NT893679, M.E. Braithwaite, 1996. Ist and 2nd records. 46/7.2x3. CERASTIUM ARVENSE x C. TOMENTOSUM = *37, Worcs.: With both parents in sandy heathland, Hartlebury Common, SO8170, W.A. Thompson, 1993, conf. J.J. Day & R. Maskew. 46/7.7b. CERASTIUM FONTANUM subsp. HOLOSTEOIDES *108, W. Sutherland: Shady seacliff, Fanagmore. NC177502, I.M. Evans, 1997. 46/7.11. CERASTIUM PUMILUM *31, Flints.: Open calcareous grassland on rocky, W facing slopes, Prestatyn, SJ067810, D. Guest, 1996. 46/8.1. MYOSOTON AQUATICUM *$1, Berwicks.: Sandy bank, River Tweed between Fireburnmill and the Lees, NT832389, M.E. Braithwaite, 1997, herb. M.E.B. 46/9.1. MOENCHIA ERECTA *28, W. Norfolk: Three colonies of c.12 plants in short grass on shingle, Snettisham, TF650304, S. Fisher, 1996, conf. G. Beckett. Ist record since 1959. = *41, Glam.: By paths in Agrostis heath, on S facing slopes of hill, Penmaen, Gower, $S537885, Q.O.N. Kay, 1992, NMW. 46/10.7a. SAGINA APETALA subsp. APETALA *67, S. Northumb.: Forest roadside, Black Rigg, NU753747, G.A. & M. Swan, 1967, herb. G.A.S., det. F.N. Hepper. 46/10.8. SAGINA MARITIMA _ 81, Berwicks.: Road verge, A68 at Renshawrig, NT477571, M.E. Braithwaite, 1996. Ist inland record. 46/15.1. POLYCARPON TETRAPHYLLUM *6, N. Somerset: Sparse open grassland (school playing field), Weston-super-Mare, ST320593, R. Higgins, 1997, herb. I.P. Green. 46/17.3. SPERGULARIA MARINA — +*77, Lanarks.: Verge of A721, Melbourne, NT0944; herb. M.E.B. Verge of A721, near Harpenhall, NT0443; both M.E. Braithwaite, 1996. lst and 2nd records. Abundantly established. 46/17.5. SPERGULARIA BOCCONEI 1, W. Cornwall: In sandy patch by inlet off the Camel Estuary, near Dennis Farm, S of Dinas and Padstow, SW919744, H.M. Meredith & M. Tempest, 1996, conf. R.J. Murphy. A completely new locality. +46/18.1. LYCHNIS CORONARIA *49, Caerns.: Naturalised by footpath, Wellington near Llanbedrog, SH3030, BSBI meeting, 1997. 46/20.9x10. SILENE LATIFOLIA x S. DIOICA (S. x HAMPEANA) *47, Monts.: One plant with S. latifolia, in waste ground on old lead-mine site, Fan Pool, SN9487, W.A. Thompson, 1986. 46/20.11. SILENE GALLICA *43, Rads.: One plant on recently constructed road verge, N of entrance to sewage works, Rhayader, SO978673, R.G. Woods, 1987. 46/25.5. DIANTHUS DELTOIDES +39, Staffs.: Eight patches in short sunny grassland, landfill site, Dudley Wood, near Mushroom Green, $0939863, T. Beynon, 1996, BM, det. F. Rumsey & A.M. Paul. Known here for 2 -3 years. *+73, Kirkcudbrights.: Well naturalised by ruined cottage, Craigaploch, Dundrennan, NX724444, O.M. Stewart, 1995. 46/25.7. DIANTHUS ARMERIA 41, Glam.: 98 flowering and 140 non-flowering plants in open stoney waste in old, disused, sandstone quarries, Cwm Clydach, Rhondda, $S961926, J.P. Woodman; 74 plants counted on dry bank by Baglan Reservoir, SS741930, C.R. Hipkin. Known in this area for several years. Both 1997. 2nd extant localities. t47/1.1. PERSICARIA ALPINA *77, Lanarks.: Grassy slope above burn, Cambuslang Golf Course, NS6460, K. Watson, 1987, GL. Seen well established in 1996 by P. Macpherson. +47/1.2. PERSICARIA CAMPANULATA *50, Denbs.: Wet wood near garden, Capel Garmon, SH8157, BSBI group, 1996. PLANT RECORDS 559 +47/1.5. PERSICARIA MOLLIS *11,S. Hants.: Steep E-facing grass bank overlooking W side of roadside pond, St Leonards Grange, St Leonards, Beaulieu, SZ406983, R.A. Barrett & E.J. Clement, 1995, herb. R.P. Bowman. 747/1.7. PERSICARIA AMPLEXICAULIS *46, Cards.: Grassy roadside verge, Upper Borth, SN607889, J.E. Halfhide, det. A.J. Silverside, 1977. Hedgebank, Capel Salem graveyard, Brongest, SN324450, A.O. Chater, 1997. lst and 2nd records. 47/1.16. PERSICARIA MINOR 50, Denbs.: Oxbow, Llanrwst, SH7962, R. Lewis. 1996. 2nd post-1930 record. +47/1.cap. PERSICARIA CAPITATA *49, Caerns.: Pavement cracks, Abersoch, SH3128 & Criccieth, SH501380, J. Hawksford, 1997. Ist records. 47/4.6. POLYGONUM RURIVAGUM *46, Cards.: Bean field, 150m SW of Pen, SN216521, A.O. Chater, 1989, NMW. = _+*81, Berwicks.: Established in stubble, Fireburnmill, NT8139, A. Espir, 1992, det. D.R. McKean. Established in arable field, Coldstream Hospital, NT832393, M.E. Braithwaite, 1997, herb. M.E.B., det. J.R. Akeroyd. 1st and 2nd records. 47/5.1x2. FALLOPIA JAPONICA x F. SACHALINENSIS (F. x BOHEMICA) *35, Mons.: Scattered along ditch / roadside, E bank of River Rhymney, NE of Michaelstone Bridge, ST2486, T.G. Evans, 1996, NMW, conf. A.P. Conolly. Some F. japonica present. *41, Glam.: Along about 30m stretch of roadside, N of Michaelstone Bridge, ST2485, T.G. Evans, 1996, NMW. *64, Mid-W. Yorks.: Large patch by old mill-race, Kirkstall Abbey, Leeds, SE261361, K. Pyne, 1996, LTR, det. J.P. Bailey & A.P. Conolly. *95, Moray: Two large stands on verge of B9010, growing with F. sachalinensis at roadside, Kellas, near Dallas, NJ163538, IP. Green, 1996, conf. A.P. Conolly. Now recorded from several sites. 747/5.3. FALLOPIA BALDSCHUANICA = *50, Denbs.: River bank, Bryn Rhyd-yr-arian, SH9568, J.A. Green, 1996. 747/7 rhaxpal.t RHEUM RHAPONTICUM x R. PALMATUM (R. x HYBRIDUM) *12, N. Hants.: Lay-by, roadside, near Wootton St Lawrence, SU598534, J. & P. Rollinson, 1997, conf. A. Brewis. 47/8.8. RUMEX LONGIFOLIUS *108, W. Sutherland: Marshy bank of Ledmore River, NC253116, I.M. Evans, 1997, conf. J.R. Akeroyd. 47/8.10x 14. RUMEX HYDROLAPATHUM x R. CONGLOMERATUS (R. X DIGENEUS) *47, Monts.: Rare in Montgomery Canal near Arddlin, SJ265165, R. Lansdown, 1997, RNG, conf. J.R. Akeroyd. 47/8.10x19. RUMEX HYDROLAPATHUM x R. OBTUSIFOLIUS (R. x LINGULATUS) *37, Worcs.: Old brick-pit, Grimley, SO8460, C.B. Westall, 1989, conf. J.R. Akeroyd. *47, Monts.: Rare in Guilsfield Arm of Montgomery Canal SSSI, $J239133, R. Lansdown, 1997. 47/8.13b. RUMEX CRISPUS subsp. LITTOREUS *35, Mons.: Sea wall near River Severn, Lamby, ST228777, T.G.Evans, 1995, NMW. *49, Caerns.: Sea-shore, Morfa Nefyn, SH2940, J. Hawksford, 1997. Sea-shore, Abersoch, SH3128; Waste ground by sea-shore, Pwllheli, SH3734; Beach, Criccieth, SH4937, J. Hawksford, 1997. Ist and 2nd records. *61, S.E. Yorks.: Welwick Saltmarsh, 1-6km S of Welwick, TA3318, P.J. Cook, 1997, det. J.R. Akeroyd. 47/8.13c. RUMEX CRISPUS subsp. ULIGINOSUS *46, Cards.: Edge of tidal ditch, Teifi estuary, Cardigan, SN14, A. Ley, 1885, BM, det. J.R. Akeroyd. Flood terrace, Leri estuary E of Borth church, SN616899, J.R. Akeroyd & A.O. Chater, 1988. 1st and 2nd records. 47/8.13x19. RUMEX CRISPUS x R. OBTUSIFOLIUS (R. x PRATENSIS) 50, Denbs.: Waste ground, Pontcysyllte, SJ266419, J.A. Green, 1997. 2nd record. *73, Kirkcudbrights.: Waste ground, Newton, Dumfries, NX9477, O.M. Stewart, 1996. 47/8.14x21. RUMEX CONGLOMERATUS x R. MARITIMUS (R. x KNAFII) *58, Cheshire: Muddy shore, Rostherne Mere, SJ740840, G.M. Kay, 1997, herb. G.M.K., det. J.R. Akeroyd. 47/8.15x9. RUMEX SANGUINEUS x R. OBTUSIFOLIUS (R. x DUFFTII) *58, Cheshire: Ditch, Adlington, SJ905796, J.H. Clarke, 1997, det. J.R. Akeroyd. 47/8.16. RUMEX RUPESTRIS 41, Glam.: 21 fruiting plants on largely inaccessible tufaceous cliff-ledge near Dunraven, SS8872, Q.O.N. Kay, 1996. Ist record since 1934. 560 PLANT RECORDS 47/8.19x21. RUMEX OBTUSIFOLIUS x R. MARITIMUS (R. x CALLIANTHEMUS) Sie WV OICS.: Damp laneside verge, Heightington, SO7671, W.A. Thompson & B. Westwood, 1996, det. J.R. Akeroyd. 47/8.20. RUMEX PALUSTRIS 15, E. Kent: On sandy banks and edges of newly dug pits, Dungeness, TRO618, E.G. Philp, 1997. 1st record since 1632; c.100 plants still present in 1997. 48/1.11. LIMONIUM PARVUM 45, Pembs.: Several hundred plants on blow-hole rim in limestone sea cliff, E of Mewsford Point, Castlemartin, SR9493, S.J. Leach, 1993, det. M. Ingrouille. Splash zone in gulley near top of limestone sea cliff, Stackpole NNR, SR9894, S.J. Leach, 1993, det. M. Ingrouille. 2nd British records. +49/1.off. PAEONIA OFFICINALIS *69, Westmorland: S side of fell road far from houses, Fell Yeat, Casterton, SD636795, C.E. Wild, 1990, LANC. *77, Lanarks.: Grassy bank, Shiels, Glasgow, NS5366, P. Macpherson, 1985, herb. P.M. Still there and flowering in 1989. *51/1.1. HYPERICUM CALYCINUM — 50, Denbs.: Edge of car park, Mold, SJ199627, J. Philips, 1997. Ist recent record. +51/1.3x4.¢ HYPERICUM ANDROSAEMUM x H. HIRCINUM (H. x INODORUM) *49, Caerns.: Old quarry, Llangystennin near Colwyn Bay, SH8279, G. Battershall & W. McCarthy, 1997. +51/1.4., HYPERICUM HIRCINUM *50, Denbs.: Cracks in stone bridge, Tervor, $J262415, J.A. Green, 1997. 51/1.6x7. HYPERICUM PERFORATUM x H. MACULATUM (H. x DESETANGSII) *35, Mons.: Side of forest track, Thicket Wood, Rogiet, ST4488, T.G. Evans, 1991; small wood, Cwmcarvan, SO4807, J. Harper, 1991. Ist and 2nd records. *51/1.for. HYPERICUM FORRESTIL *46, Cards.: Mixed woodland on N bank of Afon Ystwyth, Chwarel Goch, Pontrhyd-y-groes, SN725721, A.O. Chater, 1994, det. N.K.B. Robson. 52/1.2. TILIA CORDATA = 29, Cambs.: Regrowth from large old coppice stool c.300 years old, Ditton Park Wood, TL662570, S. Leatherdale, 1995, CGE, det. C.D. Pigott. Ist record in E of county of native tree. *73, Kirkcudbrights.: Wood, by Loch Roan, NX745693, O.M. Stewart, 1995. +*30, Roxburghs.: A single large roadside tree planted W of Fairnington, Roxburgh, NT643278, R.W.M. Corner, 1995, herb. R.W.M.C., det. C.D. Pigott. +52/l.euc. TILIA x EUCHLORA *29, Cambs.: In Park, by SW corner of Cherry Hill, Ely, TL540798, G. Crompton, 1993, det. C.D. Pigott. Widely planted in Cambridge; it is sterile and probably all one genetic clone. 752/1.tom. TILIA TOMENTOSA ~~ *S, Channel Islands: Two mature trees (planted), taller about 8m, roadside verge, La Vaurocque, Sark, WV462758, R.M. Veall, 1997, JSY, det. R.A. Waterman.. 753/6.1. ABUTILON THEOPHRASTI *42, Brecs.: 3 flowering plants in swede field, Ty Mawr, Llysdinam, SN9957, K. Hughson & R.G. Woods, 1996. 57/1.2. VIOLA HIRTA =.28, W. Norfolk: Chalk downland, Ringstead Downs, TF691402, T.C.E. Wells, 1997. 2nd extant locality in W. Norfolk. 57/1.4x5. VIOLA RIVINIANA x V. REICHENBACHIANA (V. x BAVARICA) *50, Denbs:.: Deciduous woodland, Ruthin, $J125554, S. Chambers, 1996, det. D. Moore. Ist record. 57/1.7. VIOLA LACTEA = 10, Wight: Clifftop heath, Bouldor Copse, $Z379903, T.D. Dines & C. D. Preston, 1996. 2nd extant locality. 57/1.9b. VIOLA PALUSTRIS subsp. JURESSI 46, Cards.: Damp rocks by stream, 900m W of Glasbwll, Llyfnant, SN729974, A.O. Chater & J.P. Woodman, 1995, NMW. 2nd record. +57/1.11x12xalt. VIOLA ALTAICA x V. LUTEA x V. TRICOLOR (V. xX WITTROCKIANA) *50, Denbs.: Weedy roadside verge, Llyn Brenig, SH983592, BSBI group, 1997. 61/1.1x2. POPULUS ALBA x P. TREMULA (P. x CANESCENS) _ S, Channel Islands: On wall W of Seineusie grounds, road to La Moinerie, Sark, WV463764, R.M. Veall, 1997, JSY, det. R.A. Waterman. Ist record as naturalised. Trees now about 3m tall and suckered from tree about 30m distant. 61/1.3a. POPULUS NIGRA subsp. BETULIFOLIA *44, Carms.: tMorfa, Llanelli, SS59, J.A.Webb, 1945, NMW, det. G. Hutchinson & I.K. Morgan; tree now probably destroyed. Four trees on edge of footpath adjacent to reed-swamp, Dyfatty Marsh, Burry Port, SN454008, G. Hutchinson, 1993, NMW, conf. R.D. Meikle. Ist and 2nd records. PLANT RECORDS 561 ¥61/1.3xdel. POPULUS NIGRA ‘PLANTIERENSIS’ x P. DELTOIDES ‘CORDATA’ (P. x CANADENSIS ‘“ROBUSTA’) *S, Channel Islands: One tree (planted) about 4m tall, roadside, foot of hedgebank, Rue de la Coupée between La Vaurocque & Dixcart Lane, Sark, WV461755, R.M. Veall, 1997, JSY, det. R.A. Waterman. +61/1.3xdel. POPULUS x CANADENSIS “‘MARILANDICA’ x P. x CANADENSIS ‘SEROTINA’ (P. x CANADENSIS ‘REGENERATA’ ) *S, Channel Islands: Three mature trees (planted), tallest now 18m, near stream in small, partly wooded, valley, Harbour Hill, Sark, WV473758, R.M. Veall, 1997, JSY, det. R.A. Waterman. Known to R.A.W. for many years. *+61/1.delx4. POPULUS DELTOIDES x P. TRICHOCARPA (P. xX GENEROSA) *S, Channel Islands: Two trees (planted), taller about 6m, roadside, foot of hedgebank, Rue de la Coupée between La Vaurocque & Dixcart Lane, Sark, WV461755, R.M. Veall, 1997, JSY, det. R.A. Waterman. 61/2.5lam. SALIX PURPUREA subsp. LAMBERTIANA _*77, Lanarks.: Bank of Carmichael Burn, Carmichael, NS9238, P. Macpherson, 1997, herb. P.M., det. R.D. Meikle. 61/2.10x11. SALIX CAPREA x S. CINEREA (S. x REICHARDTII) *47, Monts.: Canal tow-path, Red Bridge, Arddlin, S$J2514, F.H. Perring, 1975. Beside River Rhiw, NE of Manafon, SJ122028, E. Roberts & M. Wainwright, 1991, det. C.A. Sinker. Ist and 2nd records. 61/2.10x12. SALIX CAPREA x S. AURITA (S. x CAPREOLA) *44, Carms.: Edge of scrub, acid moorland valley, Glanrhyd Farm, Maudsland, SN332321, R.D. Pryce, 1992, NMW, det. R.D. Meikle. 1st confirmed record. 61/2.10x15. SALIX CAPREA x S. PHYLLICIFOLIA *73, Kirkcudbrights.: Wood, Hardgate Quarry, NX817672, O.M. Stewart, 1996, E, det. R.D. Meikle. 61/2.11bx16. SALIX CINEREA subsp. OLEIFOLIA x S. REPENS (S. x SUBSERICEA) *46, Cards.: Damp meadow, Peris Brook, Llanon, SN56, J.H. Salter, 1930, NMW, det. R.D. Meikle. 61/2.12x15. SALIX AURITA x S. PHYLICIFOLIA (S. x LUDIFICANS) *108, W. Sutherland: Edge of burn, Strath Feith an Leothaid, NC200212, P.A. Evans, 1997, conf. R.D. Meikle. 61/2.12x16. SALIX AURITA x S. REPENS (S. x AMBIGUA) 77, Lanarks.: Boggy ground, W of Longriggend, NS8070, P. Macpherson, 1997, herb. P.M., det. R.D. Meikle. 1st record since 1923. 61/2.14. SALIX MYRSINIFOLIA 81, Berwicks.: Scrub, Winding Burn, NT815665, M.E. Braithwaite, 1996. Only extant record. 61/2.14x20. SALIX MYRSINIFOLIA x S. MYRSINITES (S. x PUNCTATA) *94, Banffs.: Scree, Shelterstone Crag, NH999015, J. Edelsten, 1994, E, det. R.D. Meikle. 61/2.16. SALIX REPENS 47, Monts.: Damp lake margin with Calluna, Molinia and Sphagnum, Llyn Gwyddior, SH938074, T. Teearu, 1997, conf. R.D. Meikle. 1st recent record. 61/2.21. SALIX HERBACEA = 95, Moray: Barish areas on top of mountain, Creag an Tarmachain, Grantown-on-Spey, NJ154312, IP. Green, 1996, conf. P.R. Green. 2nd record. 61/2.9x10x11. SALIX VIMINALIS x S. CAPREA x S. CINEREA (S. x CALODENDRON) 137, Worcs.: Waste ground, Bury Hill, Oldbury, SO9789, W.A. Thompson, 1990, det. R.D. Meikle. 61/2.9x11. SALIX VIMINALIS x S. CINEREA (S. x SMITHIANA) *50, Denbs.: Streamside by road, Llangwm, SH9844, A.O. Chater, 1996. Riverbank, Llanrwst, SH7961, J.A. Green, 1996. Ist and 2nd records. 762/7.1. ERYSIMUM CHEIRANTHOIDES 81, Berwicks.: Established in turnip field, by Craighouse Quarry, NT605356, M.E. Braithwaite, 1995, herb. M.E.B. Ist record since 1931 and only extant locality. 762/10.lon. MATTHIOLA LONGIPETALA *49, Caerns.: Several plants on waste ground, Port Dinorwic, SH5267, G.& I. Battershall, 1996. 762/11.4. BARBAREA VERNA *89, E. Perth: New verge near Information Centre, Killicrankie, NN96, M. McCallum Webster, 1976, E. Shingle, Tay Marshes, Perth, NO129217, O.M. Stewart, 1982, E. Ist and 2nd records. 62/12.3. RORIPPA ISLANDICA 41, Glam.: Small population of about 5 plants in muddy disturbed ground in marshy grassland, Cwm Risca Meadow SSSI, SS8784, J.P. Woodman, 1996, conf. A.O. Chater. 2nd record. 62/12.4. RORIPPA PALUSTRIS 49, Caerns.: Shore of Coedty Reservoir above Dolgarrog, SH7566, R. Lewis, 1996. 2nd record. 562 PLANT RECORDS 62/12.5, RORIPPA SYLVESTRIS 93, N. Aberdeen: Garden of old farm, Old Meldrum, NJ823279, D. Elston, 1996, conf. D. Welch. 2nd record. 62/14.6. CARDAMINE IMPATIENS 50, Denbs.: Shady deciduous wood, Pandy, SJ1935, J.A. Green, 1996. 2nd post-1930 record. 762/16.1. AUBRIETA DELTOIDEA *50, Denbs.: Limestone rocks, Mold, SJ199627, J. Phillips, 1997. *81, Berwicks.: Established on town walls, Lauder, NT530476, M.E. Braithwaite, 1997. 762/2.1. DESCURAINIA SOPHIA *46, Cards.: Weed in garden of Tawelfan, Penrhyn-coch, SN641840, A.G. Bates, 1983, NMW, det. A.O. Chater. 62/21.4. DRABA MURALIS 8, S. Wilts.: In chippings at base of stone wall, Stratford Tony, SU091266, E.J.B. & P.M. Rollinson, 1994. 1st record for 60 years. +81, Berwicks.: Established on dry bank, Grantshouse Quarry, NT/810652, M.E. Braithwaite, 1996. 2nd record. 62/22.1. EROPHILA MAJUSCULA *3,S. Devon: A garden weed, Newton Ferrers, SX552480, R. E.N. Smith, 1997, det. T.T. Elkingion. *37, Worcs.: Little Malvern Quarry, SO7641, P.G. Garner, 1991, conf. T.T. Elkington. *42, Brecs.: Limestone outcrop, Ystradfellte, SN915134, M. Porter, 1997. 62/22.3. EROPHILA GLABRESCENS 3, S. Devon: Forword Point, SX9049, R.E.N. Smith, 1993, det. T.T. Elkington, 2nd record. *37, Worcs.: West Malvern, SO7644, P.G. Garner, 1990, conf. T.T. Elkington. 62/23.3. COCHLEARIA OFFICINALIS *+42, Brecs.: Roadside about 2:‘5km NW of Erwood, S0075444, R.G. Woods, 1994. Ist record. This record replaces that for Cochlearia danica in Watsonia 20: 423 (1995) +62/24.1. CAMELINA SATIVA 51, Flints.: “Y Fedwen Arian’, Cilcain, SJ1865, G. Harvey, 1996, det. T.C.G. Rich. Ist post-1930 record 62/28.4. THLASPI CAERULESCENS *89, E. Perth: Four plants only on steep, scree-strewn, craggy S-facing slope, across Lochsie Burn from Dalmunzie Hotel, Glen Taitneach, NO094714, D. Lloyd-Thomas & M. Sutton, 1992. 62/30.5. LEPIDIUM RUDERALE 46, Cards.: Minute pavement flowerbed, Llanrhystud, SN538697, A.O. Chater, 1991, herb. A.O.C. Ist record since 1941. 62/33.1. DIPLOTAXIS TENUIFOLIA *46, Cards.: Waste ground 500m SE of Aberystwyth Station, SN588811, A.O. Chater, 1997. *91, Kincardines.: Disturbed ground at abandoned railway yard, Crathes, NO7496, D. Welch, 1996, ABD. +62/34.4. BRASSICA JUNCEA *51, Flints.: Garden weed, “Gwylfa’, Licswm, SJ1771, G. Wynne, 1996, det. T.C.G. Rich. Possibly from bird-seed. +62/38.1. HIRSCHFELDIA INCANA 42, Brecs.: Road verge, Talgarth, SO153337, M. Porter, 1997. 2nd record. 62/41.1. CRAMBE MARITIMA *91, Kincardines.: Coastal cliffs, Stonehaven, NO887874, V. McAdam & U. Urquhart, 1988. 62/42.1b. RAPHANUS RAPHANISTRUM subsp. MARITIMUS 46, Cards.: Rank vegetation on edge of Ystwyth estuary, Aberystwyth, SN579806, S.P. Chambers & P. Amis, 1996, NMW, conf. A.O. Chater. 2nd record. +62/42.sat. RAPHANUS SATIVUS —_*S, Channel Islands: In field near derelict vegetable garden, La Rondellerie, Sark, WV460763, M.H. Marsden, 1983, JSY. 762/ERU.ves. ERUCA VESICARIA subsp. SATIVA *44, Carms.: Garden, below bird table, Llwynhendy, SS537993, I.K. Morgan, 1993. 51, Flints.: Disturbed ground, New Brighton, near Mold, $J2564, J. Phillips, 1996. 2nd record. 63/1.3. RESEDA LUTEA *91, Kincardines.: Roadside, Crathes, NO7496, D Welch, 1996. +65/12.arb. ERICA ARBOREA *17, Surrey: One fair sized shrub in hedgerow by footpath, Brook, SU931380, J.F. Leslie, K. Page & J.E. Smith, 1992. Parent visible in garden behind high wall opposite. 765/2.2. RHODODENDRON LUTEUM *47, Monts.: Steep side of Afon Hirddu and roadside, Lake Vyrnwy, SH974213, J. Clarke, 1997. | *50, Denbs.: Wet areas around ponds in woodland, Coed Cerrig-y-Wyallt, near Tal y Cafn, SH786702, R. Lewis, 1988, NMW. PLANT RECORDS 563 +65/8.1. GAULTHERIA SHALLON *49, Caerns.: Naturalised on verge, Llanystumdwy, SH4738, BSBI meeting, 1997. *91, Kincardines.: Policies of mansion, Glen Dye, NO649862, D. Welch, 1997. *65/8.2. GAULTHERIA PROCUMBENS *46, Cards.: Under oaks in estate grounds S of Trawsgoed mansion, SN671730, A.O. Chater & R.G. Woods, 1989. 66/1.3. PYROLA ROTUNDIFOLIA *50, Denbs.: Several patches over 50m in hazel-birch-ash wood on waste from limestone quarry, Minera, near Wrexham, SJ264519, M. Stapley, 1989. 769/1.1xpru.; PRIMULA VULGARIS x P. x PRUHONICENSIS *41, Glam.: Hybrid swarms in chapel graveyard with both parents, Rhyd y Fron, near Pontardawe, SN713059, R.G. & J.D. Woods, 1997. 69/2.1. HOTTONIA PALUSTRIS *71, Man: Boundary ditch on maritime heath, Ayres, NX432026, M. Devereau, 1988, L.S. Garrard, 1994. Means of arrival unknown but conceivably bird carried. Size of colony fluctuates. +69/3.1. CYCLAMEN HEDERIFOLIUM 46, Cards.: Shaded bank of Afon Einon by Ynys-hir bridge, Eglwys-fach, SN684957, E. Greenwood, & P. & W.M. Condry, 1995. 2nd record. *47, Monts.: Bank of Bachan Brook adjoining roadside, Aberbechan, SO1393, C.A. Small, 1989. Ist localised record. 69/8.1. SAMOLUS VALERANDI “*12, N. Hants.: Shallow pool, Longmoor Airstrip, SU8031, N. A. Sanderson, 1994. +71/DEU/sca. DEUTZIA SCABRA *49. Caerns.: Edge of woodland W side of B5106, Pont Dolgarrog, SH7766, R. Lewis, 1996, NMW. Ist Welsh record. 7+71/HYD.mac. HYDRANGEA MACROPHYLLA = _*17, Surrey: One shrub, pale blue, in woodland, Hindhead, SU881360, J.B.S. Hodge, 1993. Not obviously planted — possibly self-sown. +72/2.4. RIBES SANGUINEUM *50, Denbs.: Canal bank, Trevor, $J256413, J.A. Green, 1997. 773/1.1. CRASSULA TILLAEA *45, Pembs.: Abundant in compact soil amongst car park blocks, Manorbier car park, SS0697, J.W. Donovan, 1996. +73/1.3. CRASSULA HELMSI *44, Carms.: Abundant in pond, Wern Fendigaid, SN638455, I. K. Morgan, 1987. *45, Pembs.: Small patch with abundant Lemna trisulca in small pond on slope in woodland, Pwll Dyfrig, Glyn-y-mel, Lower Fishguard, SM967371, S.B. Evans et al., 1997. *67,S. Northumb.: Blackworth, NZ37, S. Lowe, 1990. East Cramlington, NZ292759, M. Teffner, 1992. Ist and 2nd records. *73, Kircudbrights.: Newly dug pools in unimproved meadow, Craig, NX681755, O.M. Stewart, 1995. *77, Lanarks.: The Moat, Golf Course, Cadder in Glasgow, NS6072, P. Macpherson, 1996, herb. P.M. +73/3/1. SEMPERVIVUM TECTORUM *50, Denbs.: Roadside wall, Llangwm, SH9745, A.O. Chater, 1996. Chapel wall, Llangwm, SH9645, S.P. Chambers, 1996. Ist and 2nd records. 73/5.18. SEDUM VILLOSUM 73, Kirkcudbrights.: By a small burn, Cairnsmore of Carsphairn, NX584980, D. Hawkes, 1997. Ist record since 1854. 773/5.4. SEDUM SPECTABILE *11, S. Hants.: Under bramble near car park along track, Keyhaven, SZ308917, P.A. Budd, 1997. +74/4.1. DARMERA PELTATA %*58, Cheshire: Stream bank, Adlington, $J912807, J.H. Clarke, 1996, conf. G.M. Kay. Wooded streambank, Disley, SJ969844, G.M. Kay, 1997. Ist and 2nd records. 774/5.2. SAXIFRAGA CYMBALARIA *46, Cards.: Wall, Alltyrodyn, Capel Dewi, SN450443, M. L. Lewis, c.1921, ABS, det. A.O. Chater & A.D.Q. Agnew. +74/5.9. SAXIFRAGA SPATHULARIS *46, Cards.: Mixed woodland, Plas Einion, Furnace, SN684948, A.O. Chater, 1991. 74/5.9x10. SAXIFRAGA SPATHULARIS x S. HIRSUTA (S. x POLITA) *93, N. Aberdeen: Woodland by lochan, Cortes, NK003593, D. Welch, 1997, herb. D.W. 74/5.19. SAXIFRAGA TRIDACTYLITES *94, Banffs.: Disused airfield, Boyndie, NJ615646, J. Edelsten, 1993, E, ABD, det. D. Parker. +74/7.1. TOLMIEA MENZIESII *67, S. Northumb.: Cross roads, Oakerland near Hexham, NY942624, R.M. Burton, 1994, herb. G.A. Swan. *68, Cheviot: On bank of streamlet in part shade, Ingram, NU018162, G.A. & M. Swan, 1995, herb. G.A.S. 564 PLANT RECORDS +74/8.1.+ TELLIMA GRANDIFLORA *31, Flints.: Probable garden escape in woodland, near houses, Wepre Country Park, Connah’s Quay, SJ291683, J. Phillips, 1997. | *67, S. Northumb.: Partly shaded on N bank of River Wansbeck, near Morpeth, NZ215865, H.E. Ellis, 1997, herb. G. A. Swan. *89, E. Perth: Kinnaird Burn Den, NN952585, R.E. Youngman, 1996. 775/2.1. PHYSOCARPUS OPULIFOLIUS *95, Moray: Several bushes in former garden, rough area on edge of forest, Darnaway Forest, near Forres, NJO06568, I.P. Green, 1996, conf. E.J. Clement. +75/3.1. SPIRAEA SALICIFOLIA *35, Mons.: Railside bank, Cadicot Pill, ST48, T.G. Evans, 1985, NMW. +75/3/1x2. SPIRAEA SALICIFOLIA x S. ALBA (S. x ROSALBA) *49, Caerns.: Hedgerow, near Ysbyty Ifan, SH8348, G. & I. Battershall, 1996. *77, Lanarks.: Near Abington, NS92, P. Macpherson, 1997, herb. P.M., det. A.J. Silverside. 175/3/2x3. SPIRAEA ALBA x S. DOUGLASII (S.x BILLARDII) *46, Cards.: Hedges by road at N end of Ponthirwaun, SN262452, A.O. Chater, 1980, det. A.J. Silverside. Widely planted and naturalised in hedges. +75/3.3a. SPIRAEA DOUGLASII subsp. DOUGLASII *44. Carms.: Railway cutting, Burry Port, SN444008, G. Hutchinson, 1988, NMW. *46, Cards.: Scrub, Ty n-y-garth, Cwm Einion, SN691945, A.O. Chater, 1991, NMW. Roadside hedges, 100m S of Pwll-y-whil, Pontgarreg, SN352540, A.O. Chater, 1992, NMW. Ist and 2nd records. *51, Flints.: Waste ground on site of former garden, Sealand, $J3568, G. Wynne, 1996, det. G. Hutchinson. *81, Berwicks.: Two large colonies established on roadside, Blyth road end, NT592489, M.E. Braithwaite, 1997, herb. M.E.B. +75/3.canxtri. SPIRAEA CANTONIENSIS x S. TRILOBATA (S. x VANHOUTTEI) *46, Cards.: Roadside hedges SW of Neuadd-lwyd chapel, 3-5km SSE of Aberaeron, SN473595, A.O. Chater, 1997, NMW. +75/4.1. ARUNCUS DIOICUS *42, Brecs.: Probably a garden outcast established on river bank, Sennybridge, SN9228, R.G. Woods, 1997. 75/6.1. FILIPENDULA VULGARIS *41, Glam.: Limestone grassland on cliff-top at Oxwich Point, far from roads and houses and apparently native, SS507848, B. Pawson, 1997. +75/8.2. RUBUS TRICOLOR *11, S. Hants.: Footpath by streambank, Beckton Bunny, SZ243935, V. Scott, 1997, herb. R.P. Bowman. 75/8.6. RUBUS BRIGGSIANUS *46, Cards.: One bush only in roadside hedgebank 250m ESE of Penwaun, St Dogmaels, SN157442, D.E. Allen & A.O. Chater,1995, BM, conf. A. Newton. 75/8.7x321. RUBUS IDAEUS x R. CAESIUS *42, Brecs.: Hedge, Tretower, SO188211, M. Porter, 1997, herb. M.P., conf. A. Newton. 175/8.9. RUBUS SPECTABILIS *73, Kirkcudbrights.: Wood edge and roadside, Damhead, Barnbarroch, NX8456, O.M. Stewart & A. White, 1996. +75/8.10. RUBUS LOGANOBACCUS *12, N. Hants.: Waste ground, Woolmer Forest, near Borden, SU798367, A. Brewis, 1997, herb. A.B. +75/8.11. RUBUS COCKBURNIANUS _ 11, S. Hants.: Grassy roadside by gate into wood, nearly opposite cottage, A35 Southampton road, Mallard Wood, Lyndhurst, SU319093, R.P. Bowman, 1997, herb. R.P.B. 2nd record. 75/8.20. RUBUS FISSUS *8, S. Wilts.: Wardour [Castle], near Tisbury, ST9226, F.A. Rogers, 1895, BM, det. W.M. Rogers, conf. D.E. Allen. [12, N. Hants.: the BM vouchers for the record in Journal of Botany 45: 72 (1907) are R. nessensis, det. D.E. Allen.] *27, E. Norfolk: North Walsham to Westwick, TG2727, E.F. Linton, 1893, BM, conf. D.E. Allen. 75/8.23 RUBUS NESSENSIS *73, Kirkcudbrights.: Roadside through Dalbeattie Forest, NX85, J.R.I. Wood, 1972, BM, comm. D.E. Allen. 75/8.27. RUBUS PLICATUS *37, Worcs.: Wooded heath, Devil’s Spittleful, Kidderminster, SO8075, A. Newton et at., 1989. | *H27, W. Mayo: Between Keel and Doogort, Achill Island, Fc.6507, J. Roffey, 1911, BM, conf. D.E. Allen. 75/8.32. RUBUS VIGOROSUS *37, Worcs.: Hartlebury Common, Stourport-on-Severn, SO8270, A. Newton, 1987. PLANT RECORDS 565 75/8.34. RUBUS ALBIONIS *35, Mons.: Wooded path edge, Mescoed Mawr, ST2789, T.G. Evans, M. Porter & R.D. Randall, 1996. +75/8.51. RUBUS ERRABUNDUS' _*54, N. Lincs.: Lane near Walesby railway crossing, TF1091, F.A. Lees, 1877, BM, det. D.E. Allen. 75/8.52. RUBUS GRATUS *64, Mid-W. Yorks.: Bank of Claude’s Clough, Brogden, SD848419, D.P. Earl, 1997. 75/8.53. RUBUS HESPERIUS *H10, N. Tipperary: Wooded bogland, near Slevoir Bay, Lough Derg, M8904, D.E. Allen, 1988, BM, conf. A. Newton. 775/8.55. RUBUS LACINIATUS *78, Peebless.: Wallace Hill Wood plantation, NT3136, A. Buckham, 1987, det. D.J. McCosh. 75/8.58. RUBUS LEPTOTHYRSOS *78, Peebless.: Quarry, near Innerleithen, NT3336, R. Learmouth, 1984, det. A. Newton. 75/8.61. RUBUS LUDENSIS *46, Cards.: Roadside hedgebank 100m S of Waunwhiod, St Dogmaels, SN145450, D.E. Allen & A.O. Chater, 1995, BM, conf. A. Newton. 75/8.62. RUBUS MACROPHYLLUS *37, Worcs.: Edge of deciduous plantation in Ockeridge Wood, SO7962; By woodland ride in Monk Wood, SO8060; both R. Maskew, A. Newton & M. Porter, 1995. 1st and 2nd records. 75/8.76. RUBUS PYRAMIDALIS *37, Worcs.: Trackside in coniferous woodland, Kyre near Tenbury Wells, SO06264, R. Maskew & A. Newton, 1992. 75/8.92. RUBUS BOUDICCAE *5, S. Somerset: Simonsbath, SS7739, E.S. Marshall, 1918 (as R. pulcherrimus), E, det. D.E. Allen. *40, Salop: Diddlebury Common, SO5085, A. Ley, 1909, BM, det. D.E. Allen. *46, Cards.: Hedgebank on sandy soil 200m N of Warren Farm, Penparc, SN201477, D.E. Allen & A.O. Chater, 1996, BM, conf. A. Newton. *48, Merioneth: Dolgellau, SH71, W.C. Barton & H.J. Riddelsdell, 1923, BM. det. D.E. Allen. *100, Clyde Is.: Lochranza, Isle of Arran, NR9350, M.L. Wedgwood, 1926, BM, E, det D.E. Allen. *101, Kintyre: Skipness Castle, NR908577, A.R. Church, 1997; sides of B842 in Saddell area, NR786314, G.H. Ballantyne, 1997, BM, conf. D.E. Allen. *H16, W. Galway: Near Aughnanure Castle, Oughterard, M1543, J. Roffey, 1911 (as R. pyramidalis?), det. D.E. Allen. *H27, W. Mayo: Beside Lough Beltra, M0898, D.E. Allen & G. Sharkey, 1992, BM. *H35, W. Donegal: Wood by roadside, Ardnamona, G963857, M. Dowlen & P. Hodson, 1993, BM, det. D.E. Allen. *H38, Co. Down: Meenan, Aghaderg parish, J0939, W.M. Rogers, 1901, BM, det. D.E. Allen. Field next to Rowallane Gardens, Saintfield, J414579, A. Newton & P. Hackney, 1985, BEL, det. D.E. Allen. Ist and 2nd records. *H40, Co. Londonderry: Hedge, station yard, Limavady Junction, C637255, P. Hackney, 1978, BEL, det. D.E. Allen. 75/8.93. RUBUS CARDIOPHYLLUS *100, Clyde Is.: One bush on rocky ground by shore, Kingscross Point, Arran, NS056282, D.E. Allen & A.R. Church 1997, BM. 75/8.102. RUBUS ELEGANTISPINOSUS *78, Peebless.: Roadside, A701 S of Broughton, NT13, G.H. Ballantyne, 1986. Possibly planted. +75/8.102. RUBUS ELEGANTISPINOSUS *9, Dorset: Wessex Ridgeway through woods, Iwerne Minster, ST8813, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.112. RUBUS NEMORALIS *78, Peebless.: Roadside, by junction of A703 and A72, Peebles, NT2540, G.H. Ballantyne, 1982. 75/8.113. RUBUS PAMPINOSUS *1, W. Cornwall: Among bracken below railway bridge, Ponsanooth, SW7737, W.C. Barton & F. Rilston, 1930, BM, det. D.E. Allen, conf. A. Newton. *46, Cards.: Colony at edge of larch plantation 100m W of Ty’n-y-cewm, Cwm Einion, SN699943, D.E. Allen & A.O. Chater, 1995, BM, conf. A. Newton. *47, Monts.: Llyfnant Valley, SN79, W.C. Barton, 1923, BM, det. D.E. Allen, conf. A. Newton, 75/8.118. RUBUS PROLONGATUS *37, Worcs.: Trackside in Wissett’s Wood near Bayton, S06772, R. Maskew & A. Newton, 1993. 75/8.122. RUBUS RUBRITINCTUS *8, S. Wilts.: East Copse, between Redlynch and Whiteparish, SU226219, Wiltshire Botanical Society, 1996, det. D.E. Allen. *64, Mid-W. Yorks.: Roadside, Gisburn Forest, SD7356, A. Newton, 1997. 566 PLANT RECORDS 75/8.125. RUBUS SUBINERMOIDES *8, S. Wilts.: East Copse, between Redlynch and Whiteparish, SU226219, Wiltshire Botanical Society, 1996, det. D.E. Allen. *37, Worcs.: Scrub on limestone grassland, Tinker’s Coppice, Horsham, $O7357, J.J. Day, R. Maskew & A. Newton, 1992. 75/8.132. RUBUS SPRENGELI] *37, Worcs.: Trench Wood, Oddingley, SO9258, A. Newton et al., 1988. 175/8.134. RUBUS ARMENIACUS “HIMALAYAN GIANT’ = *46, Cards.: Slope above road 200m S of Cardigan Bridge, SN178456, D.E. Allen & A.O. Chater, 1996, NMW. *H31, Co. Louth: Clogher Head, 01785, M. Campbell, 1991, BM, det. D.E. Allen. 75/8.140. RUBUS ROSSENSIS *8, S. Wilts.: Roadside between Redlynch and Whiteparish, SU221218, Wiltshire Botanical Society, 1996, det. D.E. Allen. 75/8.140x 142. RUBUS ROSSENSIS x R. ULMIFOLIUS *37, Worcs.: Scrub in woodland clearing, Papermill Coppice, near Alfrick, SO7451, P.G. Garner, R. Maskew & A. Newton, 1993. 75/8.141. RUBUS STENOPETALUS *9, Dorset: Two bushes on Wessex Ridgeway through woods, Iwerne Minster, ST8813, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.142x321. RUBUS ULMIFOLIUS x R. CAESIUS *37, Worcs.: Hedge, New Mill Bridge, Shelsley Beauchamp, SO7262, J.J. Day, R. Maskew & A. Newton, 1992. 75/8.143. RUBUS WINTERI *33, E. Gloucs.: Hailes to Farmcote, SP0529, M.A. Rogers, 1913, BM, det. D.E. Allen, conf. A. Newton. *35, Mons.: Alder carr, Henllys Fen, ST2692, T.G. Evans, 1996, det. A. Newton. 75/8.144. RUBUS ADSCITUS *17, Surrey: Queen’s Cottage grounds, Royal Botanic Gardens, Kew, SU1776, H.J. Riddelsdell, 1929, BM, conf. D.E. Allen. *37, Worcs.: Hagley Wood, Hayley Green, SO9381, A. Newton & M. Porter, 1993. 75/8.150. RUBUS CRINIGER *9, Dorset: Wessex Ridgeway near Iwerne Minster, ST8813, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.158. RUBUS ORBUS *9, Dorset: Conifer plantation, France Firs, Blandford Forum, ST880095, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.161. RUBUS SURREJANUS *35, Mons.: Woodland edge, N of Gray Hill, ST4293, R.D. Randall, 1993. *100, Clyde Is.: Bushy waste ground, Glen Cloy, Brodick, Arran, NSO11358, D. E. Allen, 1997, BM. 75/8.163. RUBUS VESTITUS *78, Peebless.: Roadside, E of A72, Peebles, NT2540, G.H. Ballantyne, 1988. 75/8.167. RUBUS EGREGIUS = *17, Surrey: Tooting Common, TQ2972, J. Roffey, 1910, BM, det. R.D. Randall, conf. A. Newton. 75/8.169. RUBUS FUSCICORTEX *46, Cards.: Llyfnant Valley, SN79 or SN69, W.C. Barton, 1923, BM, conf. D.E. Allen. 75/8.171. RUBUS MUCRONATIFORMIS *46, Cards.: Roadside hedgebank and field margin, Warren Farm, Penparc, SN201477, D.E. Allen & A.O. Chater, 1997, BM, conf. A. Newton. 75/8.174. RUBUS WIRRALENSIS *37, Worcs.: Scrub on site of disused army camp, Burlish Top, Kidderminster, SO8073, A. Newton et al., 1989. 75/8.176. RUBUS AEQUALIDENS *33, E. Gloucs.: Whiteshill, near Stroud, SO8407, E.M. Day, 1914, BM, det. D.E. Allen, conf. A. Newton. *34, W. Gloucs.: Berry Hill, SO5612, E.M. Day, 1913, BM, det. D.E. Allen. *46, Cards.: Track verge in conifers, Long Wood, Llangybi, SN610515, D.E. Allen & A.O. Chater, 1996, conf. A. Newton. 75/8.180. RUBUS DIVERSUS *8, S. Wilts.: Coniferised ancient woodland, Barnell Copse portion of Whiteparish Common, SU250218, Wiltshire Botanical Society, 1996, BM, det. D.E. Allen, conf. A. Newton. 75/8.184. RUBUS GLAREOSUS *35, Mons.: Wooded path edge, Mescoed Mawr, ST2789, T.G. Evans, M. Porter & R.D. Randall, 1996. 75/8.185. RUBUS GRIFFITHIANUS *H30, Co. Cavan: Lane on E side of Slieve Glah, H460002, D.E. Allen & P. Reilly, 1987, BM, conf. A. Newton. PLANT RECORDS 567 75/8.188. RUBUS HETEROBELUS *4, N. Devon: Track in forestry, Powler’s Piece near Melbary, $S365190, R.W. Gould, 1997, herb. R.W.G, det. A. Newton. 75/8.195. RUBUS MOYLEI *8, S. Wilts.: West Dean, SU257273. W.C. Barton, 1941, BM, det. D.E. Allen. 75/8.197. RUBUS NORVICENSIS *21, Middx.: Wooded enclosure, Holland park, London, TQ248800, D.E. Allen, 1997, BM, conf. A. Newton. *46, Cards.: Woodland by bridge 300m SSW of Capel Dewi church, SN451421, D.E. Allen & A.O. Chater, 1996, BM, conf. A. Newton. lst Welsh record. 75/8.200. RUBUS TRICHODES *11, S. Hants.: Gorse hedge by Bedhampton hypermarket, Havant, SU679077, D.E. Allen, 1992, BM, det. A. Newton. *23, Oxon: Chinnor Hill, SU7699, W.O. Focke & W.M. Rogers, 1894, BM, det. D.E. Allen & R.D. Randall, conf. A. Newton. 75/8.203. RUBUS ADAMSII *37, Worcs.: Trackside in woodland, Pitcher Oak Wood, Redditch, SP0267, R. Maskew, A. Newton & W.A. Thompson, 1996. 75/8.206. RUBUS ANISACANTHOS *97, Westerness: Streamside, Resipol, Sunart, NM76, S.M. Macvicar, 1897-8, BM, det. A. Newton. *H21, Co. Dublin: Sides of Bohernabreena larger reservoir, 00923, D.E. Allen, 1987, BM, conf. A. Newton. 75/8.207. RUBUS BILOENSIS *6, N. Somerset: Rough pasture near Stourton, ST7633, R.P. Murray, 1894, BM, det. A. Newton & M. Porter. 75/8.209. RUBUS DENTATIFOLIUS *37, Worcs.: Clent, SO9279, A. Newton & M. Porter, 1993. 75/8.211. RUBUS DREJERI *100, Clyde Is.: Roadside 2km N of Dougarie, Arran, NR871389, A.R. Church, 1997, BM, conf. G.H. Ballantyne. 75/8.214. RUBUS FORMIDABILIS *12, N. Hants.: Heathy wood margin, Wheatsheaf Common, Liphook, SU835305, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.216. RUBUS HIBERNICUS *9,. Dorset: Beech-—fir plantation, Okeford Hill, ST8109, D.E. Allen, 1996, BM, conf. A. Newton. 75/8.217. RUBUS INFESTUS *78, Peebless.: Lay-by on A72, Woodend, NT3038, G.H. Ballantyne, 1981. 75/8.218. RUBUS LEYANUS *37, Worcs.: Scrub in farmyard, Stoke Bliss near Tenbury Wells, S06463, R. Maskew & A. Newton, 1992. 75/8.224. RUBUS BLOXAMII *37, Worcs.: Trackside in woodland, Pepper Wood near Broadheath, SO9374, J.J. Day, R. Maskew & A. Newton, 1991. 75/8.225. RUBUS BOTRYEROS *37, Worcs.: Trackside in woodland clearing, Ramscombe Coppice, Great Witley, SO7666, J.J. Day, R. Maskew & A. Newton, 1990. 75/8.227. RUBUS CANTIANUS *17, Surrey: Bridleway through woodland, Croham Hurst, Croydon, TQ343629, D.E. Allen, 1997, BM, conf. A. Newton. 75/8.231. RUBUS ECHINATOIDES *37, Worcs.: Edge of woodland, Black Meadow, Chaddesley Woods, SO9173, J.J. Day, R. Maskew & A. Newton, 1991. 75/8.239. RUBUS INSECTIFOLIUS *37, Worcs.: Clent, SO9279, A. Newton & M. Porter, 1993. 75/8.243. RUBUS MALVERNICUS *46, Cards.: Replanted conifer plantation, Allt Maestir, 3-5km NW of Lampeter, SN546503, D.E. Allen & A.O. Chater, 1997, BM, conf. A. Newton. 75/8.245. RUBUS PALLIDUS *37, Worcs.: Edge of woodland, Black Meadow, Chaddesley Woods, SO9173; Trackside in woodland, Pepper Wood near Bournheath, $09374; both J.J. Day, R. Maskew & A. Newton, 1991. Ist and 2nd records. 75/8.249. RUBUS RADULA __ *78, Peebless.: W side of Walkerburn, NT3537, G.H. Ballantyne, 1981. *100, Clyde Is.: Trackside, Kilmory, Arran, NR959210, A.R. Church & G.H. Ballantyne, 1996; A.R. Church & D.E. Allen, 1997. 75/8.253. RUBUS RUDIS *37, Worcs.: Scrub in woodland clearing, Papermill Coppice near Alfrick, SO7451, P.G. Garner, R. Maskew & A. Newton, 1993. *64, Mid-W. Yorks.: Edge of woodland, Gisburn, SD821496, D.P. Earl & J. Buckley-Earl, 1997. New northern limit? 75/8.257. RUBUS SECTIRAMUS *11, S. Hants.: Upper Grenville Copse, Hambledon, S$U635172, D.E. Allen, 1996. *12, N. Hants.: Wood margin, Hartley Wood Common, Hartley Wespall, SU692581, D.E. Allen, 1994. *13, W. Sussex: Oakwood by sea, Chalkdock Point, near West Itchenor, SU790013, D.E. Allen, 1996. All BM, conf. A. Newton. 568 PLANT RECORDS 75/8.264. RUBUS ATREBATUM *11, S. Hants.: Margin of chestnut plantation, Hall Copse, near Romsey, SU329209, D.E. Allen, 1981, 1989, BM, conf. A. Newton. 75/8.266. RUBUS BERCHERIENSIS *46, Cards.: Bank below road, Cwm Degwell, 200m S of St Dogmaels Abbey, SN163454, D.E. Allen & A.O. Chater, 1995, BM, conf. A. Newton. 75/8.268. RUBUS DASYPHYLLUS *78, Peebless.: Old railway, Peebles, NT253401, G.H. Ballantyne, 1988. 75/8.270. RUBUS HYLOCHARIS *50, Denbs.: Roadside, Padog, SH8351, G. Battershall, 1996. [100, Clyde Is.: 1965 Arran specimen det. E.S. Edees in NMW is R. pictorum, det. D.E.Allen.] *101, Kintyre: Car park, Skipness, NR903579, A.R. Church, 1997, BM, det. D.E. Allen, conf. A. Newton. 75/8.273. RUBUS MARSHALLII *9, Dorset: Heath and conifer plantation, Black Down, ST6087 & 6187, D.E. Allen, 1997, BM, conf. A. Newton. 75/8.275. RUBUS ASPERIDENS *37, Worcs.: Scrub in coniferous plantation, Island Coppice, Kyre near Tenbury Wells, S06464, R. Maskew & A. Newton, 1992. 75/8.276. RUBUS MURRAYI = _*37, Worcs.: Trackside in deciduous plantation, Ellbatch Wood near Abberley, SO7366, J.J. Day, R. Maskew & A. Newton, 1990. 75/8.278. RUBUS NEWBRIDGENSIS [12, N. Hants.: Delete sole record in Fl. Hants., 158 (1996); further study has shown population is of a different but unnamed species, D.E. Allen & A. Newton. | 75/8.284. RUBUS RILSTONEI *46, Cards.: Roadside hedgebanks 250m ESE of Penwaun, St Dogmaels, SN157442, D.E. Allen & A.O. Chater, 1995, BM, conf. A. Newton. 075/8.286. RUBUS SCABRIPES [17, Surrey: Delete record published in Watsonia 20: 291 (1995).] 75/8.299. RUBUS PEDEMONTANUS *37, Worcs.: Trackside in coniferous woodland, Ribbesford Wood near Bewdley, SO7872, R. Maskew, A. Newton & W.A. Thompson, 1997. 75/8.307. RUBUS CONJUNGENS _*S, Channel Is.: Bouley Bay, Jersey, WV6654, W.M. Rogers, 1897, BM, conf. A. Newton. 75/8.308. RUBUS EBORACENSIS *78, Peebless.: Roadside, Nether Harsburgh, NT304394, D.J. McCosh, 1983, det. D.E. Allen. 75/8.310. RUBUS HEBRIDENSIS *94, Banffs.: Road junction, Dufftown, NJ323392, D. Welch, 1992, ABD, det. A. Newton. 75/8.312. RUBUS LATIFOLIUS *71, Man: Open hillside, The Creggans, Knockaloe, SC238818, D.E. Allen, 1975, BM, conf. A. Newton. 75/8.314. RUBUS PICTORUM *35, Mons.: Hedge, NW of White House, SO4222, M. Porter & R.D. Randall, 1993 75/8.315. RUBUS PRUINOSUS *46, Cards.: Disused railway embankment, Trawsgoed, SN663733, D.E. Allen & A.O. Chater, 1997. 75/8.317. RUBUS TENUIARMATUS 42, Brecs.: Hedge and lane verge, Digedi valley near Llanigon, SO2138, M. Porter, 1997, herb. M.P. 2nd record. 75/8.319. RUBUS TUBERCULATUS *100, Clyde Is.: Edge of drive, Whiting Bay, Arran, NS046254, A.R. Church & A. Smith, 1997, BM, conf. D.E. Allen. 75/8.arl. RUBUS ARICONIENSIS *35, Mons.: Woodland, Kingswood, SO4612, M. Porter & R. D. Randall, 1993. 75/8.isc. RUBUS ISCANUS *35, Mons.: Woodland, Kingswood, SO4612, M. Porter & R.D. Randall, 1993. 75/8.per. RUBUS PERCRISPUS *46, Cards.: Laneside hedge 300m WSW of Bryneithyn, Llanfarian, SN579780, D.E. Allen & A.O. Chater, 1997, BM. 75/8.vag. RUBUS VAGENSIS *35, Mons.: Woodland, Kingswood, $O4612, M. Porter & R.D. Randall, 1993. *37, Worcs.: Hay Slad, West Malvern, SO7644, P.G. Garner, R. Maskew & A. Newton, 1993. *42, Brecs.: Woodland, Park Wood, Talgarth, SO168338, M. Porter, 1991, herb. M.P. PLANT RECORDS 569 75/8.ven. RUBUS VENETORUM *104, N. Ebudes: Laig Bay, Isle of Eigg, NM4802, S.M. Macvicar, 1895, BM, det. D.E. Allen. 75/9.5. POTENTILLA ARGENTEA 95, Moray: One plant on waste ground, Rafford, NJ053551, P.R. Green, 1996. Ist record for well over 100 years. 75/9.13x14. POTENTILLA ERECTA x P. ANGLICA (P. x SUBERECTA) *37, Worcs.: Trackside verge, Lickey Hills, SO9875, J.J. Day & R. Maskew, 1992, conf. B. Harold. +*77, Lanarks.: Junction between grassy bank and gravelly shore, Hillend Reservoir, NS8467, P. Macpherson, 1996, herb. P.M., det. B. Harold. 75/9.14. POTENTILLA ANGLICA *95, Moray: Woodland ride, Quarry Wood, W of Elgin, NJ174631, P.R. Green, 1997. Ist confirmed record; possibly brought in with forestry vehicles. 75/9.14x15. POTENTILLA ANGLICA x P. REPTANS (P. x MIXTA) — *37, Worcs.: Rough grassland, Clows Top, SO7171, R. Maskew, 1994, det. B. Harold. 775/11.3. FRAGARIA x ANANASSA *49, Caerns.: Side of disused railway, Bontnewydd, SH4757, G. Battershall & W. McCarthy, 1996. Waste ground, Trevor, SH3847, G. Battershall & R. Lewis, 1996. Ist and 2nd records. +75/11.chi. FRAGARIA CHILOENSIS *S, Channel Islands: Roadside bank, Rue La Rade outside Beauvoir, Sark, WV465764, R.M. Veall, 1933, JSY, det. A. Chevalier. Established on bank until site cleared about 1980. Portions of plants planted in garden and still present (1997) 75/15.2. AGRIMONIA PROCERA 61, S.E. Yorks.: Woodland, Hagg Wood, 0-8km E of Dunnington, SE6852, M. Hammond, 1996. Only extant locality. 75/17.3. SANGUISORBA MINOR *91, Kincardines.: Disturbed ground at abandoned railway yard, Crathes, NO7496, D. Welch, 1996, ABD. +75/17.3b. SANGUISORBA MINOR subsp. MURICATA *50, Denbs.: Roadside near café, Redbrook, SJ508412, K. Watson, 1991, NMW. +75/18.4. ACAENA INERMIS *44, Carms.: Disturbed river shingle, Llangadog, SN706277, I.K. Morgan & R.D. Pryce, 1987, NMW. 75/19.10b. ALCHEMILLA FILICAULIS subsp. VESTITA *46, Cards.: Flushed SW facing slope, sheep-grazed pasture with boulders, Cwm Berwyn 800m WNW of Diffwys, SN726780, A.O. Chater, 1988, NMW. 775/19.15. ALCHEMILLA MOLLIS 49, Caerns.: Grassy bank W side of B5106, between Lodge Hotel and Bedal Inn, Talybont, SH7669, R. Lewis, 1996. Grassland near Elephant’s Cove, Great Orme, SH7783, W. McCarthy, 1996. 2nd v.c. records. 50, Denbs.: Edge of car park, Llanrhaiadr ym Mochnant, SJ1226, J.A. Green, 1996. Probable garden escape, 2nd record. 775/21.1. ROSA MULTIFLORA *11, S. Hants.: Thicket with R. canina, Linford Bottom, Linford, SU179070, G.D. Field, 1997, det. A.L. Primavesi. 175/21.3. ROSA LUCIAE *41, Glam.: Scrub at edge of old colliery tip below Kenwood Kennells, adjoining salt-marsh, W of Llanmorlais, Gower, SS526945, Q.O.N. Kay, 1995, herb. Q. O.N.K. Ist Welsh record. Presumably introduced or a garden throw-out but appears established. 75/21.4x11. ROSA ARVENSIS x R. STYLOSA (R. x PSEUDORUSTICANA) *4, N. Devon: Lane hedge, Sourton Quarry, SX521897, W.H. Tucker, 1997, conf. L.J. Margetts. Ist record under this name. 75/21.4x12. ROSA ARVENSIS x R. CANINA (R. X VERTICILLACANTHA) 4, N. Devon: Climbing through bushes by River Taw, North Wyke, SX6598, R.E.N. Smith, 1997, conf. A.L. Primavesi. 2nd record. 75/21.4x13b. ROSA ARVENSIS x R. CAESIA subsp. VOSAGIACA —_*37, Worcs.: Scrub on shore of reservoir, Upper Bittell Reservoir, Cofton Hackett, SP0275, R. Maskew, 1995, conf. A.L. Primavesi as hybrid with R. arvensis as female parent. 75/21.4x14. ROSA ARVENSIS x R. OBTUSIFOLIA (R. x ROUYANA) *37, Worcs.: Scrub at edge of woodland, Brazier’s Coppice, Stoke Bliss near Tenbury Wells, S06464, R. Maskew, 1991, det. as hybrid with R. obtusifolia as female parent. 75/21.4x16. ROSA ARVENSIS x R. SHERARDII *37, Worcs.: Hedge near Harpley, SO6860, R. Maskew, 1995. Hedge near Tanner’s Brook, SW of Wyre Forest, SO7273, R. Maskew & W.A. Thompson, 1995; both det. R.M. as hybrids with R. sherardii as female parent. Ist and 2nd records. 570 PLANT RECORDS 75/21.4x19. ROSA ARVENSIS x R. MICRANTHA (R. x INELEGANS) *37, Worcs.: Woodland, Coalpits Wood near Knightwick, $O7254, R. Maskew & W.A. Thompson, 1995, conf. A.L. Primavesi as hybrid with R. micrantha as female parent. 75/21.5x17. ROSA PIMPINELLIFOLIA x R. MOLLIS (R. x SABINIT) *78, Peebless.: Innerleithen, NT33, A. Craig-Christie, 1873, E, det. A. L. Primavesi. 75/21.5x18. ROSA PIMPINELLIFOLIA x R. RUBIGINOSA (R. x BITURIGENSIS) 93, N. Aberdeen: Roadside, Wells of Ythan, NJ623368, D. Welch, 1996, herb. D.W., det. G.G. Graham. Ist post- 1930 record. +75/21.8. ROSA FERRUGINEA *95, Moray: One bush self-sown on old tombstone, Alves, near Elgin, NJ134629, I.P. Green, 1996. 75/21.11x20. ROSA STYLOSA x R. AGRESTIS *11, S. Hants.: Heathland roadside, B3078 Godshill Ridge, SU186156, P.D. Stanley, 1997, det. A.L. Primavesi. 2nd British record. 75/21.12x13a ROSA CANINA x R. CAESIA subsp. CAESIA (R. x DUMALIS) _*37, Worcs.: Scrub on side of Tanners Brook, SW of Wyre Forest, SO7273, R. Maskew & W.A. Thompson, 1995. 75/21.12x15. ROSA CANINA x R. TOMENTOSA (R. x SCABRIUSCULA) 17, Surrey: Two bushes in hedgerow, Byfleet, TQ071606, B.W. Phillips & K. Page, 1997, conf. A.C. Leslie & P. Stanley. 2nd record. 75/21.12x17. ROSA CANINA x R. MOLLIS (R. x MOLLETORUM) *94, Banffs.: Hedge, Broomhills farm road, NJ582646, J. Edelsten, 1997, det. R. Maskew. 75/21.12x18. ROSA CANINA x R. RUBIGINOSA (R. x NITIDULA) *37, Wores:: Scrub-on limestone grassland, Walsgrove Hill, Great Witley, SO7465, R. Maskew, 1995, det. as hybrid with R. canina as female parent. 75/21.12x19. ROSA CANINA x R. MICRANTHA (R. x TODDIAE) *37, Worcs.: Calcareous grassland, Gadbury Bank near Eldersfield, SO7931, R. Maskew, 1989, det. A.L. Primavesi as hybrid with R. micrantha as female parent. *47, Monts.: Llanymynech Rocks, SJ/26.21, R. Maskew, 1995. 75/21.12x20. ROSA CANINA x R. AGRESTIS (R. x BELNENSIS) *11, S. Hants.: Heathland roadside, B3078 Godshill Ridge, SU186156, P.D. Stanley, 1997, det. A.L. Primavesi. *49, Caerns.: Wood, near Keeper’s Cottage, Penrhynside, Llandudno, SH8081, W. McCarthy, 1996. Ist Welsh record. 75/21.13x16. ROSA CAESIA x R. SHERARDII *94, Banffs.: Hedge, Mains of Durn farm road, NJ586651, J. Edelsten, 1997, det. R. Maskew. 75/21.13bx14. ROSA CAESIA subsp. VOSAGIACA x R. OBTUSIFOLIA *37, Worcs.: Scrub on limestone grassland, Walsgrove Hill, Great Witley, SO7465; Scrub on limestone grassland, Bredon Hill, S09439; both R. Maskew, 1995. 1st and 2nd records. 75/21.11x14. ROSA STYLOSA x R. OBTUSIFOLIA *37, Worcs.: Roadside bank by A44, Broadwas, SO7555, R. Maskew, 1997, conf. A.L. Primavesi as hybrid with R. obtusifolia as female parent. 75/21.14x15. ROSA OBTUSIFOLIA x R. TOMENTOSA *37, Worcs.: Scrub in deciduous plantation in Trench Wood, Oddingley, SO9259, R. Maskew, 1996, det. as hybrid with R. tomentosa as female parent. 75/21.14x19. ROSA OBTUSIFOLIA x R. MICRANTHA *37, Worcs.: Scrub on calcareous grassland near Broadwas, SO7755, R. Maskew, 1997, conf. A.L. Primavesi as hybrid with R. micrantha as female parent. : 75/21.15x16. ROSA TOMENTOSA x R. SHERARDII (R. x SUBERECTIFORMIS) *37, Worcs.: Scrub on roadside at Broad Heath, SO6765, R. Maskew, 1994, conf. A.L. Primavesi as hybrid with R. sherardii as female parent. 75/21.17. ROSA MOLLIS 50, Denbs.: Cliffs in Country Park, Mold, SJ1962, W. McCarthy, 1996. Limestone grassland, Maeshafn, SJ2061, BSBI group, 1996, det. A.L. Primavesi. 2nd v.c. records. 75/21.18. ROSA RUBIGINOSA 50, Denbs.: Limestone hillside, Rhyd y Foel, SH912763, W. McCarthy, 1997. Canal bank, Trevor, $J258414, J.A. Green, 1997. Ist & 2nd recent records. PLANT RECORDS Sal 75/21.19x20. ROSA MICRANTHA x R. AGRESTIS (R. x BISHOPII) *8, S. Wilts.: Hedge, Toyd Farm, 5km S of Coombe Bissett, SU089206, P.D. Stanley, 1995, det. A. L. Primavesi. 75/21.20. ROSA AGRESTIS 8, S. Wilts.: Hedge, Toyd Farm, 5km S of Coombe Bissett, SU089206, P.D. Stanley, 1995, det. A.L. Primavesi. Ist record since before 1957. #12, N. Hants.: Old chalk pit, Noar Hill, SU7431, A. Brewis, 1978, herb. A.B., det. R. Melville. "37, Worcs.: Scrub on Malvern Common, SO7744, P.G. Garner & R. Maskew, 1989, det. G.G. Graham & A.L. Primavesi. 49, Caerns.: Limestone quarry, Bodafon, Llandudno, SH807820, W. McCarthy, 1995, det. A.L. Primavesi. Limestone hillside, Great Orme, SH7683, W. McCarthy, 1996. Ist & 2nd records this century. *50, Denbs.: Limestone hillside, Colwyn Bay, SH829799, W. McCarthy, 1996, det. A.L. Primavesi. +75/22.13. PRUNUS LUSITANICA *46, Cards.: Estate woodland, Llanerchaeron, SN480602, A. O. Chater, 1982. Mixed woodland, Coed Penglanowen, Nanteos, SN610786, A.O. Chater, 1992. Ist & 2nd v.c. record; widely naturalised throughout Cards. *49, Caerns.: Steeply sloping copse near roadside, between Tyn-y-groes and Talycafn Bridge, SH785719, R. Lewis, 1993, NMW. Originally planted but now spreading from bird-sown plants. [The record for this species from Llanbedr-y-cennin in Watsonia 20(3): 292 (1995) should be deleted and replaced by the above]. Edge of woodland above B5106, Gwydir Castel, Llanrwst, SH/79.60, R. Lewis, 1976. Ist and 2nd records. 75/22.4x5. PRUNUS SPINOSA x P. DOMESTICA (P. x FRUTICANS) *37, Worcs.: Hedgerow, Eckington, SO9141, A.W. Reid, 1997, conf. R. Maskew & W.A. Thompson. +75/28.18. SORBUS DECIPIENS *47, Monts.: Berriew, SJ1600, P.J. Nethercott, 1990. +75/28.23. SORBUS LATIFOLIA *47, Monts.: Several plants with S. aria cut back into roadside hedge, Brooks, SO1499, M. Wainwright & V. Morgan, 1989, det. P.J. Nethercott. One medium- sized tree in hedgebank beside road, Berriew, SJ1600, P.J. Nethercott, 1989. Ist and 2nd records. +75/28.7. SORBUS INTERMEDIA 42, Brecs.: One young plant c.15cm tall in waste ground, Brynmawr, SO167119, M. & C. Porter, 1997. 2nd record. *44, Carms.: On waste ground, Llanelli, SS500999, I.K. Morgan, 1992, NMW, det. G. Hutchinson. *47, Monts.: One small tree on roadside bank, Berriew, SJ1600, P.J. Nethercott, 1989. Roadside hedge and woodland, Brooks, SO1499, P.J. Nethercott, 1990. 1st and 2nd records. 175/31.1. PHOTINIA DAVIDIANA *11, S. Hants.: Amongst gorse at edge of wood along heathland path, Fields Heath, Fawley, SU456023, R.P. Bowman, 1988, herb. R.P.B., det. R.P. Bowman & E.J. Clement. 175/32.34. COTONEASTER BULLATUS *4, N. Devon: Hedge near limestone quarry, Bampton, SS$969221, M. Greenwood, 1995, det. J.R. Palmer. Plantation, Meddon, SS281189, W.H. Tucker, 1997, det. J. Fryer. lst and 2nd record. 175/32.35. COTONEASTER REHDERI *4. N. Devon: In plantation, Windcutter Hill, Lee, SS487461, E. Stenger & W.H. Tucker; Old railway cutting, Landcross, SS459237, W.H. Tucker; both 1997, det. J. Fryer. 1st and 2nd records. 175/32.37. COTONEASTER DIELSIANUS *35, Mons.: Bird sown on old wall with C. horizontalis, Abergavenny, SO2914, I.K. Morgan, 1993, NMW, det. G. Hutchinson. *46, Cards.: Rocky scrub 150m SSW of Gwynfa, Tresaith, SN283511, A.O. Chater & P. Culyer, 1995, NMW. +75/32.39. COTONEASTER FRANCHETII *49, Caerns.: Bank in old quarry, Llangystenium near Colwyn Bay, SH8279, G. Battershall & W. McCarthy, 1997. Roadside, Morfa Bychan, Porthmadog, SH5337, BSBI meeting, 1997. Ist & 2nd v.c. record. *75/32.40. COTONEASTER STERNIANUS __*46, Cards.: Self-sown bush in rocky scrub N of road at W end of Cwm Rheidol Reservoir, SN695796, A.O. Chater, 1996, NMW. *49 Caerns:.: Limestone grassland bank, Great Orme, Llandudno, SH7782, W. McCarthy, 1997. +75/32.mai. COTONEASTER MAIREI *11, S. Hants.: Roadside in thick shade, Somerley area, SU121075, G.D. Field, 1998, det. J. Fryer. 775/32.mar. COTONEASTER MARGINATUS *50, Denbs.: Quarry floor, Llandulas, SH9177, W. McCarthy, 1994, det. J.Fryer. +75/33.1. PYRACANTHA COCCINEA *46, Cards.: Several plants in rocky scrub above road at W end of Cwm Rheidol Reservoir, SN695796, A.O. Chater, 1996, NMW. ay 2 PLANT RECORDS 75/35.7x8. CRATAEGUS MONOGYNA x C. LAEVIGATA (C. x MEDIA) 28, W. Norfolk: Roadside hedge, Rougham, TF836206, K. Beckett, 1996. 2nd record, c.600m from C. laevigata. *46, Cards.: One bush in hedgebank 130m SW of Tanllan, Llanbadarn Odwyn, SN633609, A.O. Chater, 1996. 75/35.8. CRATAEGUS LAEVIGATA 35, Mons.: Two trees on wet heath, S of Tredegar, SO151068, T.G. Evans, 1995, NMW. Ist record since 1946. +77/7.1b. ANTHYLLIS VULNERARIA subsp. POLYPHYLLA *49, Caerns.: Side of disused railway track, Bontnewydd, SH4757, G. Battershall & W. McCarthy, 1996. 77/7.1e. ANTHYLLIS VULNERARIA cf. subsp. LAPPONICA *45, Pembs.: Seacliff vegetation, E side of Porthclais Harbour, St David’s, SM742239, E. Norman, 1997, NMW,, det. J.R. Akeroyd. +77/12.1. HIPPOCREPIS EMERUS “*12, N. Hants.: Hedgerow, Monk Sherborne, SU607556, M. Taylor, 1997, herb. A. Brewis, det. E.J. Clement. 77/14.1. VICIA OROBUS 109, Caithness: Valley side, Sandside, by Reay, NC950660, R.E.C. Ferreira, 1996. Ist localised record. +77/14.3. VICIA TENUIFOLIA 28, W. Norfolk: Pine forest ride in long grass, Roudham, TL931884, K.A. & G. Beckett, 1997, conf. E.J. Clement. 2nd record. c.3 plants far from habitation or disturbance. 177/14.5. VICIA VILLOSA *35, Mons.: Newport Docks, ST38, T.G. Evans, 1985, NMW. 77/14.11b. VICIA SATIVA subsp. SEGETALIS *49, Caerns.: Road bank, Uwchmynydd near Aberdaron, SH1526, M. & A. Atkinson, 1997. Hedge bottom, Machroes near Abersoch, SH3226, M. & A. Atkinson, 1997. Ist & 2nd localised records of this subspecies. +77/14.13. VICIA LUTEA *35, Mons.: Less than ten plants on railway embankment near bridge over Afon Honddu, Llanvihangel Crucorney, SO3220, S.A. Rippin, 1987. +77/15.10. LATHYRUS HETEROPHYLLUS *35, Mons.: Extensive area on old railway ballast, Pye Corner, Bassaleg, ST2787, T.G. Evans, 1996. Ist Welsh record. +77/15.11. LATHYRUS HIRSUTUS *58, Cheshire: Looking established on rough bank, Wharton, Winsford, $J655667, J.P. Guest, 1997, herb. G.M.K., det. G.M. Kay. 177/15.12. LATHYRUS NISSOLIA *49, Caerns.: Grassy area at side of track leading to Llyn Parc, about 1.75km SW Llanrwst, SH7860, W. McCarthy, 1996. 51, Flints.: Roadside verge near crossroads, | km NW of Nercwys church, $J230613, J. Phillips, 1997. Frequent component of new amenity grassland on site of old rubbish tip, Etna Country Park, Buckley, $J285650, P. Hodges, 1997. Ist confirmed records post-1930. +77/15.6. LATHYRUS TUBEROSUS _ *42, Brecs.: Edge of disused railway, Cradoc, SO014304, E. Tompkins, 1997, NMW. At this site for at least 10 years. *+77/15.7. LATHYRUS GRANDIFLORUS 44, Carms.: Established in roadside hedgerow, N side of A476 at Gate Square, near Penygroes, SN578148, R.D.Pryce, 1991, conf. G. Hutchinson. 2nd record, still present in 1996. *95, Moray: Rough area of former garden on edge of forest, Darnaway Forest, near Forres, NJO06568, I.P. Green, 1996. +77/15.9. LATHYRUS LATIFOLIUS *50, Denbs.: Grassy hillside, Rhyd y Foel, SH914774, J.V. Robertson, 1997. 77/16.4b. ONONIS REPENS subsp. MARITIMA *46, Cards.: Dune at S end of Tan-y-bwlch Beach, Aberystwyth, SN579798, A.O. Chater, 1996, CGE, det. P.D. Sell. Sandy slope above sea, Traeth y Mwnt, SN194519, A.O. Chater, 1996, NMW. Ist and 2nd records. +77/16.mit. ONONIS MITISSIMA * *41, Glam.: Garden lawn, Dinas Powis, ST154709, J.P. Curtis, 1992, NMW, det. R.M. Burton. 177/17.3. MELILOTUS OFFICINALIS *10, Wight, Disturbed ground, Bembridge Harbour, SZ632885, S.A. Campbell, 1997, herb. C.R. Pope, det. R.M. Burton. Ist confirmed record this century, despite intensive searches. 77/18.5. MEDICAGO ARABICA +81, Berwicks.: Sandy bank, The Lees, NT840390, M.E. & P.F. Braithwaite, 1997. 2nd extant record. 77/19.3. TRIFOLIUM OCCIDENTALE *45, Pembs.: Several flowering patches on coastal cliff grassland, Stack Rocks, Castelmartin, SR/92.5944 and Short Point, Dale, SM798044, R.S. Cropper, 1995. 1st and 2nd records. PLANT RECORDS 29/5) +77/19.4b. TRIFOLIUM HYBRIDUM subsp. ELEGANS *77, Lanarks.: Coal bing slope, S of Garnkirk, NS6767, P. Macpherson, 1997, herb. P.M. 77/19.6. TRIFOLIUM SUFFOCATUM ~—_ 10, Wight: Frequent on sandy coastal bank, Cande Lake, Ryde, SZ602926, C.R. Pope, 1996, herb. C.R.P. 2nd extant locality. +77/19.10. TRIFOLIUM AUREUM _ 5, S. Somerset: Rough grassland at S end of Otterhead Lakes, Royston Water, ST226130, C.J. Cornell, 1997, conf. P.R. Green. 1st record for 73 years. Most likely sown with grass seed several years ago when work was done on lakes. *77/22.1x2. LABURNUM ANAGYROIDES x L. ALPINUM (L. x WATERERI) *44, Carms.: Pure stand in hedge, Banc Blodeuyn, Mynydd Llanllwni, SN4835, P.A. Smith & A.O. Chater, 1996. Mainly a pure stand in hedgerow, Ffos-y-Gaseg, about 4km SW of Brechfa, P.A. Smith & A.O. Chater, 1996. Ist and 2nd records. 177/23.2. CYTISUS MULTIFLORUS *49, Caerns.: Side of disused railway, Port Dinorwic, SH5267, G. Battershall et al., 1996. 1 plant on grass verge, near Bangor, SH5971, G. & I. Battershall, 1997. 1st and 2nd records. +77/24.1. SPARTIUM JUNCEUM *44, Carms.: One plant (self-sown?), roadside, Felinfoel, SN5202, I.K. Morgan, 1996. Amenity plantings, Machynys, SS5097, I.K. Morgan, 1996. Ist and 2nd records. 77/26.1x2. ULEX EUROPAEUS x U. GALLI *46, Cards.: Gorsey slope, W side of Pendinas, Aberystwyth, SN582804, A.O. Chater, 1997, NMW, conf. P.M. Benoit. +80/1.man. GUNNERA MANICATA *49, Caerns.: Riverside, Pont Eidda near Ysbyty Ifan, SH8350, G. & I. Battershall, 1997. 84/1.1x8. EPILOBIUM HIRSUTUM x E. CILIATUM (E. x NOVAE-CIVITATIS) *3, S. Devon: Near River Axe, Whitford Bridge, SY2695, L.J. Margetts, 1996, conf. T.D. Pennington. *37, Worcs.: Interfield, SO7749, P.G. Garner, 1993, det. T.D. Pennington. 084/1.8x12. EPILOBIUM CILIATUM x E. BRUNNESCENS _*44, Carms.: Acid quarry spoil heap, Pen-y-Dinas Quarry, Llansawel, SN629355, A.O. Chater, 1995, NMW, det. T.D. Pennington. Ist Welsh record. 784/1.13. EPILOBIUM PEDUNCULARE 45, Pembs.: Woodland garden, Colby Lodge, SN156078, G. Kitchener, 1995. 84/1.2x3. EPILOBIUM PARVIFLORUM x E. MONTANUM (E. x LIMOSUM) *46, Cards.: Hedgebank, Clawdd-moel, Temple Bar, SN531531, A.O. Chater, 1996, NMW, conf. G. Kitchener. *77, Lanarks.: Sloping grassland in light wood, Burnside, near Glasgow, NS6259, P. Macpherson, 1995, herb. P.M., det. J.R. Akeroyd. 84/1.2x8. EPILOBIUM PARVIFLORUM x E. CILIATUM *37, Worcs.: Garden weed, Wollaston, Stourbridge, SO8884, M.E. Smith, 1989, det. G.D. Kitchener. *49, Caerns.: Streamside gravels, Aberdaron, SH1726, M. & A. Atkinson, 1997. *58, Cheshire: By pool, Sandbach, SJ762610, J. H. Clarke, 1997, det. G. Kitchener. *77, Lanarks.: Base of wall by River Clyde, Blantyre, NS6958, A. McG. Stirling & P. Macpherson, 1996, herb. P.M., det. G.D. Kitchener. 84/1.3x6. EPILOBIUM MONTANUM x E. OBSCURUM (E. x AGGREGATUM) *47, Monts.: Laneside near v.c. 50 border, Ffridd Llanerch, SJ1422, P.M. Benoit, 1988. Ist localised record. 84/1.3x8. EPILOBIUM MONTANUM x E. CILIATUM *77, Lanarks.: Roadside, Plains, Airdrie, NS7966, P. Macpherson, 1995, herb. P.M., det. J.R. Akeroyd. *81, Berwicks.: One plant with both parents on waste ground, Lauder, NT527477, M.E. Braithwaite, 1997, herb. M.E.B., det. G. Kitchener. 84/1.4x6. EPILOBIUM LANCEOLATUM x E. OBSCURUM (E. x LAMOTTEANUM) *49 Caerns:.: Quarry near Aberdaron, SH1929, A.P. Conolly, 1959, det. C.A. Stace. 84/1.6x8. EPILOBIUM OBSCURUM x E. CILIATUM 44, Carms.: Swampy ditch between road and Afon Teifi, 300 m SE of Pont Stephan, Cwmann, SN583474, A.O. Chater, 1997. 2nd confirmed record. *46, Cards.: Shaded verge of A475, 700m E of Llandyfriog church, SN340409, C.D. Preston & A.O. Chater, 1996. Felled conifer plantation, Lodge Park, Tre’r-ddol, SN662931, A.O. Chater, 1997, NMW. Ist and 2nd records. 84/1.7. EPILOBIUM ROSEUM 89, E. Perth: By road along river, Blairgowrie, NO178457, O.M. Stewart, 1981. 2nd record. 574 PLANT RECORDS 84/1.9x12. EPILOBIUM PALUSTRE x E. BRUNNESCENS *46, Cards.: With both parents on damp shaley ground by Forestry road 2km S of Hafdre, Llyn Brianne, SN804512, A.O. Chater, 1995. E, NMW., conf. T.D. Pennington, G. Kitchener & D. McKean. Ist British record. 84/4.1x3. OENOTHERA GLAZIOVIANA x O. BIENNIS (O. x FALLAX) *28. W. Norfolk: Roadside, Garboldisham, TL990816, J. Hawksworth, 1994, det. J.C. Bowra. *51, Flints.: Between dunes and small marshy area near old lighthouse, Point of Ayr, SJ18, R.G. Ellis. 1980. NMW., det. K. Rostansk1. 84/4.3x4. OENOTHERA BIENNIS x O. CAMBRICA *95, Moray: Many plants on road bank. Blackhills, W of Fochabers, NJ276596, IP. Green, 1996, det. J.C. Bowra. Scattered along road bank. W of Forres, NJO06585, I.P. Green, 1996, conf. J.C. Bowra. 1st and 2nd records. 84/4.4. OENOTHERA CAMBRICA *44. Carms.: Churchyard, St Paul’s Church. Llanelli, SS508997, A.M. Pell, 1985, NMW, det. R.D. Pryce. 84/6.2. CIRCAEA ALPINA *46, Cards.: Wooded ravine of Llyfnant stream 200m W of Cwmrhaiadr, SN752962, A.O. Chater & T.D. Dines, 1997, NMW, conf. P.M. Benoit. 784/CLA.amo. CLARKIA AMOENA —_*35, Mons.: Rubbish tip, Newport, ST3085, T.G. Evans. 1985. 785/2.1. AUCUBA JAPONICA *50, Denbs.: Old hedge, Llangollen, $J184439, J. Clarke & A. Franks, 1997. 88/1.1. EUONYMOUS EUROPAEUS 81, Berwicks.: Riverside woodland, Birkwood Heaugh, M. E. & P.F. Braithwaite, 1995. 2nd extant record. 788/1.2. EUONYMUS LATIFOLIUS “*11, S. Hants.: Edge of wooded strip along track to beach, Cadland, Stanswood Bay, SZ470998, R.P. Bowman, 1997, herb. R.P.B., det. E.J. Clement. +89/1.1xper. ILEX AQUIFOLIUM x I. PERADO (I. x ALTACLERENSIS) S. Channel Islands: One bushy tree (planted), now about 8m tall, verge of footpath, opposite Flagpole Cottage (Beauregard), Sark, WV459756, R.M. Veall, 1997, JSY, det. R.A. Waterman. *50, Denbs.: Farm lane, Llangollen, SJ183434, J. Clarke & A. Franks, 1997. *73, Kirkcudbrights.: Wood edge, Netherlaw, NX740449, O.M. Stewart, 1996. *95, Moray: Self sown in roadside hedgerow, Forres, NJO58577, P.R. Green, 1996. 791/2.7. EUPHORBIA SERRULATA *4, N. Devon: Waste ground, Colleton Mills, SS659147, E Stenger, 1997, conf. L.J. Margetts. | *41, Glam.: Thirty plants on calcareous waste ground, The Orchards, Llanishen, Cardiff, ST1781, D.Green, 1995. 91/2.15. EUPHORBIA CYPARISSIAS +81, Berwicks.: Established 2m up wall, Lauder. NT529477, M.E. Braithwaite, 1997. 2nd record. 791/2.16b. EUPHORBIA AMYGDALOIDES subsp. ROBBIAE *46, Cards.: Roadside hedgebank S of Upper Bridge, Aberaeron, SN458623, A.O. Chater, 1996. Woodland just W of Cymerau Hall. Glandyfi, SN696962, S.P. Chambers, 1996. Ist and 2nd records. +91/2.0bl1. EUPHORBIA OBLONGATA _*44, Carms.: Established on garden rubbish tip, Llanelli, SN4901, LK. Morgan, 1995, NMW, det. E.J. Clement. Ist Welsh record. *46. Cards.: Scrub slope above church hall, Llanbadarn Fawr, SN598810, S.P. Chambers, 1991, conf. B. Wurzell & E.J. Clement. 7+93/2.2. PARTHENOCISSUS INSERTA *95. Moray: Creeping over gorse in waste area, Cummingstown, NJ135692, P.R. Green, 1996. 799/1.5. ACER SACCHARINUM *77, Lanarks.: Surviving seedling on pathside in wood, Corehouse, NS8841, P. Macpherson, .1996, herb. P.M., det. C.S. Crook. 7100/1.1. RHUS TYPHINA *46, Cards.: Well naturalised thicket of suckers in paddock, Penrhyncoch, SN646842, A.O. Chater, 1997. *49, Caerns.: Side of path, Glynllifon, SH4655, G. & I. Battershall, 1997. *77, Lanarks.: Bank of River Clyde, Linthouse, NS5466, P. Macpherson. 1992, herb. P.M., conf. J.R. Akeroyd. One plant surviving in 1998. 7102/1.3. OXALIS CORNICULATA *50, Denbs.: Wall in village, Nantglyn, SJO062, BSBI group, 1996. 51, Flints.: Abundant around disused glasshouses, Thornleigh Park, Sealand. SJ3666, G. Kay & BSBI group, 1996, conf. M.F. Watson. Ist post-1930 record. 7102/1.4. OXALIS EXILIS 73. Kirkcudbrights.: Grass by road, Mavis Grove road S of Maxwelltown, O.M. Stewart, 1996. 2nd record. PLANT RECORDS 575 *102/1.8. OXALIS ARTICULATA 46, Cards.: Roadside hedgebank N of Afon Cledan, 150m E of A487, Llanon, SN515668, A.O. Chater, 1996. 2nd record. *102/1.10. OXALIS DEBILIS 46, Cards.: Several plants in rough grass on floor of old sandpit just N of A487(T), Banc y Warren, Penparc, SN203477, A.O. Chater, 1996. 2nd record. 7102/1.14. OXALIS INCARNATA *46, Cards.: Roadside 100m W of Llangranog church, SN315540, A.O. Chater, 1996. +103/1.1. GERANIUM ENDRESSII 81, Berwicks.: An established clump in damp grassland, near Lauder, NT521481, M.E. Braithwaite, 1997. Ist localised record. +103/1.1x2. GERANIUM ENDRESSII x G. VERSICOLOR (G. x OXONIANUM) *46, Cards.: Streambank by churchyard, Llanilar, SN623751, A.O. Chater & S.P. Chambers, 1994, NMW. 103/1.4. Geranium ROTUNDIFOLIUM 51, Flints.: Spontaneous garden weed, Gronant, SJ0983, J. Phillips, 1996. 2nd v.c. record. +103/1.13. GERANIUM IBERICUM *50, Denbs.: Edge of stream, Llanrwst, SH799623, R. Lewis, 1997. 7103/1.13xpla. GERANIUM IBERICUM x G. PLATYPETALUM (G. x MAGNIFICUM) *49, Caerns.: Naturalised on a bank, Port Dinorwic, SH5267, G. & I. Battershall, 1996. 103/1.15. GERANIUM PUSILLUM 46, Cards.: Disturbed flowerbed, Art Dept. grounds, Buarth Mawr, Aberystwyth, SN588815, S.P. Chambers, 1996. Sandy swede field, Nantyferwig, SN168482, A.O. Chater, 1997. lst and 2nd extant localities. +103/1.17. GERANIUM MACRORRHIZUM __—- *46, Cards.: Footpath verge, SE of National Library, Aberystwyth, SN595814, A.O. Chater, 1992. Salix cinerea scrub, 200m NW of Cross Inn, SN542642, A.O. Chater, 1997. Ist and 2nd records. *73, Kircudbrights.: Woodland edge, Port Ling, NX883540, O.M. Stewart, 1996. 7103/1.bru. GERANIUM BRUTIUM *29, Cambs.: Melbourn by-pass, Meldreth, TL384461, P.D. sell’ 1997, CGE. 103/2.2. ERODIUM MOSCHATUM 46, Cards.: Shaley track margin and in adjacent field, 700m ESE of Capel Tygwydd, SN277432, A.O. Chater, 1996. 2nd record. 7+103A/LIM.dou. LIMNANTHES DOUGLASII *S, Channel Islands: Cliff path, Little Sark near Les Fontaines, Sark, WV451740, M.W. Angel, 1991, JSY, det. D. McClintock from photograph. *58, Cheshire: Newly-seeded roadbank, Stockport, $J904902, G.M. Kay & E. Kearns, 1997. +104/1.maj. TROPAEOLUM MAJUS = *44, Carms.: Behind rubble at edge of sea shore, St Johns Hill, Laugharne, SN302104, G. Hutchinson, 1991, NMW. *49, Caerns.: Waste ground, Porthmadog, SH5739, G. & I. Battershall, 1996, and J. Clarke & A. Franks, 1997. 105/1.1. IMPATIENS NOLI-TANGERE *46, Cards.: In a gully at Capel Bangor, SN67, C. Morley, 1939, in MS notebook, Ipswich Museum, per A.P. Fowles. 106/1.1., HEDERA COLCHICA *46, Cards.: Scrub and wood, Caerhedyn, Llyfnant, SN709974, few). Chater, 1997. *47, Monts.: Scrambling up lakeside bank and trees, Lake Vyrnwy, SH962222, J. Clarke, 1997, det. A. Rutherford (as var. dentata). 106/1.2b. HEDERA HELIX subsp. HIBERNICA *50, Denbs.: Ruin, Llangwm, SH9844, A.O. Chater, 1996. Roadside wall, Llanrhaiadr ym Mochnant, SJ1226, J.A. Green, 1996. Ist and 2nd records. 7 107/1.ran. HYDROCOTYLE RANUNCULOIDES *11, S. Hants.: Stream flowing through grounds, Southampton University, Highfield, Southampton, SU424151, C.M. Cockerill, 1997, herb. R.P. B., det. R.P. Bowman. Browndown, Gosport, SZ583996, M.J. Southam, 1997. Ist and 2nd records. *35, Mons.: 4 large patches, Broadway Reen, Marshfield, ST2681, T.G. Evans, 1996. +107/3.1. ASTRANTIA MAJOR *46, Cards.: Well established on shaley slope below Forestry road in conifer forest 2-3km SE of Eisteddfa Gurig, SN811820, A.O. Chater, 1996, NMW. With Fuchsia magellanica and Cotoneaster horizontalis and presumably deriving from throw-outs, although 2km from nearest garden. 107/16.1. BERULA ERECTA 93, N. Aberdeen: Marsh, Logie Buchan, NK0029, D. Welch, 1996, ABD. Only extant locality. 576 PLANT RECORDS 107/19.3. OENANTHE PIMPINELLOIDES *61, S.E. Yorks.: Seventy plants in neutral grassland, on edge of playing field, Ennerdale Sports Centre, Hull, TA0933, R. Middleton, 1997, HLU, det. M.J. Southam, Ist record N of Ipswich. +107/30.1. SISON AMOMUM *44, Carms.: About fifty plants, presumably introduced with material from Slimbridge, in marshy area at edge of path, S of main wildfowl lake, Penclacwydd Wildfowl and Wetland Trust Centre, Llanelli, SS530986, B. Stewart, 1991, det. I.K. Morgan. Ist post-1930 record. 107/31.1. CICUTA VIROSA —_*37, Worcs.: Marginal reed-swamp, Podmore Pool, Kidderminster, SO8477, C.B. Westall, 1989, conf. J.J. Day & R. Maskew. +107/38.1. LEVISTICUM OFFICINALE *35, Mons.: One plant in gap in pavement, Chepstow, ST5393(G), C.A. Shirley, 1994, NMW. *95, Moray: Several plants self sown, waste ground on dunes, Finehorn, NJ043643, I.P. Green, 1997. + 107/39.3. PEUCEDANUM OSTRUTHIUM *67, S. Northumb.: At W end of enclosure around long-deserted farm house, Grindon Green, NY729736, G. Simpson, 1992, herb. G.A. Swan. Only extant locality. *91, Kincardines.: Den by old graveyard, Garvock, NO743705, D. Welch, 1997. 107/40.1. PASTINACA SATIVA +42, Brecs.: Edge of railway track, Garth Station, SN953494, L. J. Smith, 1996. 2nd record. +*73, Kirkcudbrights.: 6-7 plants on roadside, A75 just E of road off to Girton, NX6154, O.M. Stewart & A. White, 1996. +107/41.2. HERACLEUM MANTEGAZZIANUM *46, Cards.: Naturalised for c. 27 years, farmyard, Ynys Edwin, Eglwys-fach, SN678962, A.O. Chater & W.M. Condry, 1997. =O1" Kincardines.: By River North Esk, Morphie, NO708636, D. Welch, 1994. Well established so present here for some years. 108/4.1. BLACKSTONIA PERFOLIATA _*77, Lanarks.: Very widespread in abandoned industrial estate, Netherton, NS7855, W. Brackenridge, 1997, herb. P. Macpherson. 108/5.2. GENTIANELLA CAMPESTRIS 83, Midlothian: About twenty plants in rich grassland, Fullarton Water, only 2m away from lay-by after bridge, NT283564, C. Dixon, 1997, det. D.R. McKean. Ist record post-1934. 108/5.5. GENTIANELLA ANGLICA subsp. ANGLICA *37, Worcs.: Limestone grassland, Tunnel Hill, SP0247, J.W. Meiklejohn, 1992, conf. A.W. Reid. Limestone grassland, Highclere, SP0147 & SP0148, D.T. Holyoak, 1992. Ist and 2nd records. 7110/1.1. NICANDRA PHYSALODES 44, Carms.: Occasional garden weed on disturbed soil, Maesquare, W of Crymlyn Manor, Bethlehem, SN653233, M. Williams, 1995, NMW, det. G. Hutchinson. 2nd record. 110/8.1x3. SOLANUM NIGRUM x S. PHYSALIFOLIUM (S. x PROCURRENS) *11, S. Hants.: Potato field, North Hayling, SU728030, P.D. Stanley & E.J. Clement, 1996, herb. R.P. Bowman, det. E. J.C. +110/8.3. SOLANUM PHYSALIFOLIUM *49, Caerns.: Flower beds, town centre, Llandudno, SH7782, W. McCarthy, 1996. +110/8.4. SOLANUM SARACHOIDES *95, Moray: One plant on waste ground, rubbish tip, Elgin, NJ236631, P.R. Green, 1997. +110/8.7. SOLANUM TUBEROSUM _ *46, Cards.: Naturalised for at least 5 years in gully on sea cliffs, 300m ENE of Mwnt church, SN198521, A.O. Chater, 1997. +110/8.ros. SOLANUM ROSTRATUM *11, S. Hants.: Edge of refuse tip in gravel extraction area, along footpath, Efford, Keyhaven, $Z310922, M.E. Young, 1997, herb. R.P. Bowman. +110/8.vil. SOLANUM VILLOSUM *35, Mons.: Farmyard, Rogiet, ST48, T.G. Evans, 1990s, NMW. 111/3.2b. CALYSTEGIA SEPIUM subsp. ROSEATA *46, Cards.: Grassy railway embankment, 500m W of Ynys Edwin, Eglwys Fach, SN673962, A.O. Chater & W.M. Condry, 1991, NMW. 111/3.2x4. CALYSTEGIA SEPIUM x C. SILVATICA (C. x LUCANA) _*37, Worcs.: Waste ground, Billesley Common, SP0880, J.W. Partridge, 1995, det. R.K. Brummitt. 111/3.3x4. CALYSTEGIA PULCHRA x C. SILVATICA (C. x HOWITTIORUM) *¥1, S. Hants.: In roadside hedge opposite house and garden, twining over Symphoricarpos, A3057 Lower Brook, Kings Somborne, SU339278, R.P. Bowman, 1997, herb. R.P.B., det. R.K. Brummitt. PLANT RECORDS a7 113/2.1. NYMPHOIDES PELTATA 751, Flints.: Established from planted stock in shallow muddy water, Greenfield Country Park, SJ193773, G. Wynne, 1997. 2nd record. +*60, W. Lancs.: Old mill pond, Fleetwood, SD327465, D. & J. Steeden, 1985. 7114/1.1. POLEMONIUM CAERULEUM *50, Denbs.: Naturalised on rough banks, Carrog, SJ1046, S.P. Chambers, 1996. 115/1.1., PHACELIA TANACETIFOLIA *35, Mons.: Horsington Yard, Abergavenny, SO21, R. Fraser, 1988, NMW. About 300 plants in strip across rape field, E of Gwern-eiddig, SO4106, J.D. R. Vernon, 1996; planted to keep away aphids. 1st and 2nd records. *41, Glam.: Established garden throw out, Whitchurch, Cardiff, ST152795, G. Hutchinson, 1991, NMW. *42, Brecs:.: Weed on garden path, Cwm Gwdi, 3km SW of Brecon, SO0226, C.H.G. Allum, 1996. Introduced with bird seed? *43, Rads.: Garden path, Presteigne, SO3164, S. Voelcker, 1996, det. D.R. Humphreys. *+116/2.4. ECHTUM PININANA *3, S. Devon: Self seeding in a lane, Ringmore, SX6545, M. Catt, 1997. : ¥116/12.2. AMSINCKIA MICRANTHA *58, Cheshire: New earth bank, Moreton, S$J263901, P.G. Gutteridge, 1997, det. G.M. Kay. 77, Lanarks.: Well established in recently planted woodland, Blantyre, NS6829, K. Watson, 1997, GLAM, det. E.J. Clement. Ist record since 1923, presumably an accidental introduction. 116/15.3. MYOSOTIS STOLONIFERA *77, Lanarks.: Flushes by the Carsehope Burn, SW of Daer Reservoir, NS9505, K. Watson, 1997, herb. P. Macpherson. 7116/4.2. SYMPHYTUM ASPERUM *830, Roxburghs.: One large established clump, roadside verge, Pinnacle, Ancrum, NT589254, H. Noltie, 1992, herb. R.W.M. Corner. +116/4.4. SYMPHYTUM GRANDIFLORUM *49, Caerns.: Edge of wood, Edern near Nefyn, SH2839, G. Battersall et al., 1996. Grassy roadside, Bethesda, SH6067, J. Hawksford, 1997. Ist and 2nd records. 7¥116/9.2. BORAGO PYGMAEA ~—_*35, Mons.: Narrow band of plants among foxgloves, bracken and brambles in strip of tall plants in middle of steep field, Yew Tree Cottage, Long Lane, Penallt, $0527092, S.J. Tyler, 1995, herb. T.G.E., det. T.G. Evans, conf. J.R. Akeroyd. +118/4.1c. LAMIASTRUM GALEOBDOLON subsp. ARGENTATUM *50, Denbs.: Road verge, Bryneglwys, SJ1446, S.P. Chambers, 1996. 118/5.2. LAMIUM MACULATUM 50, Denbs.: Established on bank, Maerdy, SJ0144, S.P. Chambers, 1996. 2nd record. 7118/17.1. SATUREJA MONTANA = *44, Carms.: About thirty-five plants well naturalised, street- side brick wall, Bryn Avenue, Barry Port, SN453009, I.K. Morgan, 1993. 118/18.5. CLINOPODIUM ACINOS *10, Wight: Fifty-six plants counted in old chalk pit, Nunwell Down, Brading, S$Z592872, A. Campbell, 1997, herb. C.R.P., det. C.R. Pope. Only extant locality. *+118/23.3x4. MENTHA SUAVEOLENS x M. SPICATA (M. x VILLOSA) *46, Cards.: Scrub above coast path, W end of Tresaith beach, SN277514, A.O. Chater, 1994, NMW. 118/23.4. MENTHA SUAVEOLENS +*81, Berwicks.: Established on riverside, River Tweed below Clintmains, NT606324, M.E. Braithwaite, 1995, herb. M.E.B., det. R.M. Harley. +118/24.1. ROSMARINUS OFFICINALIS 49, Caerns.: Sand dunes, West shore, Llandudno, SH7781, W. McCarthy, 1996. 2nd record. 118/25.4. SALVIA VERBENACA 50, Denbs.: Limestone rock outcrops, Henllan, SJO27697, C. Welsh, 1997, det. J.A. Green. Grassland, Rhyd y Foel, SH914762, J.V. Robertson, 1997. Ist recent records 120/1.2. CALLITRICHE TRUNCATA 15, E. Kent: Marsh dyke, Denge Beach, TRO517, E.G. Philp, 1996, MNE. Ist record for about fifty years. 120/1.6. CALLITRICHE BRUTIA *110, Outer Hebrides: Beinn na h-Aire, S Uist, NF8435, P. Smith, 1994. 121/1.1. PLANTAGO CORONOPUS 47, Monts.: Sandy saltmarsh, near Dovey Junction Station, SN694987, Flora project field party, 1989. 2nd record. 578 PLANT RECORDS 121/1.3b. PLANTAGO MAJOR subsp. INTERMEDIA *50, Denbs.: Sandy bank, Sutton Green, SJ417489, K. Watson, 1993, NMW. *122/1.2. BUDDLEJA DAVIDII *91, Kincardines.: Waste ground by petrol station, Banchory, NO709960, D. Welch, 1997. +123/1.susxvir. FORSYTHIA SUSPENSA x F. VIRIDISSIMA (F. x INTERMEDIA) ONE Hants: Lane, Kingsley, SU787380, A. Brewis, 1997. *35, Mons.: Bank of River Ebbw, Risca, ST29, J. Harper, 1992. The River Ebbw bank has been reinforced and raised and many strange, planted? species have appeared there; this may be one of them. 7123/2.ang. FRAXINUS ANGUSTIFOLIA *29, Cambs.: Roadsides, Cambridge, TL45, C.D. Preston, G. Crompton & G.M.S. Easy, 1986 & 1987. ?Frequently planted now. 124/1.2. VERBASCUM VIRGATUM *79, Selkirks.: Pathside (now part of Southern Upland Way), near Galafoot, Galashiels, NT511351, J. Murray, 1995, herb. R.W.M. Corner, conf. D.R. McKean. 124/1.2x7. VERBASCUM VIRGATUM x V. THAPSUS (V. x LEMAITREI) *83, Midlothian: With both parents in waste ground, opposite Duddington caravan site, NT285721, S. Maxwell, 1996, E, det. D.R. McKean. New Scottish record. +124/1.6. VERBASCUM DENSIFLORUM *35, Mons.: One plant with three main branched stems, on trackside in Council road maintenance site, Dixton, Monmouth, SO5113, T.G. & U.T. Evans, 1997, herb. T.G.E. 124/8.2. CHAENORHINUM MINUS *91, Kincardines.: Railway cess, Newtonhill, NO910935, D. Welch, 1997. 124/1.11. VERBASCUM PULVERULENTUM “*44, Carms.: Waste ground, E of Llanelli Railway Station, SS508993, T. Davies, 1991. +124/10.1. ASARINA PROCUMBENS _*58, Cheshire: Old wall, Alderley Edge, SJ851781, A.R. Franks, 1997. Present for at least 10 years. *124/11.2. CYMBALARIA PALLIDA *81, Berwicks.: Established on lower town walls, Lauder, NT529476, M.E. Braithwaite, 1997. +124/11.3. CYMBALARIA HEPATICIFOLIA *40, Salop: Garden escape, growing on a wall, Grinshill, SJ515238, T. Preece, 1994. 124/12.2. KICKXIA SPURIA *43, Rads.: On disturbed ground, S of Penlan Wood, Clyro, S0208440, R.G. Woods, 1988, NMW. 124/13.1x4. LINARIA VULGARIS x L. REPENS (L. x SEPIUM) *37, Worcs.: Railway track near Langley Green Station, SO9988, B.R.W. Fowler, 1987. 124/13.3x4. LINARIA PURPUREA x L. REPENS (L. x DOMINII) *11, S. Hants.: Edge of building site on bank between pavement and fence, Milford on Sea, SZ289914, V. Scott, 1997, herb. R.P. Bowman, det. R.P. Bowman & C.A. Stace. *37, Worcs.: Waste ground, Moseley, SP0882, J.J. Day, P.G. Garner & B. Westwood, 1993. +124/15.1. ERINUS ALPINUS *95, Moray: Well naturalised in masonry on on Dulicht railway bridge, Grantown-on-Spey, NJ024282, J.R. Edelsten, 1996. 124/16.19. VERONICA AGRESTIS 10, Wight: Cracks in paving stones, Yarmouth, $Z357897, P. & G. Stanley, 1997, herb. P.S. 2nd recent record. 7124/16.25. VERONICA LONGIFOLIA *35, Mons.: Waste ground outside walled garden, Llanfoist Church, SO2813, T.G. Evans, 1990, NMW, det. E.J. Clement. *46, Cards.: Rough grass by long-overgrown tip, Blaendolau, Llanbadarn Fawr, SN600804, A.O. Chater, 1990. 124/20.1x5. EUPHRASIA ROSTKOVIANA x E. ARCTICA *42, Brecs.: Heathy field, Coelbren, SN8612, M. Porter, 1980, herb. M.P., det. A.J. Silverside. Ist confirmed Welsh record. 124/20.3x7. EUPHRASIA ANGLICA x E. NEMOROSA *42, Brecs.: Rough grassy bank, Cwm Fforch, SO0937, M. Porter, 1982, det. A.J. Silverside. 124/20.7x19. EUPHRASIA NEMOROSA x E. SCOTTICA __*42, Brecs.: Flushes along forestry ride, Garwnant, SN9913, M. Porter, 1989, det. A.J. Silverside. 124/20.8. EUPHRASIA PSEUDOKERNERI *35, Mons.: Turf over Carboniferous Limestone, S of Little Dinham Wood, T.G. Evans, 1994, det. A.J. Silverside. PLANT RECORDS 572 124/20.9x19. EUPHRASIA CONFUSA x E. SCOTTICA *46, Cards.: Flush by drainage adit, Cwmystwyth leadmine, SN810751, S.P. Chambers, 1992, det. A.J. Silverside. *49, Caerns.: Damp montane grassland, Cwm Idwal, SH6458, A.J. Silverside, 1994. 124/20.13x14. EUPHRASIA CAMBRICA x E. OSTENFELDII *49, Caerns.: Rocky ledge, Cwm Idwal, SH6458, A.J. Silverside, 1994. 1st Welsh record. 124/20.14. EUPHRASIA OSTENFELDII *42, Brecs.: Flush by rocky outcrop, Nant Irfon, SN8453, M. Porter, 1995, det. A.J. Silverside. Ist record south of Snowdonia. 124/23.1. PARENTUCELLIA VISCOSA 10, Wight: Over 250 plants in donkey pasture, Wootton Common, $Z533918, C.R. Pope, 1997. New and only extant site. +42, Brecs.: Single plant probably introduced in grass-seed mixture on road verge, near Llyswen, SO141375, M. & C. Porter, 1997. 2nd record. 124/24.1. RHINANTHUS ANGUSTIFOLIUS *11, S. Hants.: Species-rich acidic grassland over gravel/sand surface of former tip area, W of Fort Cumberland, Eastney, Portsmouth, $Z679991, D. P.J. Smith, 1997, det. R.P. Bowman (from colour transparencies). *35, Mons.: On bank damming stream to create a duck pond, Green Meadow Community Farm, $T2994, C. Titcombe, 1992, NMW, det. T.G. Evans, 2nd Welsh record. 124/24.2f. RHINANTHUS MINOR subsp. BOREALIS *49, Caerns.: Moel Siabod, SH75, E. Roberts, 1950s, det. D.J. Hambler. 124/25.2b. PEDICULARIS SYLVATICA subsp. HIBERNICA *41, Glam.: Abundant on Molinia bog, especially in areas burnt two years previously, Broad Pool, Gower, SS5191, E. Lughadha, P. A. Smith, R. Warren & T.C.G. Rich, 1993, NMW. *125/1.2. LATHRAEA CLANDESTINA — _*48, Merioneth: Shoreline of estuary, Dwyryd Estuary, SH589357, R.G. Hughes, 1993, NMW. *5(0, Denbs.: Under rhododendron, Bodnant Garden, SH7982, A. Parry Jones, 1995. 80, Roxburghs.: Under riverside willows, River Tweed, N side below Allan Water, NT525354, J. Murray, 1997. 2nd record. 125/2.10. OROBANCHE MINOR ~~ *50, Denbs.: Bare gravelly ground, Llay, SJ3354, K. Davies, 1996. *+127/1.1. ACANTHUS MOLLIS *49, Caerns.: Large, well-naturalised clump on bridleway, Rhiw, near Aberdaron, SH2328, G. & I. Battershall, 1996. *73, Kirkcudbrights.: Waste ground, Castle Douglas, NX7562, O.M. Stewart & A. White, 1995. 128/1.4. PINGUICULA GRANDIFLORA +*48, Merioneth: Basic boggy rill with P. vulgaris, Brithdir, near Dolgellau, SH71, P.M. Benoit, 1987, NMW. Ist record for Wales. Two plants only in 1983; increased to 11 in 1987. In a completely natural habitat but probably originally planted. 128/2.1. UTRICULARIA VULGARIS 73, Kirkcudbrights.: Abundant in loch by shore, Loch Ken, NX6970, O.M. Stewart, 1996, E. Ist recent record. 129/1.10.; CAMPANULA POSCHARSKYANA *49 Caerns.: Boulders on bank of River Roe, Rowen, Conwy, SH7670, W. McCarthy, 1996. Wall base, Llandudno, SH7882, W. McCarthy, 1996. Ist and 2nd records. *50, Denbs.: Old wall at edge of waste ground,Llanrwst, SH7961, R. Lewis, 1996. 129/1.12. CAMPANULA TRACHELIUM *46, Cards.: Naturalised in thick grass and nettles, Tyglyn Aeron, SN5059, M.L. Lewis, 1908, ABS, conf. A.O. Chater. 129/1.7. CAMPANULA GLOMERATA 51, Flints.: Semi-improved calcareous grassland, Pentre- cwm, SE side of Moel Hiraddug, Dyserth, SJ0678, CCW ‘Phase 1’ survey, 1994. 2nd record. 7129/7.2. LOBELIA ERINUS 42, Brecs.: Disturbed roadside verge beside B4358, Beulah, SN9251, R.G. Woods, 1996. 2nd record. *44, Carms.: River shingle, River Teifi, N of Pentrecwrt, SN392395, G. Hutchinson, 1991. *47, Monts.: Roadside parking place, NW of Talywern, SH820009, C.A.Small, 1989. Stoney beach beside Afon Rhiw, NW of Felindre, SJ160022, E. Roberts & M. Wainwright, 1989. Ist and 2nd records. *49, Caerns.: Waste ground, Llanberis, SH5759, G. Battershall et al., 1996. Pavement crack, Abersoch, SH3128, J. Hawksford, 1997. Ist and 2nd records. 129/7.3. LOBELIA DORTMANNA *94, Banffs.: Pool near Loch Avon, NJO18026, J. Edelsten, 1994, E, det. D. McKean. 580 PLANT RECORDS 7130/5.arv. ASPERULA ARVENSIS *S, Channel Islands: Under bird-table in garden, The Kennel, Rue du Fort, Sark, WV466764, B. Gurden, 1979, JSY, det. D. McClintock. 130/6.3. Galium ULIGINOSUM 46, Cards.: Flushed slope W of Llyn Gwngu, SN/837.728, J.P. Woodman & A.O. Chater, 1997. 2nd extant locality, and the only upland one. *+130/8.tin. RUBIA TINCTORUM *53, S. Lincs.: Growing over a Jurassic limestone wall, Boothby Graffoe, SK9859, P. Porter, 1996, det. I. Weston, conf. E.J. Clement. 131/1.4. SAMBUCUS EBULUS +*46, Cards.: Waste ground, Picton Terrace, New Quay, SN388600, R.E.N. Smith, 1997. Only extant locality. 59, S. Lancs.: Well-established clump, fruiting well on disturbed mesotrophic grass/herb between industrial premises, Knowsley Industrial Park, Kirkby, SJ428972, P. Gateley, 1996. Ist record since 1960. +131/2.3. VIBURNUM TINUS *49, Caerns.: One shrub in hedgerow between Trefriw and Gower Bridge, near Llanrwst, SH7862, R. Lewis, 1997. One shrub in hedgerow of lane near Afon Crafnant, W of Trefriw, SH7763, R. Lewis, 1997. 1st and 2nd records. +131/3.orbxmic. SYMPHORICARPOS ORBICULATUS x S. MICROPHYLLUS (S.x CHENAULTII) *11, S. Hants.: Scrub, in hawthorn, gorse and brambles, Pennington Common, SZ307952, M.E. Young, 1997, det. Everton Nurseries. *39, Staffs.: One prostrate plant on old dumped spoil, W side of Bilston Cosely Road, SO944957, B.R. Fowler & IC. Trueman, 1996, det. D. Gardner. *70, Cumberland: Plantation ride, Green Rigg, Catlowdy, NY485784, R.E. Groom, 1990, LANC. 73, Kirkcudbrights.: Disused quarry, near Hardgate, NX812668, O.M. Stewart, 1995. 2nd & only extant record. 131/4.1. LINNAEA BOREALIS 93, N. Aberdeen: Sitka spruce plantation, previously Scots pine woodland, The Bin, NJ4943, E. Shand, 1989. Ist post-1930 record, confirming 1906 record for this pinewood. 7131/6.3. LONICERA INVOLUCRATA *43, Rads.: Streamside, Davids Well, SO0678, D.R. Humphreys & E.R. Dean, 1996, det. R.G. Woods. +131/6.6. LONICERA JAPONICA *46, Cards.: Scrub by old allotments, Felin-y-mor Road, Aberystwyth, SN581805, A.O. Chater, 1994, NMW. Scrub on riverbank W of Llansanntffraed church, SN511674, A.O. Chater & J.P. Woodman, 1997. Ist and 2nd records. 7131/6.tat. LONICERA TATARICA *17, Surrey: Bush 2m high in woodland strip, Send, TQ026548, J.F. Leshe & K. Page, 1993. Established garden throw-out. 133/1.2. WALERIANELLA CARINATA *108, W. Sutherland: Stony drive to house, Nedd, NC137319, ILM. Evans, 1997. 133/1.3. VALERIANELLA RIMOSA 12, N. Hants.: Over 100 plants in arable field, Bramdown Copse, Overton, SU528471, J.R. Moon, 1994, det. F. Rose. Only extant locality. 4133/2.2. VALERIANA PYRENAICA *91, Kincardines.: Policies of mansion, Glenbervie, NO768805, D. Welch, 1996. 7133/2.phu. VALERIANA PHU 46, Cards.: Roadside hedgebank 100m SW of Neuadd-lwyd chapel, 3-5km SSE of Aberaeron, SN473595, A.O. Chater, 1997, NMW. Roadside bank 850m NNE of Dihewyd, SN491566, A.O. Chater & J.P. Woodman, 1997. Only extant localities in Britain. +134/1.2. DIPSACUS SATIVUS = *51, Flints.: Probable garden escape in waste ground, near car park, Bridge Inn, Pontblyddyn, $J278604, S.P. Chambers, 1997. +134/1.lac. DIPSACUS LACINIATUS *17, Surrey: Over 20 plants in roadside verge, S of Hascombe, TQ001382, J.F. Leslie, K. Page & J.E. Smith, 1993. *44, Carms.: About twelve plants on cinders and ballast by main railway line, Pwll, Llanelli, SN463007, I.K. Morgan, 1997. +135/1.3. ECHINOPS BANNATICUS *73, Kirkcudbrights.: Several clumps in open wood, near Kenmure Castle, NX635764, O.M. Stewart & A. White, 1995. 135/3.2c. ARCTIUM MINUS 73, Kirkcudbrights.: Disused railway depot, E of Lochanhead, NX9171, O.M. Stewart, 1997. 2nd recent record. 135/4.1. SAUSSUREA ALPINA —_*77, Lanarks.: Rocky outcrop, Gana Hill, NS9501, K. Watson, 1997. 135/5.3x4. CARDUUS CRISPUS x C. NUTANS (C. x STANGII) *37, Worcs.: With both parents in horse pasture, Cookley, SO8480, B. Westwood, 1996, conf. R. Maskew & W.A. Thompson. PLANT RECORDS 581 135/11.1. CENTAUREA SCABIOSA = *91,, Kincardines.: Disturbed ground at abandoned railway yard, Crathes, NO7496, D. Welch, 1996. 7135/11.2. CENTAUREA MONTANA *81, Berwicks.: Strong colony established in railway cutting, Lauder, NT498531, M.E. Braithwaite, 1997. 135/11.3xjac. CENTAUREA NIGRA x C. JACEA (C. x MONCKTONII) *61, S.E. Yorks.: Neutral grassland, verge of bypass, near Beverley, TA03, J. Dews, 1997, det. C.A. Stace. 135/16.1b. LEONTODON AUTUMNALIS subsp. PRATENSIS [58, Cheshire: Goyts Moss, SD001718. Delete record in Watsonia 21: 396 (1997). ] 135/16.3. LEONTODON SAXATILIS 95, Moray: Grassy road verge, N verge of A96, W of Elgin, NJ166631, P.R. Green, 1996. 2nd record. 135/18.1. SCORZONERA HUMILIS *41, Glam.: Flushed marshy grassland, Cefn Cribwr Meadows SSSI, SS8583, J.P. Woodman, 1996, conf. P.D. Sell & A.O. Chater. Single patch in species-rich meadow at Little Hills Farm, $S539928, J.P. Woodman, 1997. 1st and 2nd records. 135/19.1b. TRAGOPOGON PRATENSIS subsp. MINOR *79, Selkirks.: Edge of path, near Gala Water & College of Textiles, Galashiels, NT507353, M. Little, 1994. Ist localised record. 135/21.4gla. SONCHUS ASPER subsp. GLAUCESCENS *46, Cards.: Roadside by Teifi estuary, Nantyferwig, Gwbert, SN166483, A.O. Chater, 1995. Disturbed ground by supermarket development, Parc-y-llyn, Llanbadarn Fawr, SN593806, A.O. Chater, 1995. 1st and 2nd records. +135/23.3. CICERBITA PLUMIERI *95, Moray: Many plants on road verge and open woodland, Blackhills near Fochabers, NJ276585, I.P. Green, 1996. 135/28.62. TARAXACUM FULGIDUM *42, Brecs.: Meadow, Llangasty Tal-y-Llyn, SO1225, A. J. Richards et al., 1996. 135/25.65. TARAXACUM HAEMATICUM *35, Mons.: Short turf over coal waste, Cum Du, SO2502, T.G. Evans, 1997, NMW, det. A.J. Richards. 135/25.93. TARAXACUM LANCIDENS *42, Brecs.: Meadow, Llangasty Tal-y-Llyn, SO1225, A. J. Richards et al., 1996. 135/25.96. TARAXACUM PRIONUM *42, Brecs.: Lane verge, grassy bank, 3km N of Ponsticill, SO0514, M. Porter, 1995, det. A.J. Richards. 135/25.124. TARAXACUM CURTIFRONS *42, Brecs.: Bank of old lane near Llanbedr, S$O0225193, M. Porter, 1996, herb. M. P., det. A.J. Richards. 135/25.125. TARAXACUM CYANOLEPIS *35, Mons.: River Wye meadow covered by Spring tides, N of Chepstow Castle, ST531951, T.G. Evans, 1997, NMW, det. A.J. Richards. 135/25.134. TARAXACUM EXSERTIFORME *35, Mons.: Forest road, N of Howick, ST5095, Hans @llg. et al., 1996. 135/25.140. TARAXACUM HORRIDIFRONS *35, Mons.: Roadside, Cleddon, SO5103, International Taraxacum Group, 1996, det. A.J. Richards, Hans @llg., et al. 135/25.159. TARAXACUM LUCIDUM _ *35, Mons.: Shady bank, Caerllan, SO4982, International Taraxacum Group, 1996, det. A.J. Richards, Hans Ollg., et al. | *42, Brecs.: Pasture at edge of Llangorse Lake, Llangasty Tal-y-Llyn, SO1326, M. Porter, 1995, det. A.J. Richards. 135/25.170. TARAXACUM NITIDUM ~~ *35, Mons.: Shady bank, Caerllan, SO4908, International Taraxacum Group, 1996, det. A.J. Richards, Hans @llg., et al. 135/25.173. TARAXACUM OBTUSIFRONS *42, Brecs.: River bank, Llangynidr, SO1420, A.J. Richards et al., 1996. 7135/25.189. TARAXACUM PSEUDORETROFLEXUM *35, Mons.: Cleddon Lane verge, Cleddon, $O5103, T.G. Evans, 1996. 135/25.199. TARAXACUM SINUATUM *61, S.E. Yorks.: Neutral grassland, Hollym Cross, TA3325, P.J. Cook, 1997, det. A.J. Richards. 135/25.215. TARAXACUM TUMENTILOBUM *35, Mons.: Roadside bank, New House Roundabout, M48, ST5391, T.G. Evans, 1996, det. A.J. Richards. 135/25.bre. TARAXACUM BRECONENSE *35, Mons.: Short turf over coal waste, Cwm Du, $O02502, T.G. Evans, 1997, NMW, det. A.J. Richards. 582 PLANT RECORDS 135/25.edm. TARAXACUM EDMONSONIANUM *42, Brecs.: Bank of old lane, Llangynidr, SO1419, M. Porter, 1996, det. A.J. Richards. 135/25.ope. TARAXACUM OPERTUM H.@llg. ined. *35, Mons.: Forest roadside, N of Howick, ST5095, Hans Ollg. et al., 1996. 135/26.3. CREPIS BIENNIS *46, Cards.: Abundant on roadside and field banks 500m SE of Glynarthen, SN313482, A.O. Chater, 1997, NMW. 47, Monts.: Semi-improved neutral grassland, Gweunydd Ty-Brith SSSI and Nature Reserve, Deuddwr, SJ243178, S.L. Smith & D. Guest, 1997. 2nd record. 80, Roxburghs.: One plant on road verge, Threepwood Bridge, NT519441, L. Gaskell, 1997, herb. M.E. Braithwaite. Ist record since 1878. +*81, Berwicks:: Perhaps established on road verge near old quarry, Threepwood Bridge, NT519441, L. Gaskell. 1997, herb. M.E.B. +135/26.set. CREPIS SETOSA *35, Mons.: Near meadow boundary fence, near Llanfoist, $0300119, T.G. Evans, 1991, NMW, det. R. Fraser. 7135/27.2x6. PILOSELLA OFFICINARUM x P. AURANTIACA (P. x STOLONIFLORA) *4 oN. Devon: Grass, Parracombe churchyard, $S669448, W.H. Tucker, 1997, herb. L.J.M.. conf. L.J. Margetts. *46, Cards.: E part of chapel graveyard, Tan-y-groes, SN284493, A.O. Chater, 1997, CGE, det. P.D. Sell. This population agrees with subsp. schuriana Naegeli & Peter and appears to be the first record for Britain for this nothomorph. +135/27.3a. PILOSELLA FLAGELLARIS subsp. FLAGELLARIS *80, Roxburghs.: Old road, now gated off, near A7 N of Galashiels, NT476378, J. Murray, 1995, herb. R.W.M. Corner, det. D.R. McKean. *$1, Berwicks.: Established on road verge, Al near Grantshouse, NT814655, M.E. Braithwaite, 1996. 135/28.3. HIERACIUM RIGENS *37, Worcs.: Occasional in sandy heathland, Devil’s Spittleful, SO08074, SO8174, SO8175, J. Bevan & W.A. Thompson, 1991. 135/28.6a. HIERACIUM UMBELLATUM subsp. UMBELLATUM *49, Caerns.: Side of disused railway, Caernarfon, SH4863, G. & L Battershall, 1996. 135/28.21. HIERACIUM TRICHOCAULON *37, Worcs.: Small colony in sandy heathland, Devil’s Spittleful, SO8174, J. Bevan & W.A. Thompson, 1993. 135/28.34. HIERACIUM SPARSIFOLIUM *47, Monts.: Rocky banks of Afon Trannon, E of Staylittle, SN9193, A.J. Morton, 1987, conf. J. Bevan. 135/28.67. HIERACIUM SUBAMPLIFOLIUM *37, Worcs.: Stone walls, Croft Bank, West Malvern, SO7646, J. Bevan, P.G. Garner & W.A. Thompson, 1990. 135/28.78. HIERACIUM ORCADENSE *91, Kincardines.: Outcrop of Old Red Sandstone, Dunnottar Castle, NO878838, D. Welch, 1995, herb. D.W., det. D.J. McCosh. 135/28.115. HIERACIUM SCOTOSTICTUM *37, Worcs.: Sandy laneside bank, Upper Rochford, SO06367, R. Maskew & W.A. Thompson, 1994, det. J. Bevan. 135/28.123. HIERACIUM STENSTROEMII *37, Worcs.: Small colony in woodland, Armley Bank, Broadway Hill, SP1136, J. Bevan & W.A. Thompson, 1994. 135/28.133. HIERACIUM SUBCRASSUM _*37, Worcs.: Broken wooded slope, The Dingle, West Malvern, SO7645, J. Bevan, P.G. Garner & W.A. Thompson, 1996, det. P.D. Sell. Fourth British record. 135/28.140. HIERACIUM EXOTERICUM *47, Monts.: Large population with other members of this section, railway cutting, S of Talerddig, SN9399, A.J. Morton, 1989. 135/28.147. HIERACIUM CALEDONICUM *94, Banffs.: Serpentine outcrop, Craigs of Succoth, NJ435363, D. Welch, 1992, ABD, det. D. McCosh. 135/28.165. HIERACIUM LASIOPHYLLUM *44, Carms.: One plant only on acid crag, Craig Allt- y-berau, SN7746, J. Bevan, R.D. Pryce & G. Hutchinson, 1995. 135/30.1. FILAGO VULGARIS 46, Cards.: Dry bank by road, RAE Site, Aber-porth, SN251516, A.O. Chater, 1997. 2nd extant locality. 135/30.4. FILAGO MINIMA _ 45, Pembs.: Hundreds to low 1,000s of plants scattered in short open turf on small stones, Trecwn Burning Ground, Llanychaer, SM9934, S.B. Evans et al., 1996. 2nd record. 64, Mid-W. Yorks.: Hundreds of plants on floor of old sandstone quarry, Ireland Woods, Leeds, SE254382, M. Wilcox, 1997. Ist record this century. ————— PLANT RECORDS 583 135/33.6. GNAPHALIUM LUTEOALBUM *15, E. Kent: Sandy area at edge of newly dug pits, Dungeness, TRO618, E.G. Philp, 1996, MNE. In 1997 several hundred plants & two further colonies were present in the area. +*95, Moray: One plant on waste ground on Gallow Hill, Hopeman, NJ149686, P.R. Green, 1997, herb. I.P. Green, conf. E.J. Clement. 7135/40.4. SOLIDAGO GIGANTEA *49, Caerns.: Village green (jettisoned garden relic?), Bryncroes, SH2231, J.P. Bailey & A.P. Conolly, 1985, NMW. +135/40.5. SOLIDAGO GRAMINIFOLIA *42, Brecs.: Naturalised on waste ground, Brynmawr, S$O185116, M. Porter, 1997. +135/41.3x4. ASTER LAEVIS x A. NOVI-BELGII (A. x VERSICOLOR) *46, Cards.: Saltmarsh by Afon Rheidol, Aberystwyth, SN584811, A.O. Chater, 1993, CGE, conf. P.D. Sell & P.F. Yeo. *49, Caerns.: Edge of neglected car park, Pwllheli, SH3734, J. Hawksford, 1997. *77, Lanarks.: Well established in scrubby wood, Gartsherrie, NS7266, P. Macpherson, 1996, herb. P.M., det. P. Feo: +135/41.4. ASTER NOVI-BELGI 81, Berwicks.: Established in ditch by road, A1107, near Old Cambus Wood, NT8368, M.E. Braithwaite, 1996. Large colony on road verge, A68 near Lauder Barns, NT547463, M.E. Braithwaite, 1997, herb. M.E.B. Ist and 2nd post-1930 records and only extant localities. *4135/41.4x5. ASTER NOVI-BELGI x A. LANCEOLATUS (A. x SALIGNUS) *44, Carms.: Roadside bank, Railway station approach, Kidwelly, SN401065, G. Hutchinson, 1995, NMW, det. P.F. Yeo. Ist authenticated record. +135/41.5. ASTER LANCEOLATUS *49, Caerns.: Scrub, edge of river, Pwllheli, SH3734, J. Hawksford, 1997. +135/41.con. ASTER CONCINNUS *46, Cards.: NE bank of Afon Cletwr, Tre’r-ddol, SN654929; N bank of Afon Rheidol, Aberystwyth, SN587812; both A.O. Chater, 1990, CGE & NMW,, det. P.F. Yeo & P.D. Sell. Ist and 2nd records. + 135/43.1. ERIGERON GLAUCUS *49, Caerns.: Naturalised at side of track near Aberdaron, SH1626, G. & I. Battershall, 1996. Roadside verge, near Penmaenmawr, SH7176, G. & I. Battershall, 1997. 7+135/43.4. ERIGERON KARVINSKIANUS 50, Denbs.: Lowest W facing crags above woodland, W side of Bryn Euryn, Colwyn Bay, SH8379, R. Lewis, 1993, det. Mrs R. Lever. [The determiner was eroneously given as Mrs R. Lewis in Watsonia 20(3): 298. ] 7135/44.1. CONYZA CANADENSIS *95, Moray: Two plants in set-aside field, Roseisle near Duffus, NJ156668, I.P. Green, 1996. Many plants on waste ground on new housing estate, Forres, NJ033588, P.R. Green, 1997. 1st and 2nd records. 7135/44.2. CONYZA SUMATRENSIS *12, N. Hants.: Rough grass near A331, Spring Lakes, Aldershot, SU886516, C.R. Hall, 1997, herb. A. Brewis. 7135/45.2. OLEARIA MACRODONTA _*49, Caerns.: Probably planted originally on side of track, Machroes near Abersoch, SH3226, G. & I. Battershall, 1997. 7135/48.2. TANACETUM MACROPHYLLUM _*77, Lanarks.: Well established between railings of Botanic Gardens & River Kelvin, Kelvinside, Glasgow, NS5667, K. Watson, 1996, GLAM, det. E @lement. 135/54.1. CHAMAEMELUM NOBILE 10, Wight: Non-flowering but frequent in mown amenity grassland, Puckpool Park, Ryde, SZ613922, C.R. Pope, 1996, herb. C.R.P. 2nd extant site. 7135/57.1. LEUCANTHEMELLA SEROTINA 12, N. Hants.: Bridleway called Alresford Lane, Cheriton, SU600292, R.M. Veall, 1996. 2nd record. 7135/58.lacxmax. LEUCANTHEMUM LACUSTRE x L. MAXIMUM (L. x SUPERBUM) *81, Berwicks.: Large colony established on road verge, A697 E of Fireburnmill, NT824393, M.E. Braithwaite, 1997. 135/60.1. TRIPLEUROSPERMUM MARITIMUM _ 47, Monts.: Saltmarsh, Morben-isaf, SN7098, T. Kohler, T. Teearu & E. Lomas, 1988. 2nd record. 135/60.1x2. TRIPLEUROSPERMUM MARITIMUM x T. INODORUM *51, Flints.: Behind coastal dunes, Talacre Warren, SJ1184, BSBI group, 1996. 584 PLANT RECORDS +135/61.1. COTULA CORONOPIFOLIA 11, S. Hants.: Sandy shingle bank S side of Hermitage Stream, Bedhampton Mill, SU704062, J.R.W. Hollins, 1997, herb. R.P.B., conf. R.P. Bowman. 2nd record. +135/62.1. SENECIO CINERARIA *81, Berwicks.: Established at field edge, Burnmouth, NT955611, J. Muscott, 1997. 135/62.10x11. SENECIO JACOBAEA x S. AQUATICUS (S. x OSTENFELDII) 50, Denbs.: Rough field, Cerrig-y-druidion, SH952484, J.A. Geen, 1997. 2nd record. 135/62.12. SENECIO ERUCIFOLIUS *48, Merioneth: One plant with S. jacobaea, roadside bank, near Trawsfynydd, SH7037, P.M. Benoit et al., 1994, NMW. Ist certain record. 135/62.13x15. SENECIO SQUALIDUS x S. VULGARIS (S. x BAXTERI) *46, Cards.: Garden, Llanbadarn Fawr, SN598810, J.H. Salter, 1935, (Fl. Pl. Ferns Cardiganshire (1935)). +135/62.3. SENECIO INAEQUIDENS _*77, Lanarks.: Unused cul de sac, King George V Dock, Glasgow, NS5366, P. Macpherson, 1995, herb. P.M., det. R.M. Burton. One plant 1995, three plants 1996 and 1997. +135/66.comxlax. BRACHYGLOTTIS COMPACTA x B. LAXIFOLIA (B. ‘SUNSHINE’) *46, Cards.: Gully on sea cliffs below Grogal, W of New Quay, SN373593, A.O. Chater & A.P. Fowles, 1989, BM, conf. D.H. Kent. *77, Lanarks.: Grassy, open area near river, Meadowside, Glasgow, NS5566, P. Macpherson, 1988, herb. P.M., det. D. McClintock. +135/69.1x2x3. DORONICUM PARDALIANCHES x D. PLANTAGINEUM x D. COLUMNAE (D. x EXCELSUM) *67, S. Northumb.: Roadside verge, Acumb Fell, NY955675, G.A. Swan, 1996, herb. G.A.S., det. P S. Green. 7135/71.2. PETASITES JAPONICUS *S, Channel Islands: Small shaded valley beside stream, Harbour Hill, Sark, WV473758, R.M. Veall, 1988, det. D. McClintock. Apparently planted — seen in 1988 by Botanical Group from Guernsey — still present and spreading. *135/73.1. CALENDULA OFFICINALIS 31, Flints.: Waste ground, W end of Marsh Road, Rhyl, SJ002803, J. Phillips, 1997. 2nd record and Ist since 1942. *135/73.1. CALENDULA OFFICINALIS *43, Rads.: Spoilheap in lay-by at roadside, between Newbridge & Disserth, SO030589, R.G. Woods, 1980s, NMW. *135/78.1. RUDBECKIA HIRTA *41, Glam.: Coal dump, Cardiff Docks, ST197749, G. Hutchinson, 1988, NMW. +135/79.1. HELIANTHUS ANNUUS *49. Caerns.: Demolition site, Llandudno, SH784821, W. McCarthy, 1995. +135/79.2xrig. HELIANTHUS TUBEROSUS x H. RIGIDUS (H. x LAETIFLORUS) *81, Berwicks:: Over 2m up wall, Lauder, NT528477, M.E. Braithwaite, 1997. +135/81.3. BIDENS CONNATA var. ANOMALA _*58, Cheshire: Canal bank, Sandbach, $J732505, G.M. Kay, 1997, herb. G.M.K., conf. E.J. Clement. Canal bank, Hassall Green, SJ778584, G.M. Kay, 1997. 1st and 2nd records. 135/SCO.his. SCOLYMUS HISPANICUS *46, Cards.: Two plants by waste ground, N side of Afon Rheidol, Aberystwyth, SN586811, R. Birch, 1997. 138/2.1. STRATIOTES ALOIDES *5, S. Somerset: Many plants in pond in Wayford Woods, Wayford, ST399064, I.P. Green, 1997. +*35, Mons.: Planted in artificial pond in Water Park Pond, near Ahtophon Factory, St Mellons, ST249815, G. Hutchinson, 1994. Pond in Industrial Park, Pwll-mawr, ST223787, T.G. Evans, 1996. 1st records. A series of ponds and lakes have been created between Newport and Cardiff with some strange plantings. +138/4.2. ELODEA NUTTALLII *49, Caerns.: Margin of Llyn Padarn, Llanberis, SH5662, A. Jones, 1997. *95, Moray: Very plentiful in pond, Sunbank Park, Lossiemouth, NJ231694, I.P. Green, 1996. +138/6.1. LAGAROSIPHON MAJOR *46, Cards.: In two ponds in grounds of Ffynnon Berw, lkm NNW of Ffynnonddewi, SN381538. A.O. Chater, 1994. Upland ponds 2km NNE of Ffair- thos, SN750697, M.D. Sutton, 1997. 1st and 2nd records. 141/1.1. TRIGLOCHIN MARITIMA 47, Monts.: Dovey Junction, SN7098, M. Hogan & C. Fieldhouse, 1986. 2nd record. PLANT RECORDS 585 142/1.3. POTAMOGETON COLORATUS *44, Carms.: Pond in sand dunes created about 5 years previously, Pembrey Country Park, SN3900, N.F. Stewart, 1996. 142/1.5x6. POTAMOGETON LUCENS x P. GRAMINEUS (P. x ZIZII) *41, Glam.: SW part of Kenfig Pool, SS796812, N.F. Stewart, 1996. 142/1.6x9. POTAMOGETON GRAMINEUS x P. PERFOLIATUS (P. x NITENS) *41, Glam.: Pool on Margam Dunes, SS782838, N.F. Stewart, 1996. 142/1.7. POTAMOGETON ALPINUS 81, Berwicks.: Pond, Eden water at Nenthorn House, NT673371, M.E. Braithwaite, 1995, herb. M.E.B., det. N.T.H. Homes. Large colony in slow moving burn in fen, Lauder Burn, NT516452, M.E. & P.F. Braithwaite, 1997, herb. M.E.B. Ist and 2nd recent records and only extant localities. 142/1.16. POTAMOGETON TRICHOIDES *4, N. Devon: Portledge Pool, SS392245, W.H. Tucker, 1996, conf. L.J. Margetts. 142/1.19. POTAMOGETON CRISPUS +*46, Cards.: Pond in pasture 200m N of Clogfryn, Aberaeron, SN449624, A.O. Chater, 1996, NMW, conf. C.D. Preston. 142/1.20x21. POTAMOGETON FILIFORMIS x P. PECTINATUS (P. x SUECICUS) 93, N. Aberdeen: Shallow water, Loch of Strathbeg, NK0858, C.D. Preston & P.M. Hollingsworth, 1994. Ist post- 1930 record. 143/1.1. RUPPIA MARITIMA *47, Monts.: Many plants found in four different brackish pools in salt-marsh, Dovey Junction, SN695985, F. Evans, E. Gwynn & A. Law, 1993. +147/1.1. ACORUS CALAMUS 4, N. Devon: Pond edge, Simpson, Holsworthy, SS364038, W. H. Tucker, 1997. 2nd post-1930 record. +147/2.1. LYSICHITON AMERICANUS =*12, N. Hants.: Riverside Railway Walk, River Rother, Liss, SU780283, J. Ockenden, 1997, det. F. Rose. *91, Kincardines.: Roadside ditch, Durris, NO785968, D. Welch, 1997. Probably escaped from Durris House, about 1km away. Shingle terrace on island in River Dee, Maryculter, NO815983, D. Welch, 1997. Ist and 2nd records. +147/5.2b. ARUM ITALICUM subsp. ITALICUM _*41, Glam.: Caswell, SS5897, A.S. Lewis 1997. 148/1.1. SPIRODELA POLYRHIZA 41, Glam.: Abandoned Carboniferous Limestone sheep-dip, Tythegston, SS864781, P.S. Jones & S. Thomas, 1997. Abundant in a concrete drinking-pond in the SW corner of a field near Tresilian, SS953680, Q.O.N. Kay, 1997. Ist recent records. 148/2.3. LEMNA TRISULCA *45, Pembs.: Abundant with small patch of Crassula helmsii in shady shallow pond on slope in woodland, Pwll Dyfrig, Glyn-y-mel, Lower Fishguard, SM967371, S.B. Evans et al., 1997. +148/2.4. LEMNA MINUTA —_ *41, Glam.: Locally abundant, with L. minor, in shallow water in moat & fish pond in Botanic Garden, University College of Swansea, SS6291, Q.O.N. Kay, 1996. 46, Cards.:. Garden pond, Old Castle Farm, Cardigan, SN165463, A.O. Chater, 1996. 2nd record 151/1.13x14. JUNCUS ARTICULATUS x J. ACUTIFLORUS (J. x SURREJANUS) *71, Man: Isolated pond, Ballavarran Dub, SC365972, B.A. Tregale, 1999, DGS. 151/1.25x26. JUNCUS EFFUSUS x J. INFLEXUS (J. x DIFFUSUS) *50, Denbs.: Wet ground near lake, Painters Green, SJ502499, J.A. Green, 1997, NMW. 151/1.6. JUNCUS FOLIOSUS *35, Mons.: Wet meadow, Ty’r Sais Farm, near Pen-y-fan Pond, SO1800, T.C.G. Rich, P.A. Smith & S.E. Erskine, 1996. *108, W. Sutherland: Wet trackside, Aultanrynie, Loch More, NC349356, C. Dixon, 1997. 151/1.8. JUNCUS AMBIGUUS 47, Monts.: Frequent in upper saltmarsh, Dovey Junction, with J. bufonius, SN695983, P.M. Benoit, 1987. 2nd record. *51, Flints.: Damp depression in disturbed dunes, Point of Ayr, SW of lighthouse, SJ1185, A.O. Chater, 1996, det. T. Cope. *151/2.4. LUZULA LUZULOIDES 81, Berwicks.: Open woodland, Castle hill, Thirlstane Castle, NT534477, M.E. & P.F. Braithwaite, 1997. Only extant locality. 152/11.1. CYPERUS LONGUS +*44, Carms.: SW corner of lake margin, Sandy Water Park, Llanelli, SN493004, I.K. Morgan, 1995. Source unknown, lake constructed c.5 years ago. 152/16.1x15. CAREX PANICULATA x C. REMOTA (C. x BOENNINGHAUSIANA) *80, Roxburghs.: Side of ditch, Linton Loch, Morebattle, NT793254, O.M. Stewart, 1997, E. 586 PLANT RECORDS 152/16.5. CAREX OTRUBAE 109, Caithness: Coastal turf, near Rattar, ND2574, R.E.C. Ferreira, 1995. 1st localised record. 152/16.5x15. CAREX OTRUBAE x C. REMOTA (C. x PSEUDOAXILLARIS) *45, Pembs.: Grassy verge of track (re-surfaced for motors) with C. remota, Cardigan Wildlife Park (entrance road from Cilgerran), SN185436, R.W. David, 1978, NMW, conf. A.O. Chater. The second parent (C. otrubae) is frequent on the banks of the River Teifi nearby. 152/16.9b. CAREX DIVULSA subsp. LEERSII *73, Kirkcudbrights.: Few plants on bank of waste ground and rough grassland, Castle Douglas, NX750624, O.M. Stewart, 1996, E, det. D.A. Pearman & A.O. Chater. 81, Berwicks.: One large clump in flower under beech, The Lees, NT/840390, M.E. & P.F. Braithwaite, 1997, herb. M.E.B. 2nd extant record. Status unknown. 152/16.21. CAREX LACHENALII *94, Banffs.: Cairngorms, NH994010, J. Edelsten, 1996, E, det. A.C. Jermy. 152/16.25x26 CAREX ACUTIFORMIS x C. RIPARIA (C. x Sool) *25, E. Suffolk: Large stand, Framlingham Mere, TM285637, A.C. Jermy, 1990. Ist record for Suffolk. 152/16.27. CAREX PSEUDOCYPERUS *49, Caerns.: Fen, Cors Llyferin S of Abersoch, SH3126, D. Jones, 1995, NMW. 152/16.28x29. CAREX ROSTRATA x C. VESICARIA (C. x INVOLUTA) *80, Roxburghs.: With both parents in fen, Tilery Moss, SW of Muirhouse Law, NT623274, R.W.M. Corner, 1997, herb. R.W.M.C., det. A.O. Chater. 152/16.31. CAREX PENDULA 81, Berwicks.: Spring in former woodland, Eye Water above Grantshouse, NT810650, M.E. Braithwaite, 1996. 2nd extant record. 91, Kincardines.: Marsh in woodland, Maryculter, NJ860999, D. Welch, 1997. Ist definite post-1930 record. 152/16.39. CAREX LAEVIGATA *89, E. Perth: Bridge of Cally, NO15, J.H. Penson, 1960, herb. J.H.P., conf. A.McG. Stirling. 152/16.44x46. CAREX HOSTIANA x C. VIRIDULA (C.x FULVA) 83, Midlothian: Marsh, S of Gladhouse Reservoir, NT296502, C. Dixon, 1997, E, det. D.R. McKean. 2nd record. 152/16.44x46a. CAREX HOSTIANA x C. VIRIDULA subsp. BRACHYRRHYNCHA (C. x FULVA) *79, Selkirks.: With both parents in basic flush, Roughhope Burn, Philhope, Borthwick Water, NT383078, R.W.M. Corner, 1997, herb. R.W.M.C., det. A.O. Chater. *80, Roxburghs.: With both parents in basic flush, Grindstone Burn, Upper Hindhope, Kale Water, NT768086, R.W.M. Corner, 1997, herb. R.W.M.C., det. A.O. Chater. 152/16.59a. CAREX MAGELLANICA subsp. IRRIGUA *49, Caerns.: Few plants in swamp with Carex rostata, Mignient near Ysbyty Ifan, SH7744, G. Battershall, 1996. 152/2.2b. TRICHOPHORUM CESPITOSUM subsp. CESPITOSUM *108, W. Sutherland: Schoenus nigricans mire, Ceathramh Garbh, Rhiconich, NC228515, R.W.M. Corner, 1997, conf. G.A. Swan. 152/2.2axb. TRICHOPHORUM CESPITOSUM subsp. CESPITOSUM x subsp. GERMANICUM (proliferous form) *46, Cards.: Small flush with Carex dioica, Rhynchospora alba, etc., Bryn Mawr, SE of Hafod, SN771717, A.O. Chater, 1963, NMW, det. G.A. Swan. 153/12.9. FESTUCA VIVIPARA *93, N. Aberdeen: Trackside in heather moorland, Hill of Millmedden, NJ5323, D. Welch, 1997, ABD, conf. C.A. Stace. 153/12.10. FESTUCA FILIFORMIS 47, Monts.: Hummocks in boggy field, NE of Llanfihangel- yng- ngwynfa, SJO915, P.M. Benoit, 1989. 2nd record. 153/12.4. FESTUCA ALTISSIMA .*43, Rads.: In quantity on rock outcrops & decaying wood, Bach Howey Gorge, $SO119435, R.G. Woods & M. Porter, 1980s. (81, Berwicks.: Cromwells. Brunta Burn, NT596505, M.E. Braithwaite, 1992, det. D.R. McKean. Delete record in Watsonia 19: 293 (1993). The plant has flowered and is Elymus caninus, M.E. Braithwaite, 1995, det. D.R. McKean. ] *153/12.5. FESTUCA HETEROPHYLLA *28, W. Norfolk: Edge of woodland ride, West Briggs, TF656105, S. Fisher, 1997, conf. R.M. Payne. *95, Moray: Well naturalised in grassy area under trees, in grounds of small chapel at Gordonstown School, Duffus, NJ192689, I.P. Green, 1996, conf. R.M. Burton. PLANT RECORDS 587 153/12.6. FESTUCA ARENARIA *44, Carms.: Sparse population on seaward Ammophila arenaria ridge, Pendine Ranges, Laugharne Burrows, SN311071, A.O. Chater, 1997, NMW. 153/12.7b. FESTUCA RUBRA subsp. JUNCEA *46, Cards.: Storm beach W of road, Tan-y-bwlch Beach, Aberystwyth, SN579805, S.P. Chambers, 1994, herb. S.P.C., det. C.A. Stace. Rocky sea cliff, NNW side of Foel y Mwnt, SN192521, A.O. Chater, 1994, NMW. Ist and 2nd records. *49_. Caerns.: Cliff-top turf, Porth Ceriad, near Abersoch, SH3124, M. & A. Atkinson, 1997. 153/12.7c. FESTUCA RUBRA subsp. LITORALIS *46, Cards.: Saltmarsh E of Afon Leri, Ynys- las, SN616930, A.O. Chater, 1993. Saltmarsh, Teifi estuary, 1km S of Nantyferwig, SN168480, A. O. Chater, 1997, NMW. Ist and 2nd records. 153/12.7d. FESTUCA RUBRA subsp. COMMUTATA *44, Carms.: Llansteffan Dunes, Llansteffan, SN31, H. Sealy-Lewis, 1987. 153/12.7g. FESTUCA RUBRA subsp. MEGASTACHYS *44, Carms.: Roadside verge, near Edwinsford, SN635345, A.O. Chater, 1995. Roadside near Maesteilo, Capel Isaac, SN592265, J. A. Green, D.M. George & E. Stephenson, det. A.O. Chater, 1995, NMW. Ist and 2nd records. *46, Cards.: Grassy slope in village, Ponterwyd, SN751807, A.O. Chater, 1988, NMW, det. A.K. Al-Bermani. Laneside verge just N of Pont Pant-mawr, 2km WNW of Llanilar, SN607757, A.O. Chater, NMW, 1992. Ist and 2nd records. 153/12.8c. FESTUCA OVINA subsp. OPHIOLITICOLA *44, Carms.: Rocky calcareous grassland, Pal-y-Cwrt, SN678182, A.O. Chater et al., 1997. 153/12.9. FESTUCA ‘VIVIPARA *93, N. Aberdeen: Trackside in heather moorland, Hill of Millmedden, NJ530230, D. Welch, 1997, ABD, conf. C.A. Stace. 7153/12.gau. FESTUCA GAUTIERI *3, S. Devon: Road embankment, Exeter, SX926927, R. Takagi-Arigho, 1990, herb. R.T.-A., det. T.A. Cope. 153/12.1x13.1. FESTUCA PRATENSIS x LOLIUM PERENNE (x FESTULOLIUM LOLIACEUM) *50, Denbs.: Wet meadows, near Overton, SJ396425, D. Tinston, 1987, NMW, det. P.M. Benoit. 153/12.2x13.1. FESTUCA ARUNDINACEA x LOLIUM PERENNE (x FESTULOLIUM HOLMBERGII) *28, W. Norfolk: Wiggenhall St Mary, TF575142, R.M. Payne, 1992, conf. R.M. Payne & A. Copping. 2nd record for Norfolk. 153/12.2x13.2. FESTUCA ARUNDINACEA x LOLIUM MULTIFLORUM *11, S. Hants.: Bishop’s Palace, Bishop’s Waltham, SU553175, P.D. & C.A. Stanley, 1997, herb. R.P. Bowman, det. E.J. Clement. 153/12.7x14.3. FESTUCA RUBRA x VULPIA MYUROS 35, Mons.: By railway line, Mitchell Troy, SO4910, E. Nelmes, 1951, K, det. C.A. Stace. Precedes TGE’s record of 21/7/1988 as Ist record. C.A. Stace writes ‘This .... is the earliest from any part of the world.’ 153/14.1. VULPIA FASCICULATA *51, Flints.: Fixed sand dunes, Point of Ayr, $J123851, J.A. Green, 1985, NMW, det. G. Hutchinson. 153/14.3. VULPIA MYUROS *89, E. Perth: Waste ground, Invergowrie, NO349300, A.A. Slack, 1980. Virtually on boundary between v.cc 89 & 90. Apparently Ist localised record. 153/16.2. PUCCINELLIA DISTANS +*77, Lanarks.: Verge of A73, N of Newmains, NS8256, herb. M.E.B. Road verge of A74, Melbourne, NT0844; both M.E. Braithwaite, 1996. 1st and 2nd records. 153/16.2b. PUCCINELLIA DISTANS subsp. BOREALIS *95, Moray: Road verges, Speybridge, NJ040261, P.R. Green, 1997, conf. H. Bowen. Plentiful along many of the roads and also on saltmarshes; greatly on the increase. 153/18.4. POA HUMILIS *61, S.E. Yorks.: On old brick wall, Sledmere, SE9364, M.E. Braithwaite, 1997. On old brick wall, Patrington, TA3122, P.J. Cook, 1997. 1st and 2nd records. Subsequently found on other brick walls in Holderness. 7153/18.7. POA CHAIxIE *43, Rads.: Among trees, Rock Park, Llandrindod Wells, SO055609, R.G. Woods, 1986. 47, Monts.: Park woodland under beech, Berriew, SJ1900, E. Roberts & M. Wainwright, 1989. 2nd record; possibly planted for pheasant cover. 153/20.1. CATABROSA AQUATICA 91, Kincardines.: Muddy pool close to HWM, Nigg Bay, NJ967044, D. Welch, 1997, ABD. Ist post-1930 record. 588 PLANT RECORDS 153/21.2. CATAPODIUM MARINUM *35, Mons.: On exposed former shelly beach on River Severn bank, Rumney Great Wharf, ST248786, P. Millman, et al., 1997. Though looking native in a suitable habitat, much material has been imported to build up nearby sea wall and status is uncertain. 153/22.1. SESLERIA CAERULEA *35, Mons.: Chepstow, ST59, T.G. Evans, 1988, NMW. 153/24.1. GLYCERIA MAXIMA 46, Cards.: Dominant in pond in fields 300m SSW of Coed-y- cwm, Cwm Cou, SN275428, A.O. Chater, 1996. 2nd record. +153/28.2. AVENA FATUA *46, Cards.: Several plants on disturbed ground, Ffair-Rhos, SN743679, A.O. Chater, 1991, NMW. +153/29.1. GAUDINIA FRAGILIS 11, S. Hants.: Horse pasture in closed sward of fairly rich area of grassland, on clay, Hounsdown, Eling, SU353117, J. Rowe, 1997, herb. R.P.B., conf. R.P. Bowman. 2nd record. +153/30.1pur. TRISETUM FLAVESCENS subsp. PURPURASCENS *46, Cards.: Grassy slope by pond, Gwern-medd, Llangoedmor, SN213454, A.O. Chater & L. Gander, 1997, NMW. +153/37.2. ANTHOXANTHUM ARISTATUM *12, N. Hants.: Roadside B2070, Petersfield, SU752240, P. Stanley, 1994, herb. E.J.C., det. E.J. Clement. Also present in 1995. 153/39.1x4. AGROSTIS CAPILLARIS x A. STOLONIFERA (A. x MURBECKII) 46, Cards.: River shingle by Afon Ystwyth downstream of road bridge, Llanfarian, SN589777, S.P. Chambers, 1997. 2nd record. 153/40.2. CALAMAGROSTIS CANESCENS 80, Roxburghs.: Edge of fen, Linton Loch, Morebattle, NT794255, R.W.M. Corner, 1997, herb. R.W.M.C., 2nd extant locality. +153/43.1. LAGURUS OVATUS *49, Caerns.: Waste ground, Pwllheli, SH364341, G. Battershall & W. McCarthy, 1996. 153/45.1. MIBORA MINIMA *59, S. Lancs.: In low fore-dune, NE end of Marine Lake, Southport, SD337185, D.P. Earl & J. Buckley-Earl, 1996, conf. T.A. Cope. 153/46.1. POLYPOGON MONSPELIENSIS *58, Cheshire: Old sludge bed and tracks, Wharton, Winsford, SJ656677, J.P. Guest, 1997, herb. G.M. Kay, det. E.J. Clement. +*95, Moray: Several plants on waste ground, Hopeman, NJ149686, I.P. Green, 1997. +153/48.1. BECKMANNIA SYZIGACHNE *59, S. Lancs.: Ditch bank between Marsh Farm and River Douglas, Hesketh Bank, SD452240, D.P. Earl, 1996, conf. T.A. Cope. 153/50.3. BROMUS RACEMOSUS 51, Flints.: Hay meadow, Coed Duon, Tremeirchion, SJ073715, A.O. Chater, 1997. 1st post-1930 record. +153/50.5. BROMUS LEPIDUS 12, N. Hants.: Grassy part of garden, Warren Road, Liss Forest, SU7828, det. F. Rose. Sandy tank tracks, Hogmoor Enclosure, SU7834; both A. Brewis, 1997, herb. A.B. 1st and 2nd recent records. *+153/50.7. BROMUS SECALINUS “*11, S. Hants.: Edge of bean field, S of Twyford, SU485232, P.J. Selby, 1997, herb. R.P.B., conf. R.P. Bowman & E.J. Clement. *153/52.2. ANISANTHA RIGIDA *29, Cambs.: Large patches by fence between sown area and old shelter belt, SE side Magog Down, TL490531, S.M. Walters, 1997, CGE, det. D.A. Wells. ¥153/52.4. ANISANTHA TECTORUM ~~ *58, Cheshire: New earth bank, Moreton, $J263901, P.G. Gutteridge, 1997, herb. G.M.K., conf. G.M. Kay. 7153/52.5. ANISANTHA MADRITENSIS *12, N. Hants.: Among rubbish, Woolmer Trading Estate, SU795350, A. Brewis, 1997, herb. A.B. +153/53.1. CERATOCHLOA CARINATA — *95, Moray: Very well naturalised on road verge and edge of arable field, Roseisle, NJ138670, P.R. Green, 1997. Well naturalised on verge of lane, Duffus, NJ174687, P.R. Green, 1997. Ist and 2nd records. 153/53.3. CERATOCHLOA CATHARTICA 51, Flints.: College grounds, Welsh College of Horticulture, Northop, SJ2369, J.A. Green & BSBI group, 1996. Ist post-1930 record. 153/54.1. BRACHYPODIUM PINNATUM _ 89, E. Perth: Ungrazed grassland on sloping calcareous gravels, banks of River Tay near Boat of Murthly, NO0O57396, N.W. Taylor, 1992, PTH, det. T.A. Cope. PLANT RECORDS 589 153/56.1x3. ELYTRIGIA REPENS x E. JUNCEA (E. x LAXA) 3, S. Devon: At head of creek, Salcombe, Lower Batson, SX735397, L.J. Margetts & L.M. Spalton, 1997, det. T.A. Cope. At head of saltmarsh creek, River Erme, Holbeton, SX619491, L.M. Spalton, 1997, det. T.A. Cope. Ist and 2nd post-1930 records. 153/56.2. ELYTRIGIA ATHERICA *12, N. Hants.: Sown on sandy waste ground, Bordon, SU7936, A. Brewis, 1997, herb. A.B. 73, Kirkcudbrights.: Shore of marsh, Kirkdale, NX518527, O.M. Stewart, A. White & C. Jermy, 1996. Ist recent record. 7153/59.1. HORDEUM DISTICHON *47, Monts.: Farm manure heap, SE of Llanwyddyn, S10213> CJ: Tyas, 1986. 7153/59.3. HORDEUM JUBATUM *42, Brecs.: Well established on verges of A465, between Brynmawr and Rhymney, SO1812, T.G. Evans, 1997. 81, Berwicks.: Established on road verge, A698 below Lochton, NT777385, M.E. & P.F. Braithwaite, 1995, herb. M.E.B. 2nd record. +153/60.1. TRITICUM AESTIVUM *S, Channel Islands: Roadside, La Vaurocque, Sark, WV461757, M. Marsden, 1982, JSY. *46, Cards.: Field crop, 300m W of Tegfan, near Llechryd, SN226451, A.O. Chater, 1992, NMW. 7153/62.1. CORTADERIA SELLOANA *49, Caerns.: Edge of pool, Spinnies Nature Reserve, near Bangor, SH613721, G. Battershall, 1994. 7153/62.ric. CORTADERIA RICHARDII *44, Carms.: Waste ground, below Cwmbach Road, Llanelli, SN498012, G. Hutchinson & I.K. Morgan, 1994, det. T.B. Ryves. *46, Cards.: Gorse slope above tip, Pendinas, Aberystwyth, SN584799, A.O. Chater, 1993, NMW. Streambank 100m S of Ynys-hir bridge, Eglwys-fach, SN684956, P. Condry, 1995, det. A.O. Chater. 1st and 2nd records. 7 153/65.cil. ERAGROSTIS CILIANENSIS *29, Cambs.: In scattered bird seed, near Kidman’s Plantation, Whittlesford, TL447488, A. Arbon, 1994, conf. R.M. Payne. 7153/68.1. ECHINOCHLOA CRUS-GALLI *49, Caerns.: Pavement at base of wall, Llandudno, SH794819, W. McCarthy, 1995. Roadside, Bethesda, SH6066, J. Hawksford, 1997. Ist and 2nd records. 7153/68.col. ECHINOCHLOA COLONA *4, N. Devon: Garden paving, Instow, SS473303, M. Tulloh, 1989, det. T.A. Cope. Probably introduced in bird-seed. 7 153/68.esc. ECHINOCHLOA ESCULENTA *39, Staffs.: Waste ground by RSPB centre, by River Tame, 3km E of West Bromwich, SP035927, J.P. Martin, 1989. *44, Carms.: Shingle of river, Afon Lliedi, near Buckley’s Brewery, Llanelli, SN509006, I.K. Morgan, 1991, NMW, det. G. Hutchinson. 2nd Welsh record. 7153/70.pum. SETARIA PUMILA *49, Caerns.: Base of wall, Jubilee Street, Llandudno, SH7831, W. McCarthy, 1997. 7153/71.san. DIGITARIA SANGUINALIS 39, Staffs.: Two plants on roadside, Railway Drive, Madeley, SJ772441, I.J. Hopkins, 1995. 1st record since 1932. 7153/FAR.spa. FARGESIA SPATHACEA *S, Channel Islands: Roadside verge, near fence, Dixcart Lane, near Dixcart Hotel, Sark, WV465754, R.M. Veall, 1997, JSY, det. D. McClintock. Planted; one clump now about 1-5m tall. *46, Cards.: Several large thickets in damp estate woodland, Highmead, Llanybydder, SN499431, A.O. Chater, 1985, NMW, det. C. Stapleton. [previously erroneously published as 153/1.1. Semiarundinaria anceps (under Arundinaria jansaurensis), Watsonia 16:196 (1986) and BSBI Welsh Bull. 44: 29(1986)]. Colony in wood near gardens, Y Glog, Rhydyfelin, SN593793, A.O. Chater, 1985, NMW. Ist and 2nd records. 7153/PAN.sch. PANICUM SCHINZII *29, Cambs.: Two plants by grain silo, Newmarket, TL646629, G.MLS. Easy, 1984. 154/1.1b. SPARGANIUM ERECTUM subsp. MICROCARPUM _*44, Carms.: Ditch on boundary of two farms, Carway, SN463058, R.D. Pryce, 1979, NMW. 154/1.1d. SPARGANIUM ERECTUM subsp. OOCARPUM *3, S. Devon: Stream, Thorverton Village, SS927021, L.J. Margetts, 1997. 590 PLANT RECORDS 155/1.1x2. TYPHA LATIFOLIA x T. ANGUSTIFOLIA (T. x GLAUCA) *12, N. Hants.: On the marshy edge of the promontory that extends into the pond from the W bank (probably acidic), Shortheath Pond, SU775368, B.A. Gale, 1985, herb. B.A.G., det. A.C. Leslie. Possibly introduced by an angling club. *37, Worcs.: With both parents in small reed-swamp, Broadwaters, Kidderminster, SO8478, W.A. Thompson & B. Westwood, 1994, conf. M. Briggs. 1156/5.2. HEMEROCALLIS LILIOASPHODELUS *77, Lanarks.: Rough grass at base of scrubby trees, East Flatt, East Kilbride, NS6551, P. Macpherson, 1995, herb. P.M., det. E.J. Clement. Presumably a result of fly-tipping; obviously surviving (still present in 1997). +158/6.1. KNIPHOFIA UVARIA __*46, Cards.: One non-flowering plant among gorse on heathy ground among leadmine spoil heaps, Cwmsymlog, SN698837, A.O. Chater, 1996. *81, Berwicks.: Established where cast out in rough ground near railway, Burnmouth, NT955610, V. Muscott, 1997. 7158/6.2. KNIPHOFIA PRAECOX *46, Cards.: Scrub on streambank W of bridge by Llansantffraed church, SN512674, S.P. Chambers, 1997. 158/9.1. GAGEA LUTEA = +95, Moray: Hundreds of plants on W bank of River Spey, Orton, S of Fochabers, NJ317523, P.R. Green, 1996. 2nd record, Ist native site. +158/12.bulxmac. LILIUM BULBIFERUM x L. MACULATUM (L. x HOLLANDICUM) *58, Cheshire: Sand dunes, Hoylake, $J207877, G.M. Kay, 1997. A small colony has developed from original introduction. 158/20.2x3. HYACINTHOIDES NON-SCRIPTA x H. HISPANICA *46, Cards.: Scrub slope W of church, Llanbadarn Fawr, SN598810, S.P. Chambers & A.O. Chater, 1991. *158/22.1. CHIONODOXA FORBESII *26, W. Suffolk: Old garden, now plantation of beech trees, Ousden, TL735595, R. Addington, 1988, det. by a local gardener. *49, Caerns.: Naturalised on roadside, Bodafon Lane, Llandudno, SH8081, W. McCarthey, 1997. *Q9. Dunbarton: Well established in wood with rich loamy soil, behind Cardross Park, NS3477, A. McG. Stirling & A. Rutherford, 1985. 7158/23.1. MUSCARI NEGLECTUM *46, Cards.: Grassy roadside slope, Fron-deg, Llanbadarn Fawr, SN600811, A.O. Chater, 1993. Scrub slope W of church, Llanbadarn Fawr, SN598810, S.P. Chambers, 1997. Ist and 2nd records. 7158/23.2. MUSCARI ARMENIACUM *46, Cards.: Grassy roadside verge, road to Upper Forest, Lampeter, SN576490, A.O. Chater, 1992. Copse by Primrose Hill, Llanbadarn Fawr, SN600811, A.O. Chater, 1993. 1st and 2nd records. 7158/24.12. ALLIUM CARINATUM *$1, Berwicks.: Established on limestone rocks, Birgham » Haugh Wood, NT798389, M.E. Braithwaite, 1995, herb. M.E.B. 7158/24.2. ALLIUM UNIFOLIUM *99, Dunbarton: Spreading well in damp, basic soil in deciduous woods, behind Cardross Park, NS3477, A.McG. Stirling & A. Rutherford, 1985, det C. D. Brickell. Considered first British escape. 158/31.1. LEUCOJUM AESTIVUM ~~ +*89, E. Perth: Among Phragmites in Tay Marshes, Perth, NO129217, O.M. Stewart, 1982, E. 158/31.2. LEUCOJUM VERNUM 8, S. Wilts.: Boggy ground near pond, Britford Green near Salisbury, SU1627, M. Marks, 1991, det. D.P.E. Stevens. 1st record for 90 years. +*49, Caerns:.: 7 clumps by footpath, Glynllifon Park near Caernarfon, SH4555, D. Jones, 1991. +158/32.1x3. GALANTHUS NIVALIS x G. PLICATUS *46, Cards.: Damp grassland, Aberstringell limekilns, Llanrhystud, SN519684, A.O. Chater, 1995, NMW. *95, Moray: Kincorth Wood with both parents, Kintessack, NJO11617, P.R. Green, 1997. 7 158/32.4a. GALANTHUS PLICATUS subsp. PLICATUS *95, Moray: Well naturalised in Kincorth Wood, Kintessack, NJO11617, P.R. Green, 1997. 7158/32.1ka. GALANTHUS IKARIAE *95, Moray: Naturalised in waste ground, N of Forres, NJ026608, P.R. Green, 1996. +158/33.3. NARCISSUS POETICUS *46, Cards.: Sea cliff at W side of Aber-porth, SN257515, A.O. Chater, 1992. Roadside hedgebank, Clogfryn, Aberaeron, SN449623, A.O. Chater, 1994. Ist and 2nd records. ° PLANT RECORDS 591 +158/33.3a. NARCISSUS POETICUS subsp. POETICUS *46, Cards.: Copse NW of Ffosrydygaled, Chancery, SN578765, A.O. Chater, 1994. Roadside bank at NE end of Llwyncelyn, SN443598, A. O. Chater, 1995. 1st and 2nd records. +158/33.3b. NARCISSUS POETICUS subsp. RADIIFLORUS *46, Cards.: Abundant under trees 100m E of Lovesgrove mansion, SN629816, A.O. Chater, 1995. Hedgebank of road by Tanycastell, Llanychaiarn, SN588788, A.O. Chater, 1994. 1st and 2nd records. *+158/33.3x5. NARCISSUS POETICUS x N. PSEUDONARCISSUS (N. x INCOMPARABILIS) *46, Cards.: Wooded streambank 200m W of Coed Wileirog, Wallog, SN601854, A.O. Chater, 1992. 7158/33.3x5c. NARCISSUS POETICUS x N. PSEUDONARCISSUS subsp. MAJOR (N. x INCOMPARABILIS [N. x BERNARDII]) *46, Cards.: Roadside hedgebank, Tanycastell, Rhydyfelin, SN588788, A.O. Chater, 1994. 7158/33.5c. NARCISSUS PSEUDONARCISSUS subsp. MAJOR *46, Cards.: Sand dunes by N end of road, Ynys-las Dunes NNR, SN611940, A.O. Chater, 1993. Marshy scrub 200m S of Felin Geri, Cwm Cou, SN300421, A.O. Chater, 1994. Ist and 2nd records. +158/33.5x6. NARCISSUS PSEUDONARCISSUS x N. CYCLAMINEUS (N. x MONOCHROMUS) *46, Cards.: Scrub by old cottage site 100m E of Cwmtudu, SN357575, A.O. Chater, 1994. ¥158/33.bic. NARCISSUS BICOLOR *46, Cards.: Ash wood SW of Pen-y-wern, New Cross, SN636765, A.O. Chater, 1994. Mixed woodland on ridge E of Crugiau, Rhydyfelin, SN592794, A. O. Chater, 1996. Ist and 2nd records. 7158/35.2. RUSCUS HYPOGLOSSUM *12, N. Hants.: Well-naturalised in deciduous woodland, Ashford Chace, Steep, near Petersfield, SU741261, J.D. Fryer, 1998, herb. J.D.F., det. J.M. Mullin, Probably bird-sown. +158/36.1. ALSTROEMERIA AUREA “*41, Glam.: Eastcliff, Pennard, SS5567, A.S. Lewis, 1997. 7158/ERY.den. ERYTHRONIUM DENS-CANIS _*17, Surrey: Many seedlings of varying ages in cemetery, Brookwood, SU9656, J.F. Leslie & K. Page, 1988. Seeded from grave. 7159/2.5. SISYRINCHIUM STRIATUM *35, Mons.: Two plants on earth mound in road development, Cleppa Park, Newport, ST278847, G. Hutchinson, 1994. 7159/8.1. CROCUS VERNUS *95, Moray: Well naturalised on dunes, Hopeman, NJ146698, P. R. Green, 1996. 7159/8.6. CROCUS CHRYSANTHUS *46, Cards.: Rough grass in SE corner of Llanafan churchyard, SN685721, A.O. Chater, 1997. 7159/8.7xang. CROCUS FLAVUS x C. ANGUSTIFOLIUS (C. x STELLARIS) *46, Cards.: Roadside verge 400m NNW of Moelifor, Llanrhystud, SN548710, A.O. Chater, 1997. 7 159/13.2. CROCOSMIA MASONIORUM ~— _*95, Moray: One patch on waste ground, Miltonduff, NJ201587, IP. Green, 1996, conf. P.S. Green. 162/3.7. EPIPACTIS PHYLLANTHES *37, Worcs.: Old quarry, Fish Hill, Broadway, SP1136, T. D. Knight & R. Maskew; Wooded bank on edge of Hartlebury Common, $O8270, J.J. Day; both 1987, det. R. Maskew, conf. A.J. Richards. 1st and 2nd records. 162/5.1. NEOTTIA NIDUS-AVIS_ 47, Monts.: Old valley woodland, Abermule Dingle, SO19, J. Clarke & A. Franks, 1992. 2nd record and only extant locality. *108, W. Sutherland: Deciduous woodland, Loch Dubh, Ardroe, NC072238, R.E.C. Ferreira, 1981. 162/8.1. GOODYERA REPENS *109, Caithness: Peat moor, Aukengill Loch, ND353652, J.K. & S.I. Butler, 1996. 162/11.1. CORALLORHIZA TRIFIDA %*91, Kincardines.: In Sphagnum in willow carr, Banchory, N0687946, D. Welch, 1992. 162/15.1. PSEUDORCHIS ALBIDA 50, Denbs.: Eryrys, SJ2057, M. Prosser, 1986. Ist record since 1950. 162/16.1b. GYMNADENIA CONOPSEA subsp. DENSIFLORA *78, Peebless.: West Linton, NT15, A. Craig-Christie, 1873, E, det. F. Rose. 162/16.1c. GYMNADENIA CONOPSEA subsp. BOREALIS 46, Cards.: Heathy pasture, Cae’r- meirch, Pontrhyd-y-groes, SN7573, A.O. Chater, 1996. 2nd record 592 PLANT RECORDS 162/16.1x18.1. GYMNADENIA CONOPSEA x DACTYLORHIZA FUCHSI (x DACTYLODENIA ST- QUINTINI) *39, Staffs.: One plant with parents on N facing roadside bank on limestone, Ecton, SK098586, B.R. Fowler, 1987, det. R.H. Roberts. 162/17.1. COELOGLOSSUM VIRIDE 83, Midlothian: Thirty-one, very small plants seen in heavily grazed rich grassland, Fullarton Water, NT282562, C. Dixon, 1997, det. D.R. McKean. 1st record for over 100 years. 162/18.8. DACTYLORHIZA LAPPONICA *108, W. Sutherland: Schoenus mire, Ardbhair, NC155333; Kinloch Lodge, NC566510; both R.E.C. Ferreira, post 1985. Ist and 2nd records. 162/23.3. OPHRYS APIFERA 67, S. Northumb.: Reclaimed pit-heap, near Rising Sun Farm, NZ296688, M. Jones, 1996. 2nd record. 7+162/SER.par. SERAPIAS PARVIFLORA *2, E. Cornwall: Natural, coastal grassland, Penlee Point, Watchhouse Field, SX439490, P. Cobbing, 1989, det. P. Cribb. New to the British Isles; probably deliberately introduced. Watsonia 23: 593—597 (2001) 593 Book Reviews Plant life of the Quaternary cold stages. R. G. West. Pp. xi + 320. Cambridge University Press, Cambridge, 2000. Hbk £65.00. ISBN 0-521-59397-2. Every once in a while it is important to take stock and assess the mass of palaeobotanical information accumulated. This book does that and gives the reader an excellent, much needed review of plant life in the British Isles during the cold periods that characterised a large proportion of the last two million years (the Quaternary Period). Both palynological (pollen and spores) and plant macrofossil (seeds, fruits, etc.) evidence is examined and integrated into the discussion of vegetation history of the British Quaternary cold stages, an approach that should be noted and copied by other palaeobotanists. After setting out the proposed aims for the analysis of cold stage floras there are a number of introductory chapters. One of these is an illuminating and important chapter on the type of sedimentary environments encountered in the fossil record and how investigations of taphonomic processes acting upon plant parts before they are deposited in modern analogue environments can allow more precise interpretations of cold stage fossil assemblages. A compact disc accompanies the book which contains the database which is used as the foundation for the analysis of the cold stage fossil assemblages. Chapter 4 describes the structure of this database which contains information on the fossil sites, their chronology, the botanical remains recovered from each sample, the age of those samples, certain information on the biology, habitat and present distribution of the fossil taxa recorded, and a bibliography. The latter part of the book analyses the data collected. Querying the database shows which taxa are found where, when and in what quantity. Discussions are presented about how to translate the quantity of remains into actual representation in the past vegetation. It appears that many taxa represented in cold stage deposits are present in certain parts of Britain today. Another interesting revelation is that many taxa are recorded in both temperate and cold stages, indicating how Linnean species can survive climatic fluctuations. A knowledge and understanding of the biology of the flora (life forms, life spans and variability of species) is essential in explaining such results and these are discussed. Vegetation, habitat, and climate is reconstructed for the cold periods, with the contribution of coleopteran, non-marine molluscan and vertebrate faunas to a wider view of cold stage environments also assessed. Climatic changes in a cold stage can be difficult to determine from the fossil assemblages, because extensive grassland, which is often represented in the cold stage fossil record, can exist under a range of climatic regimes. This is a difficult problem which is addressed. The reviewer recommends this book to any member of the scientific community or general public who is interested in an account of the flora that once existed in Britain during the Quaternary cold stages, and its implications for understanding present flora and environment (including climate). The price tag of £65 is not too expensive for a hard-back book (plus CD database) of this type. M. H. FIELD British Plant Communities. Volume 5. Maritime Communities and Vegetation of Open Habitats. Edited by J. R. Rodwell. Pp. xiv + 512. Cambridge University Press, Cambridge, 2000. Hbk. £70.00. ISBN 0-521—39167-9. Pbk £32.50. ISBN 0-521-64476-3. Work on the ‘National Vegetation Classification’ project began in 1975 and now with the appearance of the fifth Volume of ‘British Plant Communities’ it reaches a triumphant conclusion. Its completion is a tribute to those who initiated and fostered it, to its dedicated and indefatigable editor John Rodwell, and to his army of co-workers and helpers. The five volumes, taken together, display not only the size and complexity of the task they undertook but also the uniformly high quality of the final product. The preceding volumes have already amply proved its practical value and are widely accepted as essential tools in the hands of vegetation surveyors, managers and conservationists. Now that coastal vegetation and communities of agricultural weeds and open habitats are covered, this comprehensive and unique phytosociological treatment of British vegetation is complete. 594 BOOK REVIEWS A Preamble (as in earlier volumes) explains the background to the work and the style of presentation. The main part of the book consists of Community descriptions, arranged in four sections: Salt-marsh communities (28 in number); Shingle, strandline and sand-dune communities (19); Maritime cliff communities (12) and Vegetation of open habitats (42). Each section includes a key to the communities (described as ‘a crude guide to identifying the types of vegetation’ — helpful though not infallible!), and begins with an Introduction. These Introductions are a particularly useful feature and will appeal to the many botanists who have an interest in the flora and ecology of coastal vegetation, or in weeds and ruderals. At the close of this volume is a 14- page ‘Phytosociological Conspectus of British Plant Communities’ which users of the series have already looked for but of necessity had to await completion of the whole. This will facilitate cross- reference between different parts of the work, and will also enable comparisons with Continental schemes of classification. The dynamic nature of much coastal vegetation, notably that of salt-marshes and dunes, creates certain problems for community classification, and its application in the field. A country-wide scheme such as this may not, in all cases, fit comfortably with the floristic variation and patterning of local sites. However, with experience of working in a range of sites and with a number of examples it will normally provide the user with a convenient and instructive framework, bearing in mind that it is based solely on floristic groupings and does not necessarily reflect ecological processes very closely. Special problems also confront the task of sampling and classifying communities of open habitats and arable fields, which in any case constitute a more heterogenous grouping than most of the other vegetation types. However, they do have features in common, and in view of the scant attention paid in Britain to their phytosociology this is a most welcome component of the whole. This volume, with its predecessors, will become for the foreseeable future the standard and indispensable handbooks for survey, classification and mapping of vegetation in Britain. Their intrinsic interest also extends to all students of the British flora, for whom they contain a wealth of relevant and valuable information. It is of the utmost importance, however, to heed the Editor’s warning that this should not be treated as a ‘static edifice’ but as a working tool, which with use and in the light of further sampling and new data, will require up-dating in the future. C. H. GIMINGHAM The Flora of Dorset. Humphry Bowen. Pp viii + 373. Pisces Publications, Newbury, 2000. Hbk £45.00. ISBN 1—87—4357-16-1. Dorset may be seen as a nursery of modern botanical thought. For the first half of the twentieth century Ronald Good developed an understanding of the correlation of plant distribution with external factors such as climate, drainage and soil type. Good’s 1948 Geographical Handbook, and its subsequent revisions, have been the principal published sources of botanical information in Dorset until the recent production of Humphry Bowen’s Flora. In common with other recent floras the principal author is joined in the preparation of the introductory chapters. David Allen’s account of some earlier workers on the Dorset Flora is of interest for its perception of field botany as a social phenomenon. The section on archaeobotany of the county, by Pat Hinton, is intriguing. Post-glacial evidence of vascular plants would be a valuable addition to any other County flora in preparation. The Flora of Dorset is a flora of the tetrad recording tradition. The accounts and associated maps concentrate on post-1984 records. For each species the number of tetrads for post-1984 records are given together with historic records. The maps and the accounts, however, do not always tally. For example, the maps of Ophrys insectifera and Platanthera bifolia show eight and ten post-1984 records respectively. The text however only refers to four post-1984 sites for Ophrys and five for Platanthera. Those extracting data for use elsewhere should check the full records which are held in the Dorset Environmental Records Centre. Most species accounts only give general locations such as a parish name within a tetrad. The more sought after groups such as the localities of orchids and sedges are more precise. The absence of a general map of the vice-county, combined with not using four figure grid references to identify tetrads, adds to the difficulty in identifying localities. Judged against the two aims set out in its introduction the Flora may be judged as successful by BOOK REVIEWS 595 those searching for attractive plants. The leisure botanist will undoubtedly gain great pleasure from the Flora. Similarly keen gardeners and arboriculturalists will appreciate the many accounts of selected localities of some ornamental plants in cultivation. It is unfortunate that our vice-county floras are now drawn upon by more people than just those who love wild flowers. With the advent of legal protection for wild plants and their habitats, and with the integration of biodiversity planning into the work of government, biological data are becoming very important. Both data, and any interpretation, is now subject to detailed scrutiny by those whose interests may not be primarily botanical. It was therefore a disappointment that the excellent data that are contained within the Flora are prone to misinterpretation and challenge. At the heart of this vulnerability is the apparent lack of consistency of approach. For example, whether or not a species is extinct, or nearing extinction, in a county is of great importance. The Flora does record some species as extinct, 1.e. Pulicaria vulgaris and Oenanthe aquatica. Other species with equally long periods since being last recorded and with similar histories of habitat change are in contrast not recorded as extinct, 1.e. Eleocharis parvula and Eriophorum gracile. A clear and consistent use of criteria is not apparent. The emphasis in the Flora on aliens, garden specimens and casual occurrences similarly complicates interpretation. Complications arise where accounts reproduce all records for a species but do not differentiate between persistent native populations, casuals and possible introductions. For example, the status of Mentha pulegium described in the Flora is difficult to reconcile with that in the 1994 Scarce Plant accounts reproduced in the 1999 Red Data Book. The discovery of the important population of Mentha pulegium in Purbeck in 1996 is masked by the other parts of the account. As the presence of this protected species has considerable significance not only to botanists, but also to the landowners, a proper understanding of its status is essential. The Flora of Dorset will bring pleasure to many. It also provides a valuable point of reference upon which others may build. The interpretation of the data gathered for the Flora will need to be progressed elsewhere. The Flora of Dorset has raised many fundamental issues as to what a flora is for. If a flora intends to inform scientists or civil servants whose decisions can help or hinder the wellbeing of our wildlife and their habitats then a different form to that adopted by the Flora of Dorset is required. If we wish our county floras to serve a community wider than just botanists, we need to find clearer ways to communicate with those whose decisions determine the fate of the natural world. C. CHATTERS Wild Flowers of the Peak District. Text by Patrick Harding. Art Editor Valerie Oxley. Pp. 144. Hallamshire Press, Sheffield, 2000. Price £ 25.00. ISBN 1—874718-—53-9. Offering an interesting marriage between science and art, this book has a curious pedigree. It is the very professional outcome of a selection of work from students on a Botanical Illustration Certificate course for adults at Sheffield University. A well crafted appearance and articulate text brings credit to both the tutors and the students who produced the illustrations. Clearly the painters have worked from a basis of sound botanical knowledge. However, if you know and love the Peak District as a meeting point for highland and lowland Britain, you might be disappointed to find that it is represented by only sixty specimens of its flora, and few of the area’s rarities find a place. But, among the criteria for a plant’s inclusion have been the twin aims for a chronological treatment of the flowering year and the need to select paintable plants that can be depicted life-size. These are understandable restrictions. This reviewer misses the convention, usually found in botanical illustration, that adds enlarged details for a plant whose fruit, flower, leaf or root characteristics are particularly hard to define. Photographs are included in each species double spread and they sometimes clarify the situation; they also set the plant usefully within its environment. Regrettably, there is considerable variation between the printing of a painting, usually excellent, and its companion photograph where the colour is harsher. The standard of the painting is consistent and very good, with each page beautifully laid out. The delicate drawing of some of the root systems, done, we are assured, from archive specimens, is to be applauded. A minor carp would be that there is considerable similarity of style in the painting of each plant. A plus is the absence of any text upon the actual illustration page, which, in some cases, becomes a work of art. I particularly liked the very painterly depiction of the ubiquitous Rosebay Willowherb, a notoriously difficult subject. 596 BOOK REVIEWS The text which is scholarly in detail but relatively light-hearted in delivery, is very fine. Where Latin nomenclature has been changed in recent years both old and new names are included. There is a lot of useful information in the introductory paragraph for each plant. For those who enjoy their Grigson, the inclusion of folk-names, folk-lore, herbal, medical and domestic uses associated with the plant, make good reading. An early chapter briefly discusses the area’s climate, geology and biotic factors. This is not a volume for the pocket or rucksack, but would be a happy addition to a flagging botanist’s bookshelf, or a coffee table. Furthermore, if your fieldwork is increasingly done more by wistful memory from an armchair than by being part of the muddy boots brigade, this book could be a delight. You would not need another mortgage in order to buy it! M. M. MARTIN Zander. Handworterbuch der Pflanzennamen. Dictionary of Plant Names. Dictionnaire des noms de Plantes. 16th edition. W. Erhardt, E. Gotz, N. Bédeker & S. Seybold. Pp. 990. Eugen Ulmer GmbH & Co., Stuttgart, 2000. Hardback. ISBN 1—903257-01—8. Marketed by The Plant Press, Lewes, East Sussex. Professor Robert Zander (1892-1969) produced the first edition of his Handwéorterbuch in 1927, and went on to revise it up to the ninth edition in 1964. Since then ‘Zander’, as it is generally known, has been revised by two or more other authors; of the current four only Seybold was involved before the 16th edition, which is the first to be trilingual and which includes about 3,640 genera, 20,000 species and 10,000 synonyms. The alphabetical listing of these genera and species occupies the bulk of the book (626 pp.), but comprises only one (no. 6) of 11 chapters. It claims to cover the most important cultivated plants and nearly all the native plants of central Europe, and hopes in future editions to take in plants of western and northern Europe as well. Zander is a much used reference work, and in turn follows the most generally accepted standard publications such as the International Code of Botanical Nomenclature (1994), Brummitt & Powell’s Authors of Plant Names (1992), Kubitzki’s The Families and Genera of Vascular Plants (1990 onwards), Cronquist’s The Evolution and Classification of Flowering Plants (1988), Dahlgren, Clifford & Yeo’s The Families of Monocotyledons (1985) and Mabberley’s The Plant Book (1997). It is unique in scope. Perhaps closest to it is Griffiths’ Index of Garden Plants (1994), which covers about three times as many taxa and includes potted descriptions of them, but omits many purely wild plants, has fewer supplementary chapters (see below), is less wieldy and costs £40. Hence Zander has 43 and Griffiths 15 species of Trifolium; but the figures for Solidago are 18 and 38 respectively. To give an idea of the coverage of a purely cultivated genus, Zander has 75 and Griffiths 120 species of Opuntia. For each species Zander provides the Latin name (plus the gender of each genus); authority: German, English and French name(s); a series of symbols that denote growth form, flowering period, garden usage (e.g. border plant, hanging plant, cut-flower plant, hothouse plant, decorative- fruit plant), whether scented, officinal (its designation ‘Off is omitted in error from the list of abbreviations), poisonous or otherwise useful, and conservation/protection status; and a detailed list of its native territories. The English names were apparently compiled by Charles Quest-Ritson (Salisbury); in the main they are the familiar ones, but there are also many much less used names and many well-known ones are omitted. For example, Montia fontana is given no English name, while M. perfoliata is called Miner’s Lettuce and Winter Purslane; Spring Beauty is (?mis)applied, along with Siberian Purslane, to M. sibirica. Hence the standard English name is omitted from all three! The French names seem to me even more patchily quoted; for example Tussilago farfara is not provided with one even though its called Pas d’ane in all my French Floras. It is not difficult to find other minor errors, such as spelling mistakes, in the book, but on the whole it is well produced and authoritative. The other ten chapters provide a wealth of additional and often surprising information: 1 — biography of the authors of all 16 editions; 2 — explanation of botanical nomenclature and of the pronunciation of Latin names; 3 — systematic list of families of vascular plants (following Cronquist and Dahlgren ef al.); 4 — alphabetical list of families with the genera listed alphabetically under each; 5 — an elaborate (17 pp.) hierarchical listing of the world’s phytogeographical-regions; 7, 8 & 9 — indices of German, English and French plant names (the BOOK REVIEWS 597 above comments on these are reinforced by the 96, 71 and 19 pages respectively devoted to each); 10 — list of abbreviations and full names of authorities of plant names; and 11 — a short bibliography. All the free text in these chapters is triplicated in the three languages. For all people seriously interested in plants, Zander is a most valuable handbook of information. First and foremost it provides a list of virtually all the plants that one is likely to encounter, with a wealth of information about each. Many of the other chapters are most likely to be of use to those without access to a botanical library, enabling them to refer to, inter alia, the most commonly used systems of classification, the rules of botanical nomenclature and the standard abbreviations of authorities. Probably no other book provides quite this range of information, certainly none in such a handy compact format. This first trilingual version of Zander deserves a much wider circulation among English speakers than previous editions have enjoyed. If the next edition succeeds in improving the list of English names and in adding or emending the entries of the wild and cultivated plants that grow in the Atlantic fringes but not in central Europe, it will be indispensable. C. A. STACE Causes of Change in British Vegetation. Ecofact Volume 3. L. G. Firbank, S. M. Smart, H. M. van de Poll, R. G. H. Bunce, M. O. Hill, D. C. Howard, J. W. Watkins and G. J. Stark. Pp. 98. Dept. of Environment, Transport and the Regions, London, 2000. Price £8.00. ISBN 1—870393-51-1. This is the third volume in this series. It has as its stated aims to: 1. Identify the causes of observed changes in botanical diversity, 2. Assess the relative importance of land management and other factors, such as pollution, 3. Recommend land management practices for the maintenance and enhancement of diversity, 4. Develop predictive techniques for determining ecological impacts. With aims like these, this could well be described as being the botanical conservation holy grail — not only does it confirm in hard scientific terms what we felt we knew all along; that nutrients are up, diversity is down, with an increase in competitors, but also it offers reasons, solutions and a method for stopping this happening again (or at least for letting us know if it’s happening again). So, does it deliver? The authors identify four major drivers of vegetation change, being |. Changes in agriculture, 2. Changes in the management of roadside verges, 3. Conifer planting and 4. Inputs of nitrogen and other nutrients. These are further broken down into individual factors such as waterside management, and the impact of leisure. There is an informative table on how these drivers shifted between the surveys and their effects on various environmental factors. They then go on to define twelve Indicators of Botanical Diversity (or I.B.D.s), such as Ellenberg scores per plot, frequency of scarce species, and food plants for animal groups. These drivers and the indicators are then linked by chapters on the causes of change and policy implications. Hidden in the latter chapter is a table which is perhaps the most informative part of the document, listing policy initiatives such as S.S.S.I. designations, Nitrate Sensitive Areas or Biodiversity Action Plans against eighteen drivers such as drainage and conifer planting, then assessing the impact that these policies have had on the drivers. The remaining forty five pages consist of twenty explanatory ‘boxes’ and annexes, making this a highly readable and digestible document. Is there a down side? My only grumble was that there is much emphasis on Ellenberg’s continentality scores, which were ignored in the technical annexe, while salt tolerance was covered but not in the plots analysed; but this is pretty much a minor point. So, in answer to my question, yes, it does deliver. These are the hard data, the ‘smoking gun’ that our campaigning conservation cousins should use as evidence to influence policy, although it would be interesting to know if we will ever get the chance to see original data. This series of Ecofact publications should continue to contribute in an objective, dispassionate way to the conservation lobby’s armoury and together with the forthcoming Atlas, form a formidable data bank on which to base their case. S. J. WHILD Watsonia 23: 599-600 (2001) 599 Obituary JOHN MICHAEL MULLIN (1947-1999) Michael Mullin, or Mike as he was known to his botanical colleagues, family and friends, was born in 1947 in Tyneside into a large family of three brothers and two sisters. A few years later his family moved to Stevenage New Town, and his natural history interests were shown early; when he was five years old Michael went missing for most of a day, until found happily and totally absorbed in a study of the local frogs. As a teenager Michael’s main interest was firmly focused on botany, and with a school friend he organised an exchange scheme of insectivorous plants. At fourteen he was seriously recording wild plants, helping John Dony on his Flora of Hertfordshire project at weekends and in school holidays, and later he worked with the Upper Teesdale Group from Durham University. Schooling was at Alleyne Grammar School from which he left with eight ‘0’ level passes, but always with a tendency against authority, he left with a rude message to the headmaster watered in weedkiller onto the cricket square. Later from South Shields Grammar Technical School for Boys, he left with three ‘A’ level passes - one with distinction in Biology. Mike started work as a Biology lecturer in various schools, soon adding field trips for Extra- mural Associations, and slide presentations. As a keen walker, cyclist and camper with a passionate interest to see every British plant, by the time that he joined the staff at The Natural History Museum (then the British Museum (Natural History)) Mike had an exceptional knowledge of field localities all around the British isles. Mike had met his wife Anne on field work at Malham Tarn in 1966, and together they had continued camping, walking and hitch-hiking to field meetings or to search for plants, collecting for herbarium specimens where appropriate. From 1973-1990 Mike worked at the BM (Natural History), first as a student then as a Scientific Officer when he worked mainly in the British Herbarium curating collections. He participated in the British Museum Outer Hebrides project and was co-author with R. J. Pankhurst of the Flora of the Outer Hebrides (1991). Working in the Museum Mike acquired a phenomenal knowledge of plants from around the world, and was always prepared to spend much time researching plants brought to him for identification, and other botanical queries that came to him in the Department. As a member of the B.S.B.I. Meetings Committee, Michael had special responsibility for the organisation of the Annual Exhibition Meeting, which for many years was held in the Department of Botany each November, attended by several hundred members and guests - in the General Herbarium. Mike cycled to work and around London and his bicycle became his logo, his letters and memoranda notes being signed ‘Mike the Bike’ with a small cycle drawing below. He was also regrettably accident prone, and managed to get knocked off his bike on what surely must be a record number of times? In 1990 Michael was Assistant leader on a botanical tour to the Peloponnisos, but returned to find that staffing at the Museum had been re-organised and his post no longer existed. For a number of years Michael then worked on short-term contracts, including with London Wildlife Trust as Project Manager/Warden of Gunnersbury Triangle Local Nature Reserve, as an Ecologist with a consultancy firm, as a Lecture Course Organiser at St Mary’s College, University of Surrey, and as an Ecologist with the London Ecology Unit. In 1997 he was appointed as a Tour Leader with Naturetrek Tours, taking groups to the Himalayas, Madeira, Portugal etc. Here, Mike had really found his niche and he could have continued as a travelling lecturer for a long time. As a brilliant teacher in the field, he had a talent for making plants and natural history interesting for a wide range of people. I recall meeting him unexpectedly on Wengen Station in the Bernese Oberland, surrounded by his group and recently returned from Sikkim; he was dressed in white with a topee, a cane and looking every inch the genial traditional botanical explorer. Unfortunately, after Sikkim the first signs of the illness which was to take his life became apparent. In spite of a courageous fight against the cancer, Michael died on 2 August 1999. Colleagues from the Department of Botany attended his funeral with representatives of his other interests. Michael described himself as having ‘eclectic tastes in music ranging from classical and folk to New Age etc.’ He had a deep interest in dancing with the Cotswold Morris, and with his 600 OBITUARY family and friends travelled to dance in Britain and also to Spain and France. Michael Mullin died sadly young leaving much promise and unfinished projects. We extend sympathy to his wife Anne and their daughter Fiona, and to his many friends — he will be sadly missed. On April 9 2000, a memorial bench to Mike at Gunnersbury Triangle L.N.R. was celebrated with ‘a toast and some tales’. Friends commented ‘it was a grand party and Mike would have loved it’ but Anne said ‘I am not so sure, he would have been gnashing his teeth at not being able to get at the beer.’ MARY BRIGGS 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. 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.rbge.org.uk/data/BSBI/Activities/watsonia.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: Mr D. A. Pearman, The Old Rectory, Frome St Quintin, Dorchester, Dorset DT2 OHF. Plant Records: the appropriate vice-county recorder, who should then send them to Dr C. D Preston, Biological Records Centre, Monks Wood, Abbots Ripton, Huntingdon, PEI7 21S. Obituaries: Mrs M. Briggs, 9 Arun Prospect, Pulborough, West Sussex, RH20 IAL. Back issues of Watsonia are available from the official agents for BSBI Publications, Summerfield Books (John & Sue Atkins), Summerfield House, High Street, Brough, Kirkby Stephen, Cumbria CA17 4BxX. Tel: 017683 41577. Fax: 017683 41687. E-mail: bsbipubs @beeb.net to whom orders for all issues prior to Volume 22 part 1 should be sent. Recent issues (Vol 22 part 1 onwards) are available from Mr M. Walpole, B.S.B.L., 68 Outwoods Road, Loughborough, Leicestershire, LE11 3LY. Watsonia August 2001 Volume twenty three Part four Contents RICH, T. C. G. What is fie vulneraria L. ae corbierei aes & a Cullen (Fabaceae)? KHALAF, M. K. & STACE, C. A. ‘The Aeincnen eiasente Cerastium tomentosum L. and C. biebersteinii DC. (on bene and their occurrence in the wild in Britain , HENDERSON, S. A. ‘The Sercatan acetal aah ne ee Cham. (Orchidaceae), Irish Lady’ s-tresses, on the Isle of Coll, Inner Hebrides BLACKSTOCK, N. & ASHTON, P. A. A re-assessment of the putative Carex flava agg. (Cyperaceae) hybrids at Malham Tarn (v.c. 64): A morphometric analysis RICH, T. C. G. & MOTLEY, G. S. Conservation of Britain’s biodiversity: Hieracium linguans (Zahn) Roffey (Asteraceae), Tongue Hawkweed “ie ar SPALTON, L. M. A new subspecies of Bromus hordeaceus L. (Poaceae) SMITH, R. E. N. & MARGETTS, L. J. Festuca longifolia Thuill. (Poaceae) in Devon BAILEY, J. P. Fallopia x conollyana The Railway-yard Knotweed NOTES Stace, C. A. The interspecific hae ae of the rhizomatous couches, ies Desv. (Poaceae) be Tennant, D. J. Salix caprea L. x S. “myrsinifoia Salisb. x S. phylicifoia L. in Perthshire Bull, A. L. A new species ea Rubus Sect. Micantes fRosaceue) Sonne es as a variant of R. criniger (E. F. Linton) Rogers Allen, D. E. Dates of publication of county floras PLANT RECORDS Book REVIEWS OBITUARY Published by the Botanical Society of the British Isles ISSN 0043-1532 Typeset by D. K. & M. N. SANFORD Printed in Great Britain by PALLADIAN PRESS, UNIT E, CHANDLERS ROW, PORT LANE, COLCHESTER, ESSEX CO1 2HG 469-480 481-491 493-503 505-516 517-523 525-531 - 533-538 539-541 543-547 547-549 549-550 550-551 553-592 593-597 599-600 lend 3 M HECKMAN BINDERY, toN G. Bound-To-Please* MAY 02 N. MANCHESTER, INDIANA 46962 Sa lab, tha es tl | SMITHSONIAN il LIBRARIES 3 9088 01167 5402 PETES Mon faite veces SET RY Dag hie ns eis * YAS Seip tL 4 comets Pop ubie haus ton apt ee De ey Ainek «pan epee tse Ape tga Vat . ‘ Straw se Val eS V ol a y APRESS gee PUA DAN RININ PLEAS we aegtins chp nla ews a7 IN PE ae PUP UD LAAD Whar ryan CPUS TET PORES SU Se y ts . * t yO SAMMY Fan VERE NES, bap nie LARA bal ah jai 3 1 4 Saar ser ¢ ‘i . usin 0 ae 5 LAUER Sabo ae Penn IF ACARI? § 2 gi : hosve ne, ALR r URE ie awe ; K Paine bi Pa alain para sy eas oy ; : : ; Gr Ps Reer ‘ 7 Sue Seabee ; é < Aj p AWA STARS : gyen g M ie mee hase DRE 1 VD aOR SAR PIN Meet ‘i isn 0 i ee pkO AY 4 i eon d kta epee edn ead oe oes : area Pies y hale . , : e ' 4 : i anes j A ry ) i ° \ . f aes Lanett EEE VEU VA GAG On, 4 i "Ae ‘ 7 Riek ERED PRS gr Hah Se f ; Rey BM UN Sy UNAS ae Sesion “Says SPURINY F The oe SHY fatis ? Sag suey Cggaaiu tpi! EA ns Pp NER ERUAEEDER aS Fe a gal ar tt aly Coen. SSAA ARATE NIP ap b 4 are Bars i Abril Ly Hy weap V in NE ne : : " Pa ne al Pres TEA NPR bee ag Be wert pe sox {eer oe Vy aa a ea Vn Vea, } h iene)