BtoSb Ac. ti' ■ - x- Shelve as: B.S.B.I Conference Report — A EUROPEAN FLORISTIC AND TAXONOMIC STUDIES CONFERENCE REPORT Edited by S. M. WALTERS with the assistance of C. J. KING B\0L0GY LIBRARY 101 B'JRR'ti- BA-L BOTANICAL SOCIETY OF THE BRITISH ISLES UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIQN BIOLOGY HAY 0 1 Digitized by the Internet Archive in 2018 with funding from BHL-SIL-FEDLINK https://archive.org/details/bsbiconferencere1519bota EUROPEAN FLORISTIC AND TAXONOMIC STUDIES B.S.B.I. Conference Report No. 15 A Conference held in Cambridge 29 June to 2 July 1974 Edited by S. M. WALTERS with the assistance of C. J. KING Published for THE BOTANICAL SOCIETY OF THE BRITISH ISLES By E. W. CLASSEY LTD © 1975 Botanical Society of the British Isles Published for the Botanical Society of the British Isles by E. W. Classey Ltd, Park Road, Faringdon, Oxon SN7 7DR ISBN 0 900848 90 1 Printed in Great Britain at the University Printing House, Cambridge (Euan Phillips, University Printer) CONTENTS \ Preface Conference programme page iv viii History of the British contribution to the study of the European flora. W. T. Stearn Floristic studies in Greece. W. Greuter Apomixis in a sterile hybrid species of Potentilla. R. Czapik The Veronica hederifolia group: taxonomy, ecology, and phylogeny. M. Fischer Caryosystematic study of some species of the genus Centaurea L. in the western Mediterranean basin. M. J. Fernandez- Morales and C. Gardou Studies in the genus Crocus. B. Mathew and C. A. Brighton Studies in the genus Myosotis. J. Grau Cytotaxonomy of the genus Leucanthemum in Yugoslavia. D. Pape§ Hybridization in yellow-flowered European Rorippa species. B. JONSELL Wild hybrids in the British flora. C. A. Stace Taxonomic problems in the fern genus Polystichum caused by hybridization. G. Vida and T. Reichstein Appendix I Summaries and short papers based on demonstrations at the Conference Appendix II Addresses of contributors 1 18 38 48 61 76 82 90 101 111 126 136 144 [iii] PREFACE The content of this Preface needs a little explanation. Normally in a Conference Report the reader might expect a separate Editor’s Preface and an Introduction, but on this occasion such a division seemed unwieldy, because both these contributions were expected from the pen of the same author! I hope, therefore, that I may be excused for conflating Preface and Presidential Introduction into a single block of prose. When the Botanical Society of the British Isles did me the honour of electing me President in 1973, 1 began to discuss with colleagues the possibility of holding a European Conference in Cambridge for the following year. There were several reasons why such a Conference seemed particularly appropriate. The first was that I had just been appointed to succeed John Gilmour as Director of the University Botanic Garden after his retirement on 30 September 1973, and such a Conference, using the facilities of the Botany School and the Botanic Garden, seemed to be a suitable inaugural activity for my first year. The second reason was that I had been much impressed by the joint Conference of the Society and the Royal Horticultural Society in 1972, the success of which owed much to the zeal and enthusiasm of my predecessor David McClintock; I asked myself what would be the most appropriate contribution which I might make to the series of Con¬ ferences held approximately biennially in recent years, and decided that the theme, then much under public discussion, of Britain as a part of the European Continent would be particularly appropriate. In 1973, the role of British taxonomic botany in Europe as a whole was very much in my mind because of my work as a member of the Editorial Committee of Flora Europaea ; we were all much exercised by the problem of financing the final stages of this co-operative European project, and were heartened by the evidence of real practical support from colleagues in many European countries. So the idea of the Society running a European Conference took shape , but an obvious difficulty remained. The Botanical Society of the British Isles has no paid officials, and even though the Honorary Secretary and all the other officers perform marvels of courteous efficiency in admin- [iv] PREFACE v istering the Society’s increasingly ambitious affairs, the burden of organizing a full-scale residential conference out of London could not and should not fall on the already overworked officers. I therefore early in 1973 approached Mrs Gigi Crompton, a Cambridgeshire botanist from whom amateur and professional alike in Cambridge have in recent years come to expect much devoted and efficient service, with the question - would she act as Honorary Conference Organizer? I felt that, if Mrs Crompton were prepared to act, I also could see my way to undertake the planning and preparation required; when she agreed, we were all set to begin the invitations, the bookings, the applications for financial help, and the many other chore s which needed attention , if B . S . B . I . members and invited foreign guests alike were to be satisfied. One other happy conjunction presented itself early in the planning. Several of our distinguished Honorary Members, including most appropriately John Gilmour himself, were achieving, or had already achieved, that ‘Valhalla’ state of 50 years’ Membership, and the Council of the Society agreed to honour seven such members by an invitation to the Conference Dinner in King’s College. In the event, two only were able to be present, but the Latin Oration in their honour delivered by the University Orator, Mr Patrick Wilkinson, to a text supplied in English by Mr John Raven (both Fellows of the College) added a special lustre to the proceedings. On the Sunday, after the formal sessions at which the main papers were presented, a picnic lunch was served in the Research and Experi¬ mental Area of the Botanic Garden. The weather was kind, and the occasion took on a Continental flavour almost of a French dejeuner - an illusion which was certainly helped by the bottles of wine unexpec¬ tedly available. In the afternoon the demonstrations were inspected in the Laboratory, and after tea Dr Yeo and I showed the participants round the Garden, concentrating in particular on the Conservation role which we hope to develop in the next few years. After dinner. Dr Perring regaled a relaxed and sympathetic audience with a light-hearted illustrated talk on some of the ‘Consequences of Mapping the Flora of Europe ’ . Much more could be recorded of those peripheral events which did much to make the Conference such a pleasant experience. My own selection is perhaps a peculiar one, in that I was behind the scenes at various times, and formally presiding at others. I certainly derived great satisfaction from the obvious success of the Conference in providing, as we had hoped, an opportunity for our European guests to meet and get to know our own members, both ‘professionally ’, as it were, in the sessions of the Conference itself, and more light-heartedly on the field excursions, at the reception in the Fitzwilliam Museum, or in the VI PREFACE much-appreciated bar of King’s College. It was a particular pleasure to find that, in addition to our guest speakers, several foreign members of the Society came specially for the Conference, and that our own internal membership was excellently represented by a very satisfactory mixture of amateur and professional, young and old. Several of our foreign guests expressed admiration, astonishment, and even envy, at the vitality and range of our Society - a reminder that, in the amateur tradition of field botany in Britain, we have a most valuable asset most other countries lack. In all we had about 140 participants at the Conference, including 16 guest speakers from 11 different European countries. In editing the material of the Conference for this publication, Mr Clive King and I have taken certain liberties, which should be briefly ex¬ plained. Firstly, we have slightly altered the title, introducing the word ‘ taxonomic ’, so that the general reader would be prepared for the range of subject-matter covered by the papers. The order of the papers has been retained more or less as presented, beginning with the two general papers, one by Dr Stearn on the British contribution to European floristic studies, and the other by Dr Greuter on the history of floristic studies in Greece (where the contribution of foreign botanical work is, of course, particularly important), and proceeding to the particular taxonomic studies of familiar European genera. In these more specialist invited papers, we have permitted, or even encouraged, some re-drafting in the interests of clarity, but have tried to retain the particular flavour of the papers as delivered to the Conference. One of the conditions imposed on, and willingly accepted by, the invited speakers was that they should give their paper in English; it was, I know, a matter of considerable admiration on the part of many B.S.B.I. members to find how many of their European colleagues could satisfy this requirement with some distinction. I must confess to a single area of disappointment - namely that, in spite of all our efforts, four of our invited foreign speakers were unable to get permission to attend. I had hoped that the general spread of the philosophy of detente in Europe would have made our task easier; it did not, alas, prove to be so, and the entirely free movement of people and ideas, even in such a ‘ neutral ’ science as ours, is still a dream for the future. It remains for me to now record my personal thanks, and the appreciation of the Society, to a number of organizations and individ¬ uals with whose help we were able to run such a successful Conference. In the first place, we would wish to thank the Royal Society for a generous grant towards the fares, and the British Council for subsis¬ tence and (in some cases) travel grants, for our invited foreign speakers. PREFACE Vll In addition, we gratefully acknowledge a grant for entertainment received from the University of Cambridge, and permission from Professor Percy Brian, who welcomed guests and participants at the opening session, to use the facilities of the Botany School and the Botanic Garden during the whole Conference. Thanks are also due to the Director of the Fitzwilliam Museum, Professor Michael Jaffe, and members of his staff, who arranged with us the excellent evening reception to see the remarkable exhibition of flower drawings from the Broughton Collection. The two field excursions which followed the main meetings were greatly appreciated, in spite of ‘Atlantic’ weather appropriate to the vegetation types visited, and our special thanks must be recorded to our members, Mr R. P. Libbey, Mr P. D. Sell and Mr E. L. Swann, leaders of the Norfolk field trip, and to Dr F. H. Perring and Mr M. Schofield (Monks Wood Experimental Station), and Mr G. Mason (Woodwalton Fen Nature Reserve) for the second day. A special word of thanks should also be recorded to Mr John Raven, who helped in three quite separate capacities, firstly in preparing the Latin Oration, secondly with botanical queries arising from the selection of the Broughton Collection drawings, and lastly in kindly receiving at his beautiful house, Docwra’s Manor, Shepreth, a small but highly appre¬ ciative party of foreign guest botanists and their hosts on an extra excursion at the end of the Conference. Last, but by no means least, I must record my appreciation of the excellent work of the staff of my own College, King’s, which housed and fed the Conference. It was very pleasant, after the event, to receive so many words of thanks particu¬ larly for the catering arrangements. I have already expressed my total indebtedness to our excellent Conference Organizer, Mrs Gigi Crompton. To help her in the complex hour-by-hour detailed running of the Conference she had many willing colleagues, four of whom deserve particular mention - Mrs Mary Briggs, Mr and Mrs John Dony and Mr J. C. Faulkner. To these and all the ‘ B.S.B.I. team ’ I can only say, in tribute, that the President was left with no worries and no unexpected duties and therefore, to his great relief, found he had time to relax and enjoy the whole occasion! A final, and special word of thanks is due to my assistant editor, Clive King, whose skill and patient attention to detail are revealed in this end-product of many hours’ work. S. M. WALTERS CONFERENCE PROGRAMME Held at King's College, the Botany School, and the Botanic Garden, University of Cambridge. Saturday, 29 June OPENING SESSION Chairman: Professor P. W. Brian, F.R.S., Professor of Botany, University of Cambridge 1 1 . 15- 1 1 .20 president’s opening remarks Dr S. M. Walters 11.20- 12.20 HISTORY OF BRITISH CONTRIBUTION TO EUROPEAN STUDIES Dr W. T. Stearn 12.20- 13.00 FLORISTIC STUDIES IN GREECE Dr W. Greuter 13.15- 14.15 Lunch, King’s College STUDIES IN COMMON EUROPEAN GENERA Chairman: Professor C. D. Cook, Professor of Botany, University of Zurich 14.30-15.10 APOMIXIS IN A STERILE HYBRID SPECIES OF POTENTILLA Dr R. Czapik 15.10-15.50 THE VERONICA HEDERIFOLIA G ROUP Dr M. Fischer 15.50-16.20 Tea Chairman: Professor D. H. Valentine, Professor of Botany, University of Manchester 16.20- 17.00 CARYOSYSTEMATIC STUDY OF SOME SPECIES OF THE GENUS CENTAUREA Dr C. Gardou 17.00-17.40 aspects of crocus TAXONOMY Mr B. Mathew and Miss C. Brighton 17.40-18.20 STUDIES IN THE GENUS MYOSOTIS Professor Dr J. Grau 19.30 conference dinner, King’s College, in honour of the Honorary Members of the Society of 50 years Membership standing [ viii ] PROGRAMME IX Sunday, 30 June CYTOTAXONOMIC AND HYBRIDIZATION STUDIES Chairman: Professor D. H. Valentine 09.30-10.10 HYBRIDIZATION IN YELLOW-FLOWERED EUROPEAN RORIPPA SPECIES Dr B. Jonsell 10.10-10.50 WILD HYBRIDS IN THE BRITISH FLORA Dr C. Stace 10.50-11.20 Coffee 11.20-12.00 TAXONOMIC PROBLEMS IN THE FERN GENUS POLYSTICHUM CAUSED BY INTERSPECIFIC GENE EXCHANGE Dr G. Vida 12.00-12.40 CYTOTAXONOMY OF THE GENUS LEUCANTHEMUM IN YUGOSLAVIA Dr D. Papes 13.00-14.00 Buffet lunch, Botanic Garden 14.00-16.00 DEMONSTRATIONS IN THE BOTANIC GARDEN LABORATORY Opening remarks by Mr J. S. L. Gilmour, Cambridge 16.00-16.30 Tea 16.30-18.00 THE ROLE OF BOTANIC GARDENS IN CONSERVATION AND TAXONOMIC RESEARCH Dr S. M. Walters: perambulation of the Botanic Garden 19.30 Dinner, King’s College 21.00-22.00 THE CONSEQUENCES OF MAPPING THE FLORA OF EUROPE (Illustrated talk in Keynes Hall, King’s College) Dr F. H. Perring, Monks Wood Experimental Station Monday, 1 July 09.00-18.00 excursion to saltmarsh, fen and bog communities of NORFOLK Leaders: Mr R. P. Libbey, Mr P. D. Sell, Mr E. L. Swann 19.30 Dinner, King’s College 21.00-22.30 reception in Fitzwilliam Museum at inauguration of exhibition of botanical illustration from the Broughton Collection Tuesday, 2 July 09.00-17.00 excursion to monks wood experimental station, includ¬ ing THE BIOLOGICAL RECORDS CENTRE, AND WOOD WALTON NATURE RESERVE Leaders: Dr S. M. Walters, Dr F. H. Perring, Mr M. Schofield 19.30 Dinner, King’s College ■ I HISTORY OF THE BRITISH CONTRIBUTION TO THE STUDY OF THE EUROPEAN FLORA William T. Stearn A conference on floristic studies in Europe must necessarily be inter¬ national both as regards its participants and its subject-matter. My allotted task for this Conference has been to provide a cursory survey of the British contribution to these studies. The contribution of any one nation to knowledge of the European flora cannot, however, be con¬ sidered entirely apart from the work of others, because plants attained their ranges under the influence of factors by no means the same as those which have determined modern national boundaries; hence they often extend across them, thereby being made vegetable citizens of more than one country. Moreover, modern botany has been an inter¬ national science from its very beginning. Work on the plants of one country has so often become relevant to those of others that botanical scholars have always tried to keep abreast of the contributions of their fellow-workers elsewhere, examining, assessing, criticizing, using and appreciating them according to their merits; in this way all these botanists have increased and shared a common inheritance of know¬ ledge. Their work collectively manifests the cultural unity of Europe. Situated in the north-western corner of Europe, Britain possesses, on account of its Atlantic climate and its greatly varied topography, a diversity of plants less common elsewhere in northern Europe; the British contribution to European floristic studies has been based pri¬ marily upon these but has certainly not been confined to them. It began in the sixteenth century when botany developed as an independent study out of herbalism, being an outgrowth of enquiry into vegetable materia medica. It pushed forward under the general questioning spirit and dissatisfaction with the immediate past which then found its theological expression in the works of Erasmus and Luther and its scientific expression in those of Copernicus, Galileo, Vesalius and lesser men. Among these men was William Turner (1508-68), the ‘ father of English botany’. The area in which modern botany originated has been indicated on a map published by Isaac Henry Burkill (1870-1965) in 1953 and included in his posthumous Chapters on the History of Botany in India (1966). On his map (Fig. 1) Burkill marked all the places at which 2 THE BRITISH CONTRIBUTION Fig. 1. The area in which modern botany originated (based on Burkill, 1953). Places named are those where three or more botanical books were printed between 1500 and 1623. Numerals indicate places where one or two botanical books were printed: 1. Leyden; 2. Middelburg; 3. Arnhem; 4. Louvain; 5. Berlin; 6. Leipzig; 7. Bautzen; 8. Gorlitz;9. Prague; 10, 11. Niirnberg and nearby Altdorf; 12. Augsburg; 13. Bergamo; 14. Mantua; 15. Padua; 16. Ferrara; 17. Bologna; 18. Florence; 19. Naples; 20. Salamanca. Bologna under Luca Ghini and Montpellier under Rondelet were important centres of botanical teaching. (Reproduced from Proc. Linnean Soc. London , 169: 179 (1958) by courtesy of the Linnean Society of London.) botanical books were printed between 1500 and 1623, the year of Gaspard Bauhin’s Pinax; this important work listed virtually all the plants then known, with their numerous synonyms, and was compiled from virtually all the books then published. The area encompassing then- places of publication stretches from Salerno in Italy northward to London, with an out-station at Salamanca in Spain. At this period, as Burkill noted, the printing of a book at a given place is reasonable evidence of its having been written in the neighbourhood and also of enough interest there to justify its printing. The delimited area covers W. T. STEARN 3 territory occupied by the Dutch, English, French, Germans and Italians, but Latin provided them with a common language. Because of the close association of botany with medicine, those universities having the highest repute for medical teaching, Padua and Montpellier, influenced most the development of botany. Students from distant lands, going to them for anatomical and other medical instruction, learned the new technique of herbarium-making invented by Luca Ghini and the old classical names of plants as interpreted by their teachers. These two universities owed their sixteenth-century eminence to a spirit of enquiry which could flourish only in a tolerant or at least non-hostile academic environment such as is more commonly associated with a trading and manufacturing community than with a purely academic or ecclesiastical one. Money was money for the Venetian republic, of which Padua was the university, whatever the race, creed or beliefs of those providing it. Pecunia non olet. WILLIAM TURNER For England, indeed for the British Isles, botany began with William Turner, ‘unquestionably the earliest writer among us, that discovered learning and critical judgment in the knowledge of plants as Pulteney said in 1790. His Libellus de Re herbaria (London, 1538) and the Names ofHerbes in Greke, Latin , Englishe, Duche and Frenche (London, 1548) are the first two printed books relating to the British flora which have any claim to originality. Only two copies of the first and less than twelve of the second are known to have survived, their format being too humble to ensure careful preservation, but they are available in a low-priced facsimile, with extensive introductory matter, published by the Ray Society in 1965. Turner followed these in 1551-68 with A New Herball , which is fortunately not so scarce. His aim was essentially medicinal, to equate English vernacular names of plants with Latin equivalents and thereby to make such plants available as materia medica. Thus incidentally he recorded for the first time some 238 species of British plants. The publication of his works implies a public for them in England; indeed, Turner himself stated in 1538, with probable exag¬ geration, that ‘there are six hundred of us Englishmen who in this kind of learning would precede me (as the saying goes) on white horses’. Turner was a doughty Protestant controversialist, the author of such books as The Huntyng and Fyndyng out of the Romyshe Fox , which in the troubled sixteenth century meant the risk of being burnt as a heretic (which Turner, unlike his friends Thomas Ridley and Hugh Latimer, escaped through his second exile), and of having one’s publications banned, confiscated and burnt, which happened. His 4 THE BRITISH CONTRIBUTION Libellus (1538) shows him to have been acquainted with the works of Ruel and Brunfels. In 1540 he and his wife found it prudent to leave England and go into exile. His wanderings took him up the Rhine, his wife probably being left at one of the Rhineland cities where English exiles congregated, and into Italy, where he certainly visited Venice, Ferrara and Bologna, studying here under the great teacher and reluctant writer Luca Ghini; he also visited Milan and Como; thence he went to Switzerland, crossing the Alps to Zurich, here meeting Conrad Gessner, then going to Basel and down the Rhine to Cologne. Presumably here he rejoined his wife; their son Peter was born in 1542. They later settled in East Friesland, Turner there serving the Earl of Emden as his physician, then moved into the Low Countries. After the death of Henry VIII in 1547, Turner returned to England and became physician to the Protector Somerset at Syon House across the Thames from his own garden at Kew. In 1548 he published his Names of Herb es which contains many references to plants seen on his continental travels, and in 1551 the first part of his New Herball. The accession of Queen Mary in 1552 and the consequent persecution of Protestants caused Turner, his wife and their three young children to leave England again for the safety of Germany, and he did not return again until the accession of her Protestant half-sister Queen Elizabeth I in 1558. He died in 1568. The influence of these enforced travels on the continent of Europe upon Turner’s publications from 1548 onwards is very evident in the references he makes to plants growing there. They indicate correlation between the plants of England and elsewhere. Thus Turner, through his direct acquaintance with continental plants, litera¬ ture and botanists, brought British botany from the start into the main stream of European botanical learning. The major contribution which any scholar could make then to the floristic botany of Europe was to record the plants of his homeland. Turner did not, however, set out to record all the plants of England; his aim was to provide correct names for those that could be used in medicine, the same intent as that of his contemporaries Euricius Cordus and Leonhart Fuchs. According to W. A. Clarke, First Records of British Flowering Plants , 2nd ed. (1900), his works contain notices of 238 native flowering plants. Since his interest was primarily medicinal, they include few grasses and no sedges but many members of the Compositae and Umbelliferae. The association of scientific names with vernacular names undoubtedly aids the popularization of botany: Turner recorded such English names as he knew and, when none existed, he proposed names of his own, writing as he was ‘unto the English my countremen an Englysh herball’. Thus of the herb called by herbalists Saponaria and grown in German gardens he remarked that W. T. STEARN 5 ‘I never sawe it in England, therfore I know no englishe name for it. However if we had it here, it myght be called in english sopewurt or skowrwurt.’ Thus Saponaria officinalis acquired its English name Soapwort. Similarly he proposed that the herbarist’s Pyrola ‘maye be called in englishe wyntergrene’ and that Larix ‘maye be called in englishe a Larche tree’. Turner’s most important immediate successor was Matthias de l’Obel (1538-1616) from Flanders, the author of Plantarum seu Stirpium Historia (1576) and physician to William the Silent. He settled in England about 1586 and lived here until his death. Clarke credits him with no less than 80 first records of British plants. De l’Obel, who is commemorated by the genus Lobelia, had a hand in the preparation of John Gerard’s Herball (1597). This added 182 species to the British list. Thus, as Clarke pointed out, by 1600 about 500 species of British plants were known and described and most had been correlated with those of the Continent; they included 21 grasses and three sedges. THOMAS JOHNSON AND THE SOCIETY OF APOTHECARIES Turner had collected much information about British plants on his travels in England. Early in the seventeenth century medical men and apothecaries began to make deliberate ‘ public excursions and simpling- tours ’ in order to acquaint themselves at first hand with medicinal and other plants. These botanical excursions became a regular activity of the Society of Apothecaries soon after they had dissociated themselves in 1617 from the Grocers’ Company and become an independent body ‘Corporate and Politic’ with a charter from King James I. The best documented journeys are those of Thomas Johnson (1604-44) and his friends because he published accounts describing their travels and finds. In 1629 they made a trip into Kent, followed by one to Hampstead Heath, described in Latin in Johnson’s Iter Plantarum Investiga- tionis...in Agrum Cantianum Anno Dom. 1629. . .Ericetum Ham- stedianum (1629). In 1632, they made a longer Kent journey described in Johnson’s Descriptio Itineris Plantarum Investigationis. . .in Agrum Cantianum Anno Dom. 1632 (1632). These two rare booklets are now available in a facsimile, together with translations and commentaries, edited by J. S. L. Gilmour (1972). Such excursions were evidently jovial, as well as scientific, and educational, and the Society of Apothecaries continued them down to 1834, which resulted in the flora of the London region’s being fairly well recorded before its extermina¬ tion or reduction by urban development. Some excursions extended to fifty miles from St Bartholomew’s Hospital or Apothecaries’ Hall, London, the usual meeting places. For 55 years they were conducted 6 THE BRITISH CONTRIBUTION by Thomas Wheeler (1754-1847), a teacher much esteemed both for his geniality and learning. He remained an active field botanist while in his seventies and possessed a profound knowledge of grasses, sedges, rushes, umbellifers and Compositae as well as of less critical groups. On these excursions, when over 70, he wore a threadbare black coat and breeches and long leather gaiters. It is related by Field and Semple that on one occasion, when the party was returning in an open coach near Maidstone, Wheeler sat on the box beside the driver, with his hat off, his thin light hair blowing about his face and his large spectacles on his nose, alternately laughing and chatting with the driver and diving into his hat with his huge pocket-knife, separating and examining a bundle of wild plants. Such a figure naturally attracted attention along the road, and when stopping at a turnpike-gate, the party were rather surprised by the evident interest and eagerness of the toll-keeper, as he scratched his head, and, pointing to Mr. Wheeler, exclaimed in his blunt Kentish dialect, ‘ So ye ha’ got him at last!’ This was incomprehensible to all the party until they arrived at a small inn close to the parish of Barming, where they read a placard offering a reward for the capture of an escaped lunatic! The American botanists John Gill Lemmon and his wife Sara, collecting plants where Indians were on the warpath, later owed their lives to being similarly identified through their addiction to such a mad activity. In 1634 the Apothecaries led by Johnson made an even longer journey, going to Bristol, by way of Reading, Marlborough and Bath. The most important of their journeys, however, was to North Wales in 1639, by way of Chester, Holywell, the Great Orme’s Head, Aberconway, Bangor and Caernarvon, ascending Snowdon on the 3rd of August, then to Machynlleth and Montgomery, and back to England, through Ludlow, Hereford, Gloucester and Oxford. Edward Morgan, one of the party, was a Welshman. Without him as an interpreter their journey, difficult enough on account of the wildness of the ways through the mountains, might have yielded little on account of linguistic difficulties and suspicion of strangers by its equally wild inhabitants, who in fact treated them very hospitably, as they later did Willughby and Ray, the latter making the remark, ‘The Welch people generally are extremely civil and well bred, very honest and courteous to strangers ’ . Johnson’s purpose in making such journeys and listing the plants found was to assemble material for a complete descriptive account of British plants in collaboration with his friend John Goody er (1592-1664), who was no such traveller, but a diligent, scholarly and sharp-eyed student of Hampshire plants, described in 1657 as ‘the ablest Herbarist now living in England ’. The Civil War of 1642-6 tragically brought their plan to nothing. Johnson joined King Charles’s forces at Oxford and fought as a Royalist in the defence of Basing House near Basingstoke W. T. STEARN 7 during its long siege by Parliament forces ; here he died in September 1644 from a gunshot wound received during a sortie. The production of a much-amended edition of Gerard’s Herball in 1633, which remained a standard work well into the eighteenth century and became known as Gerardus emaculatus , and the accounts of his travels with their lists of plants stand to his credit. In these he recorded some 170 species as new for the British flora. THE INTERREGNUM, 1640-67 The death of so enthusiastic and learned an apothecary and field botanist as Johnson at the age of 40 was a bad but not fatal blow to British botany at a time when interest in it was steadily growing. Others, though not of the same ability, kept it alive. Thus six years later William How produced Phytologia Britannica (1650), an uncritical compilation based on Johnson’s Mercurius botanicus (1634) but con¬ taining new records from How’s friends, which was popular because there existed nothing better. This added 16 species to the British list. How died in 1656 before he could produce a new edition, but his publisher obtained the services of a physician, Christopher Merrett (1614-95). Merrett could not travel much himself, but he did the best he could in getting additional information. Thus he employed an old soldier, Thomas Willisel, used to rough living and travel on foot, uneducated but with a keen eye for plants, and sent him every year for five years on collecting trips all over England. In 1666 Merrett published a Pinax Rerum naturalium Britannicarum , with a second edition in 1667; this added 46 species. Among his friends was William Harvey, famous for his work on the circulation of the blood, but also keenly interested in botany, and thus characteristic of the many British physicians and apothecaries who provided a public for such books on British plants. Merrett took some material from John Ray’s Catalogus Plant arum circa Cantabrigiam nascentium (1660), a little book which Charles Raven has described as ‘ initiating a new era in British botany ’. JOHN RAY A conference in Cambridge on European floristic botany can honour no man more fittingly than John Ray (1627-1705), who travelled widely both in Britain and on the Continent for the study of plants and whose works in the seventeenth century provided a foundation for those of Linnaeus in the eighteenth. Ray, like Linnaeus, was of humble rural origin, his father being a blacksmith at Black Notley, Essex. He 8 THE BRITISH CONTRIBUTION received, however, a good education at the grammar school in nearby Braintree and, supported by a Braintree scholarship for ‘hopeful poor students’ of ‘sober and Christian conversation’, he entered the Uni¬ versity of Cambridge in 1644, the year of Thomas Johnson’s death. He graduated B.A. in 1647/8, was appointed Greek lecturer in 1651, mathematical lecturer in 1653, and humanities lecturer in 1655 and was ordained as a clergyman in the Church of England in 1660. Botany had then no official place in university education at Cambridge, but there existed nevertheless a group which studied it for their own private interest. Ray’s own study began in 1650 when convalescence from illness gave him leisure to examine the plants growing around Cam¬ bridge, aesthetic delight leading on, as it so often has, to critical research into their characteristics and differences. In 1654 a young nobleman Francis Willughby (1635-72) became his pupil. Their friend¬ ship blossomed into a fruitful scientific collaboration which ultimately enabled Ray to devote all his time to natural history. In 1660 he published his Catalogus Plantarum circa Cantabrigiam nascentium , a list of the plants growing around Cambridge, preparatory to a list of those of Britain as a whole. This was intended to replace How’s Phytologia Britannica ; ‘to which purpose’, he told Willughby in 1660, ‘ I am now writing to all my friends and acquaintance who are skilful in Herbary to request them this summer to search diligently his country for plants and to send me a catalogue of such as they find, together with the places where they grow ’. Thus began the co-operative record¬ ing, area by area, of the distribution of British plants. This culminated in the nineteenth century in the publication of two works by Hewett Cottrell Watson (1804-81), Cybele Britannica (1847) and Topographical Botany (1873), followed in the twentieth century by F. H. Perring and S. M. Walters, Atlas of the British Flora (1962) based on one and a half million field records contributed by some 1,600 collaborators, most of them amateur botanists. This enquiry by Ray led to his Catalogus Plantarum Angliae (1670) and that to his Synopsis methodica Stirpium Britannicarum (1690). These works rested upon a first-hand acquaintance with the British flora gained by extensive travel as well as by extensive correspondence. Thus in 1660 Ray and Willughby journeyed to the north of England and over to the Isle of Man. In 1661 Ray and Philip Skippon went to Scotland, travelling along the eastern side of England by way of Lincoln, Harrogate, York, Newcastle, Berwick, to Edinburgh, then westward to Glasgow and by way of Carlisle and the Lake District back to Cambridge. In 1664 Ray and Willughby made a journey through Wales and the south-west of England. Such travel involved much discomfort and some danger, but Ray, Skippon and Willughby were unique only W. T. STEARN 9 in undertaking it for the study of natural history; many a gentleman and occasionally even a young lady such as the courageous and inquisitive Celia Fiennes set out then to visit the great houses, the remains of antiquity, and the market towns, to note the manufactures and general state of their country. The habit of touring their native land [states Esther Moir in her The Discovery of Britain ] began in the sixteenth century: it is a Tudor phenomenon. . .the motive force was pride in the greatness of Tudor England, and a curiosity both in the historic roots of that greatness and its contemporary manifestations. Bound by no tradition or convention, lacking the established routes and rituals of the Grand Tour, it became a popular pastime amongst gentlemen of leisure to travel for weeks, even months, in the discovery of their own country. The roads they traversed were mostly appalling, rough in dry weather, deep with mud and treacherous in wet weather, sometimes ‘little better than ponds of liquid dirt with a scattering of loose flints just sufficient to lame every horse that travelled them’, as Arthur Young described Wiltshire roads, and the flea-infested inns were often little better as regards comfort. Enthusiasts like Ray and Willughby took such dis¬ comforts and hazards as a matter of course in getting to know the natural history of their native land. In all Ray, with the aid of his friends and correspondents, added some 200 species to the British list. These helpers were numerous (cf . Stearn, 1973, pp. 34-41) and among the most enthusiastic of them was a young Welsh scholar, Edward Lhwyd (1660-1709), who not only searched Snowdon for rare plants, finding among them Lloydia serotina , Ceras- tium arcticum and Isoetes lacustris , but also lesser mountains and hills, including Craig Breiddin (his ‘Craig Wreidhin’) in Montgomeryshire, where he discovered Potentilla rupestris and Veronica spicata subsp. hybrida. The contemporary rendering of his surname as ‘Lloyd’, ‘Floyd’, ‘Lhuyd’ and ‘Luidius’ may be confusing. The natural son of an eccentric and dissolute squire Edward Lloyd and a Caernarvon¬ shire gentlewoman Bridget Pryse, he retained the usual anglicized patronymic ‘ Lloyd ’ (whence the generic name Lloydia commemorating him) until about 1686, then adopted with antiquarian zeal the older Welsh spelling ‘ Lhwyd ’ (grey), which he latinized as Luidius to accord with the pronunciation of ‘ Lhwyd ’. The university of Oxford and Jesus College, Oxford having inexplicably refused after his death to purchase his many volumes of notes and drawings relating to the natural history and antiquities of Wales and Ireland, these passed into private hands and were destroyed in two disastrous fires, but botanical specimens from him are preserved in the British Museum (Natural History) and Oxford herbaria (cf. Clokie, 1964; Dandy, 1958). The big crisis in Ray’s life came in 1662 with the enforcement of the 10 THE BRITISH CONTRIBUTION pernicious Act of Uniformity; rather than assent to this, which implied that an oath was not binding, some 2,000 clergymen of the Church of England forfeited their livings and among them was Ray, with the prospect before him of extreme poverty and frustration. Willughby came to his aid and by thus freeing him for research and travel made possible his further immense contribution to European botany. To¬ gether they had travelled many thousands of miles around England and Wales. In 1663 they set out on a Continental tour lasting until 1666, which took them through the Low Countries, Rhineland Germany, Austria, Italy, Switzerland and France and out to Sicily and Malta. No botanist before, except Clusius, had had the opportunity of acquir¬ ing so wide a knowledge of European plants, and Ray made full use of it. Willughby died prematurely at the age of only 36, leaving an annuity to Ray, which enabled him to devote most of his life thereafter to preparing the works on birds, fishes, insects and plants they had planned together. The most massive of these was his Historia Plan- tarum in three thick folio volumes (1686, 1688 and 1701), which incorporates the results of their European travels. This work and Dillenius’s edition (1724) of Ray’s Synopsis methodica Stirpium Bri- tannicarum constitute a very important contribution to European floristic botany and have a lasting value because of the information they provided Carl Linnaeus (1707-78) when preparing his Species Plan- tarum. From the Synopsis Linnaeus derived most of his knowledge of the occurrence of species in Britain. The Historia gave him details about distribution in Europe generally. Thus Linnaeus stated that Epimedium alpinum ‘habitat in Alpium Euganeorum, Ligurinorum, Pontebarum umbrosis’. This information came from Ray’s Historia Plantarum 2: 1330 (1688), where Ray states that he himself had observed the plant ‘non longe a Ponteba oppido quod territorium Venetum et Imperiale distermat ’ . In 1754 Linnaeus published a dissertation. Flora Anglica , which attempted a correlation of the Swedish and British floras by applying binomial names to the plants listed in the Synopsis (1724), which used, of course, pre-Linnaean phrase-names (cf . Stearn, 1973, pp. 42-68). His ‘ Caput Quartum ’ deals concisely with the history of botany in Britain. Here he noted that at the beginning of the seventeenth century the English nation seemed little fitted for, and indeed almost alien to, the study of botany, whereas by its end they had produced as many botanists as the whole of Europe, and among these Ray stood at the head. As a result of their combined efforts, by 1700 some 970 species of flowering plants were known from Britain; collecting and study by their successors in the eighteenth century raised this to about 1 , 145 by 1800. W. T. STEARN 11 GENERAL CONTRIBUTION The work of Ray illustrates three of the five main kinds of contribution that botanists of a given country can make to the floristic botany of Europe as a whole. I. The investigation of the plants of their own country, defining and naming its species and ascertaining their distribution and ecology, is exemplified by Ray’s Catalogus Plantarum circa Cantabrigiam nascentium (1660), his Catalogus Plantarum Angliae (1670) and his Synopsis methodica Stirpium Britannicarum (1690, 1696). Such investigation has often brought to notice species hitherto unknown but later found elsewhere. The type-locality of a species is the place whence came the material on which the protologue of its name is based. When this is merely stated in broad terms such as ‘Europa’, ‘India’, ‘Nova Hollandia’, it may nevertheless be possible to make this more precise by ascertaining such information as the name of the collector, his routes and collecting areas, or by excluding the areas from which no material could be available at the time of publication. A locality thus delimited within a wider area originally designated has become known, particularly among ornitholo¬ gists, as the ‘restricted type-locality’. Thus the restricted type-locality for Geranium lucidum L., of which Linnaeus gave the distribution as ‘Habitat in Europae rupibus umbrosis’, is Torsburg on the island of Gotland, where Linnaeus collected it on 3 July 1741 (cf. Biol. J. Linn. Soc. 5: 7-9 (1973)). A remarkably large number of species were first named from British material, thanks to the work of Ray, Miller, Hudson, Sibthorp, Curtis, J. E. Smith, Babington and others. Species with their type-locality in Britain include: Hymenophyllum tunbrigense (L.) Smith Asplenium viride Hudson Cystopteris dickieana Sim Woodsia alpina (Bolton) Gray Dryopteris pseudomas (Wollaston) Holub & Pouzar (D. borreri (Newman) Tavel) Ranunculus circinatus Sibth. Meconopsis cambrica (L.) Vig. Fumaria occidentalis Pugsley Fumaria purpurea Pugsley Rhynchosinapis monensis (L.) Dandy Rhynchosinapis wrightii (O. E. Schulz) Dandy Cochlearia alpina (Bab.) H. C. Watson Cochlearia micaceaE. S. Marshall Cochlearia scotica Druce Cochlearia anglica L. Cardamine flexuosa With. Arabis brownii Jordan Viola lactea Smith Viola lutea Hudson Hypericum e lodes L. Cerastium nigrescens Edmondston ex H. C. Watson 12 THE BRITISH CONTRIBUTION Cerastium pumilum Curtis Sagina maritima Don Sagina boydii F. B. White Herniaria ciliolata Melderis Chenopodium ficifolium Smith Chenopodium botryodes Smith Atriplex glabriuscula Edmondston Halimione pedunculata (L.) Aellen Arthrocnemum perenne (Miller) Moss Salicornia dolichostachya Moss Salicornia ramosissima Woods Salicornia pusilla Woods Tilia cordata Miller Malva pusilla Smith Linum bienne Miller Genista anglica L. Trifolium ochroleucon Hudson Trifolium dubium Sibth. Trifolium occidentale Coombe Vida orobus DC. Potentilla fruticosa L. Potentilla anglica Laich. Rubus microspecies (more than 110) Alchemilla conjuncta Bab. Alchemilla minima Walters Rosa arvensis Hudson Rosa sherardii Davies Sorbus microspecies (about 15) Sedum anglicum Hudson Sedum forsterianum Smith Umbilicus rupestris (Salisb.) Dandy Ribes spicatum Robson Drosera anglica Hudson Physospermum cornubiense (L.) Burnat Berula erecta (Hudson) Coville Oenanthe fluviatilis (Bab.) Coleman Ligusticum scoticum L. Euphorbia portlandica L. Polygonum minus Hudson Rumex hydrolapathum Hudson Rumex palustris Smith Ulmus glabra Hudson Ulmus procera Salisb. Salix nigricans Smith Limonium humile Miller Limonium binervosum (G. E. Smith) C. E. Salmon Limonium recurvumC. E. Salmon Limonium transwallianum (Pugs- ley) Pugsley Limonium paradoxum Pugsley Primula scotica Hooker Primula vulgaris Hudson Centaurium latifolium (Smith) Druce Centaurium littorale (D. Turner) Gilmour Gentianella anglica (Pugsley) E. F. Warb. Verbascum virgatum Stokes Euphrasia microspecies (about 14) Calamintha sylvatica Bromf. Galeopsis speciosa Miller Scutellaria minor Hudson Phyteuma tenerum R. Schulz Senecio cambrensis Rosser Arctium pubens Bab. Carduus tenuiflorus Curtis Hieracium microspecies (more than 100) Alisma lanceolatum With. Eriocaulon aquaticum (Hill) Druce Polygonatum odoratum (Miller) Druce Scilla verna Hudson Juncus castaneus Smith Luzula forsteri (Smith) DC . W. T. STEARN 13 Luzula sylvatica (Hudson) Gaudin Allium ampeloprasum L. Narcissus obvallaris Salisb. Epipactis purpurata Smith Epipactis leptochila (Godf .) Godf . Epipactis dunensis (T. & T. A. Stephenson) Godf. Epipactis phyllanthes G. E. Smith Ophrys apifera Hudson Ophrys sphegodes Miller Orchis purpurea Hudson Dactylorhiza fuchsii (Druce) Soo Dactylorhiza praetermissa (Druce) Soo Dactylorhiza purpurella (T. & T. A. Stephenson) Soo Eleocharis multicaulis (Smith) Smith Carex laevigata Smith Care x binervis Smith Carex extensa Gooden. Carex sylvatica Hudson Carex depauperata Curtis ex With. Carex rostrata Stokes Carex riparia Curtis Carex pendula Hudson Carex strigosa Hudson Carex disticha Hudson Carex divisa Hudson Carex divulsa Stokes Carex spicata Hudson Carex curt a Gooden. Carex ovalis Gooden. Carex pauciflora Lightf. Festuca pratensis Hudson Festuca tenuifolia Sibth. Puccinellia maritima (Hudson) Pari. Puccinellia rupestris (With.) Fer- nald & Weatherby Catapodium rigidum (L.) C. E. Hubbard Poa flexuosa Smith Poa balfourii Parnell Poa subcaerulea Smith Bromus erectus Hudson Bromus ramosus Hudson Brachypodium sylvaticum (Hudson) Beauv. Agropyron donianum F. B. White Hordeum marinum Hudson Helictotrichon pratense (Hudson) Pilger Deschampsia setacea (Hudson) Hackel Calamagrostis scotica (Druce) Druce Agrostis setacea Curtis Agrostis tenuis Sibth. Alopecurus myosuroides Hudson Spartina maritima (Curtis) Fernald This list does not include names based on British material which have passed into synonymy because the species concerned had earlier been described on the Continent. Thus Bromus britannicus I. A. Williams (1929) was found to have been earlier described in Sweden as B. lepidus O. R. Holmberg (1924), although its recognition as a species hitherto unnoticed in Britain reflected great credit on that enthusiastic, cultured and sharp-eyed amateur botanist Iolo Aneurin Williams. Inclusion of taxa now commonly given infraspecific rank and first detected in Britain, e.g. Sparganium neglect um Beeby (= S. erect um subsp. neglectum (Beeby) Schinz & Keller), Veronica humifusa Dick- 14 THE BRITISH CONTRIBUTION son (= V. serpyllifolia subsp. humifusa (Dickson) Syme), would also have lengthened the above list. II. The second kind of contribution is that made by botanists visiting countries other than their own and diligently collecting material for study. Here Ray had a predecessor in Thomas Penny (c. 1530-88), like Turner and Ray a Cambridge scholar. In 1565 he turned from theology to medicine and went to Zurich; here he met Conrad Gessner, to whom he gave drawings and specimens of British plants. He also corresponded with Clusius and gave him drawings reproduced in his Rariorum aliquot Stirpium per Pannoniam Historia (1584) and Rariorum Plan- tarum Historia (1601). These include Anemone narcissiflora found in the Jura near Geneva and Rubus chamaemorus from Ingleborough (‘ Engelborow ’) in Yorkshire, but the most interesting of his finds is that called Myrto-cistus pennaei by Clusius, ‘ex Balearium insularum majore, vulgo Majorica nuncupata’, which is now Hypericum bal- earicum, and probably the first species recorded from the Balearic Islands. There being no terminal y in Latin, Clusius referred to Penny as ‘Thomas Pennaeus, Londinensis medicus’. The main British collecting effort of the last century was, however, made outside Europe, in the expanding British Empire, notably in Australia and India, and led to such important works as W. J. Hooker’s Flora Boreali- Americana (1829-40), Grisebach’s Flora of the British West Indian Islands (1850-64), Bentham’s Flora Australiensis (1863-78), J. D. Hooker’s Flora of British India (1872-97), etc. Never¬ theless, material gathered on short trips and holidays by a diversity of persons and now in the British Museum, Cambridge, Edinburgh, Kew, Manchester and Oxford herbaria, particularly from Spain, Italy and the Balkan peninsula, has provided a valuable basis for much work on the Flora Europaea. Such specimens have included new species. Centaurea ebenoides , for example, was discovered- on Euboea by the English philosopher and economist John Stuart Mill when touring Greece in 1862. III. A third contribution is the preparation of Floras or regional lists. Here again Ray made a notable contribution by including so many European records in his Historia Plantarum derived from his own travels as well as from the literature. The major British activity in this has been, as indicated above, outside Europe and has borne fruit not only in the major Floras mentioned but also in smaller works such as Bentham’s Flora Hongkongensis (1861), Gamble & Fischer’s Flora of Madras (1915-36), Dalzell & Gibson’s Bombay Flora (1861), J. L. Stewart’s Punjab Plants (1809), Haines’s Botany of Bihar and Orissa ( 1921 — 4), Theodore Cooke’s Flora of the Presidency of Bombay (1901-3), Duthie’s Flora of the Upper Gangetic Plain (1903-29), Prain’s Bengal W. T. STEARN 15 Plants (1903), Trimen’s Handbook of the Flora of Ceylon (1893-1900), etc., some of these covering areas as big as many European countries. The British contribution to European regional Floras has been essen¬ tially the work of well-informed and enthusiastic amateurs , among the first being Bentham’s Catalogue des Plantes indigenes des Pyrenees et du Bas-Languedoc (1826) followed by P. B. Webb’s Iter Hispaniense (1838), J. H. Moggridge’s Flora of Mentone (1867), Clarence Bicknell’s Flora of Bordighera (1896) and A. H. Wolley-Dod’s Flora of Gibraltar and the Neighbourhood ( J . Bot., Lond. 52: Suppl. (1914)). The most notable British contribution to a European Flora is, however, Sibthorp and Smith’s Flora Graeca (1806-40). John Sibthorp travelled in Italy, Greece, Asia Minor and Cyprus from 1784 to 1787 accompanied by the young botanical artist Ferdinand Bauer, who made superb drawings of the plants seen to illustrate a Flora of Magna Graecia. Sibthorp died in 1796, leaving illustrations, muddled notes and income from the Sibthorp estate for the publication of a Flora Graeca in ten volumes. Thanks to the devoted care of his executor Thomas Hawkins, the engraving of the plates by James Sowerby and the preparation of the text for Vols. 1-7 part 1 by James E. Smith and for Vol. 7 parts 2-10 by John Lindley, publication of this magnificent and costly work was achieved in 1840, the original edition reaching only 25 complete sets, each of which cost about £620 to produce, though sold at £254. Between 1845 and 1856 Henry Bohn published a re-issue of about 40 copies. The total cost of the undertaking was £15,572 (cf. Stearn, 1967). IV. A fourth kind of contribution is the preparation of monographs embracing all the European species. In general, British botanical effort has been directed to the preparation of Floras of British overseas possessions rather than of monographs covering other areas. Here again George Bentham stands out by his work Labiatarum Genera et Species (1832-6), but the most remarkable are George Maw’s Monograph of the Genus Crocus (1886), and W. R. Dyke’s The Genus Iris (1912), based on years of study of living plants assembled in their gardens. Neither Crocus nor Iris can be studied satisfactorily from dried material. They are, however, groups of considerable horticultural value with a high aesthetic appeal, and as such have attracted the painstaking attention of an especially British type, the gardener-botanist, whose particular characteristic is to take up a group of garden interest, collect and cultivate together all the available kinds, study their characteristics in a living state and, after years of intimate acquaintance with them, produce a handbook, survey or revision. J. G. Baker, originally a draper, then a botanist in the Kew Herbarium, did useful work in revising the Liliaceae in a long series of papers in the J. Linn. Soc., 16 THE BRITISH CONTRIBUTION Bot. 11-18 (1870-80). More recently (1941-50) W. W. Smith and H. R. Fletcher have published a series of papers in the Trans. Proc. bot. Soc. Edinb. 33-5 (1941-50), J. Linn. Soc., Bot. 52 (1942) and Trans. R. Soc. Edin. 50-1 (1942-3) revising section by section the genus Primula. V. From these regional floras and world monographs it is one step forward to the major synthesis, the Flora of all Europe, Flora Euro- paea. Grisebach attempted such a work but achieved only a fragment published in 1882. Werner Rothmaler put forward plans for a Flora Europaea in 1944, but it is undeniable that the major impetus and most of the money for this great international work have come from the British Isles. Throughout its preparation taxonomists in every part of Europe have contributed advice, constructive criticism and accounts of genera. Contributors to Vols. 1-3 number 125. Of these Britain has provided 60, Germany and Czechoslovakia each 7, Portugal and Swit¬ zerland 5, Poland and Spain 4, France, Austria and Hungary 3 , fourteen other countries 1 or 2. The work is thus part of the British contribution to European floristic botany, but it is also a European contribution to British floristic botany. Such fruitful interaction has its roots in the sixteenth century. The international membership of the present con¬ ference on European floristic botany in Cambridge is evidence of its happy continuance and growth. Some sources of further information Burkill, I. H. (1953). Chapters on the history of botany in India. J. Bombay nat. Hist. Soc. 51: 846-67 (reprinted in 1966 in book form). Clarke, W. A. (1900). First Records of British Flowering Plants, 2nd ed. London. Clokie, Hermia N. (1964). An Account of the Herbaria in the Department of Botany in the University of Oxford. London: Oxford University Press. Dandy, J. E. (ed.) (1958). The Sloane Herbarium. London: Trustees of the British Museum. Emery, F. V. (1971). Edward Lhuyd F.R.S. 1660-1709. Cardiff: Gwasg Prifysgol Cymru. Field, H. & Semple, R. H. (1878). Memoirs of the Botanic Garden at Chelsea. London. Gilmour, J. S. L. (1944). British Botanists. London. Gilmour, J. S. L. (ed.) (1972). Thomas Johnson, Botanical Journeys in Kent and Hampstead. Pittsburgh: Hunt Botanical Library. Gunther, R. W. T. (1945). Life and Letters of Edward Lhwyd ( Early Science in Oxford, 14). Oxford. Kew, H. Wallis & Powell, H. E. (1932). Thomas Johnson, Botanist and Royalist. London. Moir, Esther (1964). The Discovery of England: the English Tourists, 1540-1840. London: Routledge & Kegan Paul. W. T. STEARN 17 Raven, C. E. (1942). John Ray , Naturalist, his Life and Works. Cambridge. Raven, C. E. (1947). English Naturalists from Neckam to Ray. Cambridge. Rothmaler, W. (1944). Aufforderung zur Mitarbeit an einer Flora von Europa. Reprium nov. Spec. Regni veg. 53: 254-70. Stearn, W. T. (1958). Botanical exploration to the time of Linnaeus. Proc. Linn. Soc. Lond. 169: 173-96 (extensive bibliography). Stearn, W. T. (1967). Sibthorp, Smith, the Flora Graeca and the Florae Graecae Prodromus. Taxon 16: 168-78. Stearn, W. T. (1973). Ray, Dillenius, Linnaeus and the Synopsis methodica Stirpium Britannicarum (prefixed to Ray Society facsimile of John Ray, Synopsis methodica, 3rd ed.). London: Ray Society. Turner, W. (1965). William Turner, Libellus de Re Herbaria, 1538. The Names of Herbes, 1548. Facsimiles with introductory matter by J. Britten, B. D. Jackson and W. T. Stearn. London: Ray Society. Welch, Mary A. (1972). Francis Willoughby, F.R.S. (1635-1672). J. Soc. Bibl. Nat. Hist. 6: 71-85. 18 FLORISTIC STUDIES IN GREECE W. Greuter ABSTRACT A historical survey of the floristic exploration of Greece, followed by some suggestions for an effective and harmonious international colla¬ boration. The present state and future prospects of Greek floristics are illustrated by four maps. Dr Stearn has spoken to us about the contribution of a single nation, Great Britain, to the botany of Europe. The lecture I have been asked to give deals with a fairly complementary subject: the exploration of the flora of a single country, Greece, by students from all over Europe. Let me state from the very beginning that this survey of the floristic exploration of Greece can by no means aim at completeness. I had to attempt to operate a selection and, at the same time, some sort of classification of the countless contributors to Greek floristics. A strictly chronological sequence did not seem very satisfactory to me. I have adopted thematical headings which coincide to a fair extent, however, with particular though partly overlapping chronological periods. 1 . THE CLASSICS One can find several good reasons for the choice of Greece, in preference to other countries, as a subject for this lecture - beginning with the trivial one that it is the region with which I am most familiar, and ending with the statement that no other part of Europe exemplifies to such a striking degree the past and present interwovenness of floristic research all over Europe, irrespective of political boundaries. But the most valid of arguments, at least on an emotional level, is that Greece is rightly considered the birthplace and cradle of scientific botany. Theophrastos (372-287 b.c.), born at Eressos on the East Aegean island of Lesbos (Mitilini), was Aristotle’s most famous student and his successor as a head of the Peripatetic school at Athens. He was a philosopher in the most comprehensive sense of the word, his interests ranging from metaphysics and politics to the natural sciences. [ 18] W. GREUTER 19 It happens that one of his few works to have survived is the Histories on plants (mpi cpuTcov ioropioa). This incredibly fascinating, genial text brings scientific botany into existence and raises it, at the same time, to a level of perfection which was not to be reached again for about two thousand years. The roots (and more than the sheer roots) of most of the modern disciplines of botany can be found in the Histories on plants , namely systematics, morphology, anatomy, phyto¬ geography and ecology. Still better known than Theophrastos’s masterly treatise, and with a much more immediate impact on the further development of botany, is a later work of a very different kind: the Materia medica (nepi uAris icrrpiKris) of Pedanios Dioskorides. This Greek from Asia Minor (born in Anazarbos in Cilicia), an army doctor under Nero, lived in the first century a.d. The scope and intent of his work was an eminently practical one: it is a sort of handbook, giving a review of the very impressive professional knowledge of the widespread and highly es¬ teemed school of Greek herbalists and physicians. Much of its contents, obviously, is existing tradition forged into a new frame; but some parts - as, for example, the synonymies in many contemporary languages - appear to be original and due to Dioskorides’ s wide range of knowledge and experience. Theophrastos and Dioskorides are names which stand for many others: botanists and herbalists whose works are more obscure, partly or entirely lost. Doubtless botany - together with all other fields of science and philosophy - reached a first and very marked climax in ancient Greece. Through the Byzantine Empire, the medieval monastic scholarship and some Arabic and other sources, a few sparks and sprinkles of this venerable and admirable tradition have been saved for us; as to the remainder, we are left to imagine what it may once have been. 2. THE HUMANISTS The return to the classical sources which characterized the Renaissance period favoured the onset of the botanical exploration of Greece. The old Greek and Latin plant names of Dioskorides and Pliny had been currently applied to a variety of Central and North European species by the medieval herbalists. It was now felt that these interpretations needed a severe reappraisal: the original sources were again to be used, and checked against the plants growing in the classical countries, first of all Greece. At the beginning of this new trend stands a very rare book on the simples and their interpretation, written by Luigi Anguillara (1561), who 20 FLORISTIC STUDIES IN GREECE had himself visited Italy and the Balkans, down to the island of Crete, on the tracks of the ancients. He was followed by many others, some of them prominent naturalists like Tournefort and Sieber (see section 4). John Sibthorp is a prominent example who, before setting out on his Greek expedition, thoroughly studied a world-famous, beautifully illustrated copy of Dioskorides: the ‘ Codex Aniciae Julianae ’, a Byzan¬ tine manuscript dating back to a.d. 512, kept at the Imperial Library at Vienna. Sibthorp’ s contribution to Greek botany is a very prominent one which Dr Stearn has outlined in his paper on the British contribution to European floristics (see also Stearn, 1967). Karl Fraas (1845) published a book which represents both a masterly synthesis of and a conclusion to the humanistic investigations of the Greek flora. His Synopsis plantarum florae classicae is a compendium of the plant names of the classics with their modern equivalents, resulting mainly from the author’s own, extensive Greek field studies. Fraas was, between 1838 and 1843, the first professor of botany of Athens University, then newly founded by King Otho I. His departure coincided with the dismissal of all foreign (specifically Bavarian) state ministers and higher officials, which was enforced by a Greek rebellion. This is just one of many examples of the strong interrelation between politics and floristics, and leads us to say some words on Greek political history. 3. THE HISTORICAL BACKGROUND This is in fact an extremely complex and manifold subject, and only a very generalized, rudimentary picture can be given here. Following the fall of Constantinople and the dismemberment of the Byzantine Empire (1453), Greece was taken over by the Turks and by Venice. The former gradually widened their domain, first to the whole continental area whose conquest was virtually achieved in 1460 (except for a few harbours), then to the islands: Rhodes fell in 1522, Crete in 1669 and Tinos as late as 1715. Only the Ionian Islands remained Venetian up to the French Revolution, and were afterwards transformed into a British protectorate. Greek national independence had thus to be won at the expense of the Turks. This was achieved in the Greek liberation war (1821-9), succeeded by the establishment of a Greek kingdom under the Bavarian prince Otho in 1832. This kingdom originally comprised Central Greece, the Peloponnese, Euboea and the Cyclades, and was later gradually extended to its present shape and size: in 1864 the Ionian Islands were handed over by Britain, in 1881 Thessaly and the Arta province became Greek; the Balkan Wars led to the annexation, in 1913, of Epirus, W. GREUTER 21 Macedonia, Crete and the East Aegean islands; in 1923, after the conclusion of the rather catastrophic Greco-Turkish war, western Thrace was added; finally, as a result of World War II, the Dodecanese was taken over from the Italians, who had occupied it since 1912. When considering the historical development of Greek floristics, one must acknowledge the fact that the Greek (and, earlier, the Venetian) parts of the territory were much more easily accessible than the Turkish domain. Especially the inland areas under Turkish administra¬ tion were far from safe: political instability, corruption and inefficiency of local administrations, tribal wars and brigandage made travelling in those countries troublesome and dangerous. The means of inland communication were also rudimentary, while sailing was relatively safe. Thus, most of the early botanical explorers of the Ottoman provinces travelled by sea and concentrated on the islands and some coastal regions of the mainland; indeed, the only Greek mountain areas which had been botanically explored to a satisfactory extent, prior to the 1830s, were those of Crete. 4. THE NATURALISTS The early explorers of the Greek flora were no thoroughbred botanists: they displayed a wide range of interests in natural history, geography, ethnography, archaeology and the humanities. The country was so unexplored in every respect that one could hardly afford to concentrate on a single, specialized subject at that time: everything was new and noteworthy and had to be investigated, gathered and published. Travel reports from the sixteenth to the nineteenth century are fascinating reading, although the information they convey lacks many details, and sometimes even accuracy. The first prominent example of a naturalist to explore Greece was the French Pierre Belon du Mans. During his travels to the Levant, in 1546 and 1547, he visited Crete, the North Aegean islands Lemnos and Thasos, Mt Athos, the coastal areas of eastern Macedonia and Thrace and the islands of Lesbos, Hios, Samos, Patmos, Kos and Rhodes in the eastern Aegean. His observations were published in the form of a highly informative and original book (Belon, 1553), whose French text was subsequently translated into Latin by Charles de l’Ecluse. Another Frenchman, Joseph Pitton de Tournefort, explored Crete and a considerable number of the smaller Aegean islands in 1700, together with the painter Aubriet and a German doctor, Gundelsheimer. This expedition is particularly important to us since Tournefort’s primary interest was in plants: he was Professor of Botany at the Paris Museum, and author of a skilful and highly esteemed plant system. His 22 FLORISTIC STUDIES IN GREECE thorough knowledge of the vegetable kingdom and of its classification enabled him to digest rationally the huge quantity of material he found and collected, and to allot their proper places to the many new species he discovered. The general account of his travels was published posthumously, in the form of letters, together with some beautiful copper engravings of the most notable plant species (Tournefort, 1717). A further most readable travelling report is that of Franz Wilhelm Sieber (1823) who had explored Crete in 1817. Captain (and later Rear-Admiral) Dumont d’Urville’s Aegean cruise in 1819 yielded only a purely botanical account (Urville, 1821). The next notable enterprise was due to the French habit (initiated by Napoleon Bonaparte on his Egyptian campaign) of connecting scientific and military expeditions. France had despatched an army corps to the Peloponnese (the Moree, as they called it) during the later stages of the Greek War of Independence. This was paralleled in 1829 by the ‘expedition scientifique de Moree’ under Colonel Bory de Saint- Vincent, who later edited the results in an impressive multi-volume series (Botanique by Fauche et al. 1832-6). The newly created Greek kingdom with its safer travelling oppor¬ tunities attracted many students, especially Germans. Among the authors of corresponding narratives, some had a marked interest in botany: Emanuel Ritter von Friedrichsthal (1838), the ‘ Bergkommissar ’ Karl Gustav Fiedler (1840-1) and the well-known Viennese paleo- botanist Friedrich Unger (1862) may be cited in this context. August Grisebach deserves a special mention: he was the first to defy the dangers of the mountainous inland regions of the still Turkish portions of the Balkans. He crossed northern Greece in 1839, coming from Constantinople: from the Evros estuary he went to Maronia, crossed to the island of Thasos and to Mt Athos, continued through the Halkidike peninsula to Thessalonica, Edessa and Arnissa and ascended Mt Kajmakcalan or Vorras (Nidze planina); he afterwards visited the now Yugoslavian part of Macedonia, climbed Mt Pelister and the Sar planina and, having cut across northern Albania, finally reached the Austrian territory of Montenegro. The whole journey is very nicely described in his two-volume report (Grisebach, 1841). A French geographer and geologist, Victor Raulin, scanned Crete very thoroughly in 1845, collecting plants which were partly determined by Boissier. His masterly treatise on the island (Raulin, 1869) includes a sizeable chapter on botany, based on both his own and Heldreich’s results (see the next section). Last but not least among the naturalists, Forsyth Major - mainly known as a vertebrate paleontologist - collected plants on several Aegean islands between 1886 and 1890, sponsored by Boissier’s son- W. GREUTER 23 in-law William Barbey at Geneva. From these expeditions two mono¬ graphic works on the natural history of Samos and Karpathos resulted (de Stefani et al. 1892, 1895). 5. THE FIELD BOTANISTS From approximately 1840 onward, the naturalists were gradually re¬ placed by explorers with purely botanical interests. Specialization had become a real necessity if further progress was to be achieved. It made it possible to investigate the subject in a much more thorough manner, to collect more material and, above all, to convey more accurate and much more detailed information about this material thanks to an improved labelling procedure. There are, of course, exceptions, but in general one can say that the naturalists’ specimens are valuable only as historical documents (and often as nomenclatural types), while those collected by the botanical explorers can serve as a reliable base for floristic records and distribution plots. Edmond Boissier, the well-known author of the Flora orientalis (1867-88), was the father and initiator of Greek field botany. Having visited Greece himself in 1842, he convinced a virtually unknown young German botanist to settle in Athens and to start exploring the different regions of the country. This talented youngster was to stay in Greece for almost 60 years, from 1843 to his death in 1902, and to become the most famous and knowledgeable botanist who ever resided in this country. His name was Theodor von Heldreich. Heldreich’s contribution to the development of Greek botany can hardly be overrated. Apart from his official duties (he was director of the Athens Botanical Garden from 1851 to 1902, and keeper of the Natural History Museum from 1858 to 1883), he made innumerable collecting expeditions and excursions all over Greece, of which only the most important can be mentioned: to the Peloponnese in 1844, to Crete in 1846 and again in 1870, to the Thessaloniki area and Mt Olympus (then still Turkish) in 185 1 , across the Central Greek mountain areas in 1879 and to the freshly annexed territory of Thessaly in 1882, 1883 and, together with Haussknecht, in 1885. Moreover, Heldreich was invaluable as an adviser, mentor and sometimes companion to other field botanists and professional collectors. His activity resulted in the discovery of a great number of new species (mostly described together with Boissier), of innumerable new plant localities, and in classical, widely distributed sets of exsiccata of which the Herbarium graecum normale is the best known (see Halacsy, 1902). Eduard Formanek, from Brtinn (now Brno) in Bohemia, was one of those fearless explorers who defied the Turks, the fierce natives 2 UBG 24 FLORISTIC STUDIES IN GREECE and the brigands of the Balkan interior. On several journeys between 1889 and 1899 he visited, apart from many areas of present-day Yugoslavia, the now Greek western Macedonia, Thessaly and Epirus. His botanical skill and care were not, alas, equal to his courage and perseverance: his specimens were poor, and his determinations even worse, so that his various floristic papers are best ignored. It is fortunate that Vandas (1909) revised critically the whole Formanek collection, so through his book the many interesting new records are now safely available. As unwearied and intrepid a traveller as Formanek, and a much better botanist and collector, was Antonio Baldacci of Bologna. Best known as explorer of Albania, he made significant contributions to Greek botany too, visiting the Epirus in 1892, 1895 and 1896, and Crete in 1893 and 1899. Even in the latter already rather well-known area he made several interesting finds owing to his predilection for unusual itineraries and remote spots (see, e.g., Baldacci, 1895, 1903, 1917). Eugen von Halacsy, an Austro-Hungarian physician and botanist, is most famous as the author of a flora of Greece (Halacsy, 1900-8, 1912) which is still, in my judgement, the best of its kind, in spite of its age and of the fact that its boundaries exclude most of the more recent Greek territories like Macedonia, Thrace and the eastern islands. Halacsy acquired his own Greek field experience on three journeys in 1888, 1893 and 1911, of which the second was the most important. It led through Epirus, Thessaly, Aetolo-Acarnania and the northern Peloponnese and yielded a series of very thorough floristic accounts (Halacsy, 1894). An equally important series of papers resulted from the Greek expeditions of Rene Maire in 1904, 1906, 1908 and 1917 (the second one together with M. Petitmengin). They are important for both phanero- gamists and cryptogamists, especially mycologists, to whom Rene Maire’ s name is a very familiar one indeed (see Maire, 1906-9, 1922). Incidentally, the original set of specimens from these trips is not in the Maire herbarium, now deposited at Montpellier (MPU), but at the Faculty of Sciences of Nancy University (NCY). Other familiar names of botanists who visited Greece are Michel Gandoger (in Crete from 1914 to 1917, possibly with some kind of pseudo-diplomatic or secret service activity; see Gandoger, 1916, 1919, 1920); Johann Bornmuller in 1926 (Bornmiiller, 1928; he had already visited Mt Athos in 1891, together with Sintenis); and Peter Davis, who travelled extensively on the islands and the continent in the early stages of World War II, prior to the German occupation, and visited Crete and Karpathos again in 1950 (see Davis, 1953, 1955). Several Bulgarians (e.g. Urumov, Stojanov, Stefanov, Kitanov; see W. GREUTER 25 Kuzmanov, 1971, for further details) made basic contributions to the knowledge of the flora of northern Greece, although their publications have been somewhat neglected because they are written in Bulgarian, and many of them are not readily available. One may remember that parts of Macedonia and Thrace were Bulgarian from 1912 to 1918, and again occupied by Bulgarian forces during World War II. In recent times Greek field botany has been very much intensified. Improved travelling facilities and stable political conditions are partly responsible for this development. But we must not forget the persons who are behind this general trend, and first of all Karl Heinz Rechinger who, having first visited Greece in 1927, has been backsliding fairly regularly to this sin of his youth - at least twenty times up to the present day. His main attention was devoted to the islands and peninsulas of the Aegean sea (Rechinger, 1943, 1949, 1961), and notably to Crete, which he visited with a German wartime scientific expedition in 1942 (Rechinger, 1944; a parallel expedition led Rothmaler to the Peloponnese, but the collected materials were destroyed at Berlin, following an Allied air raid, before being worked up). Several excur¬ sions to the mainland, from Peloponnese to Epirus and Thrace, must also be mentioned, but only two of them yielded floristic accounts (Rechinger, 1936, 1939). Rechinger’ s are certainly by far the most extensive Greek collections ever brought together by a single person. They are outnumbered, however, by those made by a collective enterprise which one may call the Lund undertaking, or Runemark’s team: a group of young Swedish botanists who have been scanning one by one all the countless islands of the Aegean, from 1957 onward (Runemark et al. 1960-74), and are now doing the same with the mountain peaks of the mainland. Once all the results of this collecting are available in a published form, we will be justified in stating that the preliminary chapter of the floristic investigation of Greece has been successfully closed. 6. THE PROFESSIONAL COLLECTORS The contribution of plant dealers and professional collectors to Greek floristics is far from negligible. They have certainly set new standards in the techniques of collecting and drying plant specimens. Unfortun¬ ately their labels are not always prepared as carefully as the correspond¬ ing exsiccata. In a few cases (Reverchon!), they are indeed so carelessly prepared that one could easily speak of fakes. . . We find forerunners of the ‘professionals’ in the first half of the nineteenth century, with Sieber in Crete in 1817 (see section 4), Frivaldszky’s hired collectors in Crete and northern Greece, and 26 FLORISTIC STUDIES IN GREECE Aucher-Eloy in Attica, on Mt Athos and on some islands. But the high periods of professionalism were the second half of the nineteenth century - with names such as Balansa, Bourgeau, Pichler, Reverchon and Sintenis - and the early years of our century: Adamovic, Dimonie, Dorfler and Leonis. Apparently the First World War settled the fate of this profession: plant dealing is no longer, it seems, financially rewarding. 7. THE GROUP EXCURSIONS Collective excursions are an invention of our century. Today, commer¬ cial tourist cruises with partly naturalistic aims are current and very popular, especially in Britain, but originally such group excursions were the prerogative of universities. The first to make a group attack on Greek botany were the Viennese, who organized two ‘Universitatsreisen ’ in 1911 and 1914; Zurich followed in 1921, under the leadership of Martin Rikli and Eduard Riibel. Nowadays, almost every year botanical university excursions swarm over Greece. They are not usually of a basic importance to floristic research: they tend to follow trodden paths, mostly around touristic centres with hostels and opportunities for bathing, shopping and archaeological sight-seeing. To some extent, this seems to have been true even for the pioneer enterprises mentioned above, if we believe what Samuelsson (1933) wrote on the subject of the Zurich excursion in which he had participated (translation mine): ‘Unfortun¬ ately I had in a way misunderstood the purpose of this trip - it turned out to be to a large extent of a purely touristic nature, with a pronounced tendency to swallowing kilometres - and had also over¬ rated the knowledge of the leader (Rikli) with respect to the Medi¬ terranean flora. . As a rule, the scientific results of group excursions are not published. In some cases, duplicated plant lists are produced, often with rather provisional determinations. The Vienna and Zurich travels, however, yielded extensive scientific reports which belong to the traditional Greek floristic literature (Vierhapper, 1914-16, 1914-19; Rikli & Riibel, 1923). A more recent example is the International Phytogeographical Excursion to Greece in 1971, whose results will be published shortly (Dafis & Landolt, 1975). W. GREUTER 27 8. THE AMATEURS The amateurs have actually and chronologically succeeded the natural¬ ists - indeed, in a way, they are naturalists, just denoted by a less ambitious term. The amateurs’ contribution to Greek floristics was a very prominent one, although it is still imperfectly known since most of their results remain unpublished. It is very difficult, if not impossible, to keep a record of botanical amateur activity in Greece, especially during these last few years since tourism has so tremendously in¬ creased. Citing any names, in this context, would mean being unfair to the many equally prominent amateur botanists left unmentioned. (By the way, I am proposing no special category for trained botanists on their holiday trip, without a specific Greek research programme, who may just as well be included together with the amateurs.) There are, however, three persons who must be specifically men¬ tioned here. They owe this special treatment to the fact that they have all been living in Greece for a long time: Shirley Atchley and Fernand Guiol in Athens, from where they have explored many and often remote parts of the country; and H. G. Tedd in Xanthi who has mainly investigated western Macedonia, Thrace and the northern Aegean area. Guiol’s herbarium is now at the British Museum (Natural History), and Atchley ’s and Tedd’s are at Kew. All three are rich in new, mostly unpublished findings which certainly deserve to be put on record. 9. THE PHYTOSOCIOLOGISTS They are the youngest members of the big family of Greek botanists - but certainly not the most unproblematic ones. The good botanical amateur, as we have just seen, can be characterized as a remarkable collector, a keen connoisseur of plants, but a very lazy author. Conversely, one might depict the phytosociologists - even some of the best! - as poor collectors with an inadequate floristic knowledge, but with a tremendous printed output. After this slightly disrespectful statement (for which I apologize to all those colleagues - and they exist indeed - who do not deserve it), I will certainly be forgiven for not dwelling on this subject and for not giving any concrete examples. But I want to make it quite clear that data from phytosociological releves (and in a few cases, alas, even from floristic publications by a phytosociologist) should not be recognized as floristic records unless specimens have been checked. 28 FLORISTIC STUDIES IN GREECE 10. THE GREEK BOTANISTS Having begun our survey with the Greek fathers of botany, we have been lingering on several categories of students from abroad. It is now time, as a conclusion, to revert to the genuine Greek botanists. The first to gain well-deserved reputation and fame in this field was Theodoros Orphanides, Fraas’s successor in the Chair of Botany of Athens University which he held from 1849 to 1882. Orphanides was successful both as a botanical explorer and as a scientist, collaborating with Boissier and Heldreich but also working independently. He distributed a well-known series of exsiccata on an exchange basis, and assembled a considerable world-wide personal herbarium which now constitutes the main body of the Athens University herbarium. After Orphanides’ retirement, floristics and systematic botany prac¬ tically disappeared from the Greek universities, the botanical profes¬ sors concentrating on other fields of research, or on teaching activities (see Phouphas, 1940). Some successful collecting was done in the early years of our century by the successor of Heldreich at the Athens Botanical Garden, B. Tountas. We must proceed to the 1930s to find a renewed, though still timid, rise of our discipline in Greece, mainly centred on Thessalonica University where Konstantinos Ganiatsas, who was to become Professor of Botany and a very successful teacher of plant systematics, began publishing a series of valuable floristic contributions. Today, of the several chairs of botany at the three Greek universities only one, at Patras, is occupied by a phanerogam taxonomist: Dimitrios Phitos, well known for his skilful and thorough contributions to floristics and cytotaxonomy. But there is a striking renewal of interest in the subject at the other botanical institutes too, and several young students and scientists are now enthusiastically proceeding to the exploration of the floristic riches of their home country. 11. THE GREEK AMATEURS Greece too has had, and still has, some prominent amateur botanists. To begin with, we may mention Miss Sophia Topali, whose specimens are now at the Geneva Conservatoire botanique (G). Her main area of activity was Mt Pelion, close to Volos in Thessaly where she was living. She also visited other parts of Greece, partly together with Gustave Beauverd , the keeper of the Boissier Herbarium (see Beauverd & T opali , 1937). She was executed during the German occupation in World War II, being suspected of collaboration with the Greek partisans. Two of the most sizeable Greek herbaria, each of about 25,000 W. GREUTER 29 specimens, have been built up in fairly recent times by local amateurs. Both are of considerable importance to the student of the Greek flora, since they contain unpublished and inadequately studied material often collected in remote, ill-known areas whose access even nowadays remains difficult. The first is due to the skill of Leonidas Pinatzis: it is a beautiful, remarkably well-arranged collection, still kept and cared for by his family. The second has been assembled by Constantine Goulimis, a specialist of international and commercial law who, having decided at the age of sixty to devote himself to the plant life of Greece, ceaselessly explored all the parts of this country from 1946 to his death in 1963. His herbarium contains innumerable new findings and many novelties, a few of which were described and published by Turrill & Rechinger (see also Goulimis, 1956, 1960; Goulandris, Goulimis & Stearn, 1968); it is now at the Goulandris Natural History Museum whose foundation, in fact, was directly motivated by its bequest. 12. THE GOULANDRIS MUSEUM This institution, which was founded in 1964 in Kifisia near Athens as a botanical museum and was later expanded to include other branches of natural history as well, has the status of a private foundation made and endowed by Mr and Mrs Angelos Goulandris. By now it plays a major role in Greek botany. It has its own collecting staff: the most active collaborators are Mrs Elli Stamatiadou, who is a very keen and gifted plant hunter and has been successfully prospecting the whole country since 1967; and Paul Haristos, who lives in Thessalonica and has been concentrating on the flora of Macedonia from 197 1 onward. On the other hand, a scientific library is now being built up, publication activity has started and several research programmes, monographic as well as floristic, are under way. The Goulandris Museum has also prepared a public exhibition in order to increase the knowledge and love of nature of the Greek population, and above all of the young generation who will have to resolve the increasingly urgent problems of the conserva¬ tion and management of nature and the human environment. Besides its own activity as a centre for public education, collecting, documentation and research, the Goulandris Museum sees one of its most important and noble tasks in the establishment and maintenance of contacts between the many students of Greek botany, and in a co-ordination of their efforts. It serves as a strengthening point and working base to many botanists, including the amateurs who have been largely neglected by the university institutes. It favours the exchange of information and aims at establishing a correlation between the independent research programmes of different students and teams. 30 FLORISTIC STUDIES IN GREECE Most urgently, it endeavours to strengthen and improve the relations between Greek and foreign botanists. 13. THE PROBLEMS OF INTERNATIONAL COLLABORATION The Greek botanists, quite naturally, tend to have an inferiority complex when faced with the numerous foreign specialists successfully working in their home country, having a training, an experience and - last but not least - domestic research facilities such as libraries and herbaria which cannot yet be obtained within Greece. This feeling of frustration usually turns into blank fury when it appears that those foreign botanists have been working, and are publishing, in then- personal research fields, without giving them the slightest chance of knowing beforehand and of delimiting appropriately the respective subjects. I believe that we must understand these reactions, and try to prevent them. Incidentally, they are by no means limited to Greece, but widespread in most or all the so-called developing countries. Unfor¬ tunately in many cases - not yet in Greece! - they have led not only to a regrettable general animosity against foreign research but, more concretely, to deplorable and counter-productive administrative restric¬ tions to scientific activity, especially to fieldwork. Faced with similar problems in Central and South America, a group of biologists from the United States has produced a sort of chart of good behaviour for biologists abroad, which is basic reading for all those, amateurs and especially professionals, planning fieldwork in foreign countries: no severe, weary rules, just some self-imposed imperatives and restrictions which, at a first glance, seem so self- commendatory to a decent-minded scientist that it is hard to believe how often one has unconsciously neglected them in the past. These ‘Guidelines for biological field studies’ have been reprinted in Taxon (19: 950-1 (1970)), and I should like to forcibly recommend everyone who has not yet done so to study them carefully, paying special attention to the seven points of the second section dealing with ‘ guest scientists’. The establishment of close contacts with colleagues of the host country, the exchange of information, materials and publications and the co-ordination of research are essential prerequisites for a fruitful international collaboration. W. GREUTER 31 Map 1. The floristic investigation of Greece. A, well or reasonably well explored areas. B, ill-explored areas. C, unexplored areas. Further explanations in the text. (The inset in the left-hand bottom corner of this and the following maps represents the Greek island group of Kastellorizo off the S. Anatolian coast.) 14. THE PRESENT STATE OF THE FLORISTIC EXPLORATION OF GREECE A report on this topic by W. Greuter, D. Phitos and H. Runemark was recently presented at the CNRS Colloquium on the ‘Flore du bassin mediterraneen ’ (Montpellier, 4-8 June 1974). Its conclusions were summarized in the form of four maps which are reproduced here. The complete report will be published in the Proceedings series of the ‘Colloques internationaux du CNRS’; it will include literature 32 FLORISTIC STUDIES IN GREECE Map 2. An estimate of the relation of published to unpublished floristic information in Greece. A, areas from where more than two-thirds of the available information has been published. B, areas from where one- to two-thirds of the available information has been published. C, areas from where less than one-third of the available information has been published. references relating to map 3, and details concerning the research projects of map 4. Map 1 is based on an overall estimate of the total amount of available information, both published and unpublished (such as herbarium speci¬ mens and reliable field notes). Islands and mainland areas where an estimated 80 per cent of the existing wild species, or more, have been collected or noted are considered to be ‘reasonably well explored’. Most of the other regions are ‘ ill-explored ’ , the exceptions being a few small islands and some mountains at the Albanian and Bulgarian W. GREUTER 33 Map 3. Greek areas covered by recent floristic publications (see the text for additional explanations). A, thorough coverage (by a comprehensive study or several minor contributions). B, incomplete coverage. frontiers which, to my knowledge, have never been visited by a botanist or collector. Although the criteria applied are less severe on the mainland (the ‘ areas ’ considered there are, on average, much bigger than the individual islands), it is obvious from the map that the insular parts of Greece were much more thoroughly explored than the continent and especially its lowlands. Map 2 shows to what a considerable extent the available floristic information on Greece is still unpublished, and is designed as a warning not to rely on printed sources only when dealing with the 34 FLORISTIC STUDIES IN GREECE Map 4. Greek areas for which floristic surveys are in progress. distribution of plants in this country. The situation will be considerably improved by the publication of the results of the research projects plotted on map 4. As to map 3, which shows the areas covered by recent floristic publications, one must note that the word ‘recent’ has been defined differently in different regions. Basic regional Floras or floristic reports, giving references to the previous literature, have been used as starting points (except for northern Greece where no such work exists): Greuter (1972) for Kriti, Ciferri (1944) for the Dodecanese, Rechinger (1943) for the other parts of the Aegean area, Hayek (1924-33) for W. GREUTER 35 Macedonia and Thrace and Halacsy (1900-8, 1912) for the remainder of Greece. It is hoped that this set of maps, imperfect as it is, will contribute to a rationalization of future floristic activity, and entail an increased effort to investigate the still imperfectly known or unexplored Greek areas. References Anguillara, L. (1561). Semplici, liquali in piu pareri a diversi nobili huomini scritti apaiono. Vinegia. Baldacci, A. (1895). Risultati botanici del viaggio compiuto in Creta nel 1893. Malpighia 9: 31-70, 251-79, 329-55. Baldacci, A. (1903). Itinerari fitogeografici del mio secondo viaggio in Creta (1899). Memorie R. Accad. Sci. 1st. Bologna ser. 5, 10: 253-74. Baldacci, A. (1917). Itinerari albanesi. Rome. Beauverd, G. & Topali, S. (1937). Excursions botaniques en Grece (Pelion, Eubee et Peloponese). Bull. Soc. bot. Geneve 28: 94-183. Belon du Mans, P. (1553). Les observations de plvsievrs singvlaritez et choses memorables, trouuees en Grece, Asie, Judee, Egypte, Arabie, et autres pays estranges. Paris. Boissier, E. (1867-88). Flora orientalis sive enumeratio plantarum in Oriente a Graecia et Aegypto ad Indiae fines hucusque observatarum. 5 vol. & suppl. Geneva, Basle and Lyons. Bornmiiller, J. (1928). Ergebnis einer botanischen Reise nach Griechenland im Jahre 1926 (Zante, Cephalonia, Achaia, Phokis, Aetolien). Reprium nov. Spec. Regni veg. 25: 161-203, 270-350. Ciferri, R. (1944). Flora e vegetazione delle isole italiane dell’Egeo. Atti 1st. bot. Univ. Lab. crittogam. Pavia, ser. 5, suppl. A. Dafis, S. & Landolt, E. (1975), editors. Zur Flora und Vegetation von Griechenland (Ergebnis se der 15. Internationalen Pflanzengeographischen Exkursion durch Griechenland). Veroff. geobot. Inst., Zurich in the press. Davis, P. H. (1953). Notes on the summer flora of the Aegean. Notes R. bot. Gdn Edinb. 21: 101-42. Davis, P. H. (1955). Mediterranean itineraries. Notes R. bot. Gdn Edinb. 21: 271-8. Fauche, M., Brongniart, A., Chaubard, L. & Bory de Saint- Vincent, J. B. (1832-6). Botanique. In J. B. Bory de Saint-Vincent, editor: Expedition scientifique de Moree, vol. 3(2) (text and atlas). Paris. Fiedler, G. K. (1840-1). Reise durch alle Theile des Konigreichs Griechenland. 2 vols. Leipzig. Friedrichsthal, E. Ritter von (1838). Reise in den siidlichen Theilen von Neu- Griechenland. Leipzig. Gandoger, M. (1916). Flora cretica. Paris. Gandoger, M. (1919). Troisieme voyage botanique dans Tile de Crete. Bull. Soc. bot. Fr. 63 (‘1916’): 219-43. Gandoger, M. (1920). Quatrieme voyage botanique en Crete (1917). Bull. Soc. bot. Fr. 64 (‘1917’): 110-22. Goulandris, N. A., Goulimis C. N. & Stearn, W. T. (1968). Wild flowers of Greece. Kifisia. 36 FLORISTIC STUDIES IN GREECE Goulimis, C. N. (1956). Nea eide tes ellenikes hloridos. New additions to the Greek flora. Athens. Goulimis, C. N. (1960). New additions to the Greek flora. Second series. Athens. Greuter, W. (1975). Floristic report on the Cretan area. Flora europaea: Floristic & taxonomic reports. VII. Symposium Coimbra Portugal May 22-31, 1972. (Mem. soc. Brot. \915 in the press). Grisebach, A. (1841). Reise durch Rumelien und nach Brussa im Jahre 1839. 2 vols. Gottingen. Halacsy, E. von (1894). Botanische Ergebnisse einer im Auftrag der hohen kaiserl. Akademie der Naturwissenschaften unternommenen Forschungs- reise inGriechenland. I-IV. Denkschr. Kaiserl. Akad. Wiss. math.-naturwiss. Kl. 61: 217-68, 309-22, 467-86, 487-535. Halacsy, E. von (1900-8). Conspectus florae graecae. 3 vols. and suppl. Leipzig. Halacsy, E. von (1902). Theodor von Heldreich. Ein Nachruf. Magy. bot. Lap. 1: 325-36. Halacsy, E. von (1912). Supplementum secundum Conspectus florae graecae. Magy. bot. Lap. 11: 114-202. Hayek, A. von (1924-33). Prodromus florae peninsulae balcanicae. Beih. Repert. Spec. nov. Regni veg. 30 (1-3). Kuzmanov, B. A. (1971). A catalogue of the taxa (Pteridophyta-Angiospermae) described by Bulgarian botanists from neighbouring countries (1888-1959). Fragm. flor. geobot. 17: 177-210. Maire, R. (1906-9). Materiaux pour servir a V etude de la flore et de la geographie botanique de V Orient. Premier (-sixieme) fascicule. Nancy (also in Bull. Soc. sci. Nancy, passim). Maire, R. (1922). Contribution a l’etude de la flore grecque. Bull. Soc. bot. Fr. 68: 370-80. Phouphas, C. (1940). La botanique en Grece. Act. Inst. bot. univ. Athen. 1: 15-25. Rechinger, K. H. (1936). Ergebnisse einer botanischen Sommerreise nach dem agaischen Archipel und Ostgriechenland. Beih. bot. Zbl. 54 B: 577-688. Rechinger, K. H. (1939). Zur Flora von Ostmazedonien und Westthrazien. Bot. Jb. 69: 419-552. Rechinger, K. H. (1943). Flora aegaea. Flora der Inseln und Halbinseln des agaischen Meeres. Akad. Wiss. Wien math.-naturwiss. Kl. Denkschr. 105(1). Rechinger, K. H. (1944). Neue Beitrage zur Flora von Kreta. Akad. Wiss. Wien math.-naturwiss. Kl. Denkschr. 105(2). Rechinger, K. H. (1949). Florae aegaeae supplementum. Phyton (Horn) 1: 194-228. Rechinger, K. H. (1961). Die Flora von Euboa. Bot. Jb. 80: 294-465. Rikli, M. & Riibel, E. (1923). Uber Flora und Vegetation von Kreta und Griechenland. Vjschr. naturf. Ges. Zurich 68: 103-227. Runemark, H. et al. (1960-74). Studies in the Aegean flora, I-XXI. I, Bot. Not. 113: 421-50; II, id. 451-7; III, id. 114: 453-6; IV, id. 115: 357-75; V, id. 116: 323-5; VI, id. 118: 104-22; VII, id. 139-65; VIII, Opera Bot. 13; IX, id. 14; X, Bot. Not. 120: 9-16; XI, id. 84-94;XII, id. 161-76, 486;XIII, id. 121:233-58; XIV, id. 122: 38-56; XV, id. 123: 52-60; XVI, Opera Bot. 28; XVII, Bot. Not. 123: 371-83; XVIII, id. 384-93; XIX, id. 124: 399-418; XX, Opera Bot. 33; XXI, id. 34. Samuelsson, G. (1933). Symbolae ad floram graecam. Ark. Bot. 26A(5). W. GREUTER 37 Stearn, W. T. (1967). Sibthorp, Smith, the ‘Flora graeca’ and the ‘Florae graecae prodromus’. Taxon 16: 168-78. Stefani, C. de. Major, C. J. F. & Barbey, W. (1892). Samos. Etude geologique, paleontologique et botanique. Lausanne. Stefani, C. de. Major, C. J. F. & Barbey, W. (1895). Karpathos. Etude geologique, paleontologique et botanique. Lausanne. Tournefort, J. P. de (1717). Relation d'un voyage du Levant, fait par ordre du roi. . . 2 vols. Paris. Unger, F. (1862). Wissenschaftliche Ergebnisse einerReise in Griechenland und in den jonischen Inseln. Vienna. Vandas, C. (1909). Reliquiae Formanekianae. Enumeratio critica plantarum vascularium, quas itineribus in Haemo peninsula et Asia Minore ( Bithynia ) factis collegit Dr. Ed. Formanek, professor gymnasii brunensis bohemici. Brno. Vierhapper, F. (1914-19). Beitrage zur Kenntnis de Flora Griechenlands. Bearbeitung der anlasslich der zweiten Wiener Universitatsreise im April 1911 in Griechenland gesammelten Pflanzen. A. Anthophyta und Pterido- phyta. Verb zool.-bot. Wien 64: 239-70; 69: 102-312. Vierhapper, F. (1914-16). Beitrage zur Kenntnis der Flora Kretas. Aufzahlung der anlasslich der fiinften Wiener Universitatsreise im April 1914 auf Kreta gesammelten Blliten- und Farnpflanzen. Osterr. bot. Z. 64: 465-82; 65: 21-8, 50-75, 119-40, 204-36, 252-65; 66: 150-80. APOMIXIS IN A STERILE HYBRID SPECIES OF POTENTILLA R. Czapik ABSTRACT Hexaploid Potentilla mixta Nolte (= P. reptans f. mixta Krause = P. anglica Laich.xP. reptans L.) from England and Poland is highly sterile. It reproduces, however, abundantly by runners which show morphological seasonal variation in special conditions. A few seeds developed by some clones after open pollination were raised to plants of various ploidy level ranging from 5jc to 9x {In = 35, 42, 56, 63) and a single individual with c. 93-7 chromosomes. Cytological differentia¬ tion of the progeny is connected with a tendency to apomixis of the maternal plants. Thus the reproduction of P. mixta represents a complicated system of vegetative propagation, sexuality and apomixis covered by the sterility of the hybrid taxon. In this paper I shall try to present some of the special problems we meet in highly sterile hybrids of a genus rich in apomictic taxa. Polymorphism, apomixis and polyploidy are combined in Potentilla with comparatively often noted interspecific hybridization. The taxon was engaging a good deal of attention in the works of Mlintzing (1928-58) , Rutishauser (1943-67), Hunziker (1954), Smith (1963 a, b ), Smith, Bozman & Walters (1971), Asker (1966-71) and many others who studied cytology, embryology, apomictic phenomena and their regula¬ tion, as well as intraspecific variability. There is one basic number x = 7 in the genus. The euploid numbers range here from 14 to 112. Several numbers are recorded in certain species which comprised euploid and aneuploid plants and in a few cases also plants with fragments of chromosomes. The aneuploids which are rare in nature are noted very often among experimental plants from inter- or intraspecific crosses. The occurrence of aneuploids shows that gametes with aberrant chromosome numbers are able to function in Potentilla , and aneuploid plants may survive in special conditions. In spite of the fact that in published records polyploid apomicts prevailed there is no simple correlation between the degree of poly¬ ploidy and apomixis in Potentilla on account of the complicated and still [38] R. CZAPIK 39 not well recognized genetic regulation of the apomictic processes. On the diploid level apomixis is determined by independent systems of recessive genes, which cause the elementary apomictic processes; apospory, diplospory and parthenogenesis (Rutishauser, 1967; Asker, 1971). The genes have a quantitative effect. The systems of apomictic genes were resistant toward the mutagenic treatment of X-rays and ethyl methane sulphonate in experiments of Asker (1966 a). However, they may be broken down or restored as a result of hybridization or polyploidization. Most of the apomictic Potentillas belong to the pseudogamous group and require fertilization of the central nucleus for the development of endosperm and a viable seed. Single fertilization which occurs in pseudogamous plants is not the only element of sexuality in apomictic Potentilla. They are as a rule facultative apomicts. Both diplosporous as well as aposporous plants may develop unreduced and reduced embryo sacs in various percentages, and their egg cells may be fertilized. In the progenies of these facultative apomicts triploid and diploid hybrids occur among matroclinous plants. Thus polymorphism in Potentilla is in the first place connected with this special mode of reproduction apart from hybridization and some cytological mech¬ anisms which seem to act occasionally in the course of developmental processes (Asker, 1966 b, 1971). According to the last review of apomixis done by Khokhlov in 1967 apomixis was recorded in 30 species of Potentilla , and since then at least three further examples may be added to this list. Among those taxa 19 species belong to the European representatives of the genus. This rather high number of noted apomictic species is in some way connected with the attractiveness of the genus for students, but it is still low in comparison with the number of unexamined taxa. To classify a plant as an apomict many data are needed. The embryological examination reveals the origin of young embryo sacs and degree of apomeiosis as well as early development of autonomically parthenogenetic embryos. The frequency of functional embryo sacs with reduced and unreduced chromosome number in facultative apo¬ micts may be established only after cytological and morphological analysis of progenies, obtained after several controlled pollinations with plants which have various chromosome numbers and belong to various taxa. The examination of progeny from open pollination or from one cross only is insufficient. According to the experiments of Rutishauser the degree of pseudogamy and the number of fertilized unreduced egg cells depends on the pollen plant used in a cross. Such a complex investigation meets additional obstacles in the case of highly sterile hybrids. 40 HYBRID SPECIES OF POTENTILLA Figs. 1-6. Archesporium and initial cells in P. x mixta. 1, young ovule (stage II) at the beginning of meiosis; 2, archesporium and aposporous initial cells; 3, tetrad; 4, meiosis arrested in early prophase; 5, diplosporous initial cells; 6, degenerated tetrad. The sterile natural and experimental hybrids in Potentilla were not embryologically examined. There were, of course, observations on meiosis in anthers and degree of pollen and seed sterility. However, histological backgrounds of the seed sterility are unknown. There are no remarks about the reproductive tendencies interrupted by the sterility. Let us see some aspects of such an investigation using the example of P. x mixta. R. CZAPIK 41 P.x mixta Nolte ex Reichenb. shows the characteristics of the genus: polyploidy and morphological variability together with a hybrid origin. Typical P.x mixta has 42 chromosomes and is believed to be a complicated hybrid of the group Tormentillae, highly sterile but repro¬ ducing abundantly by runners. In its genetic constitution four genomes of P. reptans and two of P. erecta are involved. So the domination of P. reptans in many characters of P.x mixta is understandable. The hybrid might originate from a successful pollination between P. reptans and P. anglica (the latter is a stabilized hybrid of P. reptans and P. erecta ), or directly from a cross between P. reptans and P. erecta. In the last case a hexaploid P.x mixta would be formed as a consequence of the fusion of a tetraploid gamete of P. reptans with a diploid gamete of P. erecta. The experimental confirmation of one of these ways was done by Matfield, Jones & Ellis (1970). The authors obtained two plants morphologically similar to natural P. x mixta from the cross of octoploid (autoploid) P. reptans as the maternal plant with a tetraploid P. erecta , the pollen plant. They mentioned also a syn¬ thesized hexaploid P.x mixta obtained by Valentine from the cross P. reptans as the maternal plant and P. anglica as the pollen plant. It is interesting to mention that Schwendener (1969) obtained two hybrids after pollination of a tetraploid P. reptans with tetraploid P. erecta. The plants had 42 and 44 chromosomes respectively and originated, as the author assumed, from unreduced gametes of P. reptans fertilized by reduced gametes of P. erecta. The ploidy level and genetic constitution especially of the euploid hybrid seemed to be in accordance with the expected constitution of P.x mixta. Nevertheless, both experimental plants show morphological similarities to P. anglica in their habit and branching. We know rather little about the variability of P.x mixta. The examples from Monks Wood showed, for instance, a seasonal variation in the shape of leaves in some degree connected, among other possible factors, with variable watering. Three types of runners differing in shape and size of leaves, presented in the paper of Matfield et al. (1970) for three different plants, were developed by the same individuals from Monks Wood. On the other hand the plants from two other localities in Poland, Rybnik and Muzakow (Czapik, 1968) did not show such striking morphological changes. The high sterility of P.x mixta is the main obstacle in successful pollination experiments. The degree of sterility varies within the population and within the same specimen, but it is always very high. In the experimental field some plants did not develop any seeds after open pollination during several years, but others had 0-3 seeds in one flower. Most of these seeds were, however, empty, only 9-11 per cent 42 HYBRID SPECIES OF POTENTILLA Figs. 7-11. Embryo sacs of P.x mixta. 7, binucleate ES; 8, abnormal four- nucleate ES; 9, two uni- and one binucleate ES, cells arrested in the I prophase; 10, micropylar part of an adult ES, central nucleus short after the fusion of polar nuclei; 11, nuclei from the same ovule: one of a somatic cell and two pairs of nuclei from two binucleate ESs. R. CZAPIK 43 of seeds germinated. The attempts of controlled pollinations failed in about 200 pollinated flowers. Acetocarmine tests showed from 100 to 87 per cent unstained pollen in particular flowers. Extremely poor seed-setting connected with very effective vegetative reproduction classifies P.x mixta as a vegetative apomict. However, agamospermy could also be expected according to the records of Forenbacher (1913) for P. erecta and Schwendener (1969) for P. reptans , P. erecta and P. anglica, for the putative parental species of P.x mixta. These three taxa are able to develop diplosporous and aposporous initial cells, and in P. anglica and P. reptans partheno¬ genesis was noted. So the genes for apomixis occur in the group, and the probability of agamospermy or at least of some elementary apo- mictic processes must be taken into consideration. Some suggestions are expected from the progeny test. The plants were obtained after open pollination in the experimental field, the only technically possible way of getting more seeds from one plant (Table 1). The chromosome number of the pollen plant is of course speculative in each case. One cannot even tell if the plants are self -sterile or self -fertile on account of the difficulties in pollination of highly sterile individuals. The number of plants was low but they were cytologically differentiated; euploids and aneuploids occurred among them. The aneuploid numbers pointed to meiotic disturbances in P. x mixta, in the pollen plant or in both. The plants from Rybnik, Silesia, and from Monks Wood, Hunting¬ donshire as well as from Wessington, Derbyshire (the last examined by Matfield et al. 1970) developed seeds, which germinated and their seedlings survived, after meiosis and fertilization as a rule. However, such chromosome numbers as 42, 56, 61, 63 and 93-7 suggested occasional formation of unreduced gametes. There is also a probability of some mechanisms which might occasionally raise the ploidy level of a plant. Nevertheless facultative apomixis in P.x mixta would be possible, but the supposition demands a confirmation of embryological facts. The sterility of pollen in P. x mixta was connected with disturbances in meiosis: in diakinesis from 2 to 13 univalents occurred and occa¬ sionally one trivalent was formed. One to four chromosomes were left beyond equatorial plates in the I and II metaphases, laggards were visible in the I and II anaphase. As the result of these irregularities additional, small nuclei were formed in interkinesis and in the II telophase. Polyads had one to two microcytes and pollen degenerated after meiosis. The most characteristic feature of the developmental processes in ovules of P.x mixta from the investigated populations were distur- 44 -si s o a c o c §■ ■> si I Cx, -s: s2 -Ch a c o s§ 8 2* -si a . ^ Jh x C3 H § S' C/3 s- X 3 C . a <*> .5 1 c •gC J3 6 ° a > 2 x OfS S .2 * C/3 > - J3 •(-> C« U "* 1/5 £ ’T * ) ' ^ Cd |ss C/3 ~ X O C/3 C/3 X -< X o D -4-) ^ 13 £ D -2 X !l8 £ G • 0) c -a E *- go o c CO O -G £ J3 "a o X C/3 £ O o w c/3 -ra ll O x C/3 s a co C '> o ^ (-1 o a o t-H C/3 o o c/1, proportion for each chromosome of short to long arms. When c/1 is lower than 0*75 there is a heterobrachial iromosome, and when c/1 is equal to or greater than 0-75 there is an isobrachial chromosome. These proportions represent the mean of thirty metaphase plate measurements. Chromosomes with satellites are enclosed in a box. 6 A 10 // 10// 8 Figs. 6 and 7. Irregular meiosis in C. granatensis Boiss., population Gr., 2 n = 20. In Fig. 6 the 20 chromosomes form 6 bivalents and 2 quadrivalents, one of them indicating the beginning of the formation of a ring of four chromosomes (A), the other one a zigzag chain of four other chromosomes, signs of translocation heterozygote (B). In Fig. 7 the 20 chromosomes form 8 bivalents and 4 univalents (indicated with an arrow). Figs. 8, 9 and 10. Irregular meiosis in C. ornata Willd. var. microcephala Wk., population Mo, 2 n = 20. Fig. 8, anaphase I with chromatid bridges; Fig. 9, metaphase I with a cross-shaped quadrivalent, sign of a translocation heterozygote (indicated by an arrow) ; Fig. 10, metaphase I, the 20 chromosomes form 9 bivalents and 2 univalents (indicated by an arrow). 12 13 10 n Fig. 12. Somatic chromosomes of C. omata Willd. var. microcephala Wk., population K., 2 n = 20. Fig. 13. Somatic chromosomes of C. omata Willd. var. macrocephala Wk., population 1337, In = 40+5B (the supernumerary chromosomes indicated by an arrow). Fig. 14. Somatic chromosomes of C. omata Willd., morphological intermediate form, population 1295, 2 n = 40. M. J. FERNANDEZ-MORALES AND C. GARDOU 69 variable karyotype and its occasionally irregular meiosis. It is probably a taxon undergoing transformation. A more extensive study of a large number of populations in the whole Iberian peninsula would cast light on the genetic problems revealed by the karyological study. (3) Centaurea saxicola Lag. This taxon is only found in the south-east of Spain: it is, as its name indicates, a saxicolous species. Rivas-Goday (1962) considers that it is a characteristic species of the Asplenietea rupestris in the sierras of Callosa, Segura, Orihuela, Crevillente, Carthagena and Carrascoy. We have gathered these plants in the following place: 1326-7: Callosa de Segura, province of Alicante, calcareous rocky walls above St Roch Church, about 100 m altitude. In this population there were two kinds of plants: (a) Some of them were prostrate with short stems (1326), agreeing with the type Centaurea saxicola Lag. var. saxicola. We counted on them In — 60+2^1 B (Fig. 15), Gardou (1972). ( b ) The other ones (1327) were tallest with longer stems, according to the diagnosis of Esteve-Chueca (1965) Centaurea saxicola Lag. var. littorale Est.-Chu. On these plants we observed 2 n = 60 chromosomes without supernumerary chromosomes (Fig. 16), Gardou (1972). We consider that C. saxicola Lag. is a hexaploid and that the supernumerary chromosomes represent the only size difference be¬ tween the two kinds of plants forming the two varieties. In this way it is not right to speak about two varieties in C. saxicola Lag., we only have two forms of the same species. Morphologically these taxa are very close to Centaurea ornata Willd. and both species present supernumerary chromosomes. It is difficult to compare idiograms of both species because we have seen that there were variable karyotypes in C. ornata Willd. But we think that C. saxicola Lag. could have originated from hybridizations between diploid and tetraploid C. ornata Willd. section Cyanus c ass. We studied 6 populations belonging to this section in the following localities: 1236: Mont Lachens (France), Var, south slope near the top about 1700 m altitude; 2 n = 40 (Fig. 3). 1293: Sierra de Guadarrama (Spain), province of Madrid, between Navacerrada and Cercedilla, about 1200 m altitude; 2 n = 20 (Fig. 4). 15 16 \ Fig. 15. Somatic chromosomes of C. saxicola Lag. var. saxicola, population 1326, 2 n = 60+4B (supernumerary chromosomes indicated by an arrow). Fig. 16. Somatic chromosomes of C. saxicola Lag. var. littorale Est.-Chu., population 1327, 2 n = 60. Fig. 17. Somatic chromosomes of C. granatensis Boiss., population 1456, In = 20. M. J. FERNANDEZ-MORALES AND C. GARDOU 71 C. 7: Sierra Nevada (Spain), province of Granada, Dehesa de Dilar, near the river Dilar in the side of the path, about 1800 m altitude; n = 10 (Fig. 22). C. 91: Sierra de Tejeda (Spain), province of Granada, on the nearest mountain in the side of the farms of the Hoyos gardens, about 1700 m altitude, calcareous soil; 2 n = 20. 1376: Djebel Azrou Akchar (Morocco), eastern part of the Rif, calcareous rocks near the top, about 2020 m altitude; 2 n = 20. 1437: Djebel Ayachi (Morocco), eastern part of the Haut Atlas, Imi-n-Thand valley, calcareous crumbled-down stones, under the cedar woods, about 2500 m altitude; 2 n = 20. All these populations can be classified in the polymorphous taxon Centaurea variegata Lmk. (Lamarck, Encycl. meth. 1: 668 (1784)) = C. lingulata Lag. (Lagasca, Gen. et spec. pi. nov. p. 32; 1816) = C. triumfetti All. var. seuseana Gugler (Gugler in Schinz & Keller, FI. Schw. ed. 3, 2: 353 (1914)). This perimediterranean calcicolous taxon possesses a very wide geographical area of distribution. In the western Mediterranean basin it occurs frequently in the south of France (Provence Alps), Spain, Italy, Greece and Morocco. In the eastern Mediterranean basin the total distribution is larger, including Turkey, Syria, Lebanon and eastwards to Armenia and Iran. Within the whole area this taxon occurs on calcareous rocky steppes except for the Guadarrama population (1293) in Spain which is in a siliceous forest community. Five of the six populations are diploid: they are gathered from Morocco and Spain. We observed mitosis in populations 1293, 1376, 1437 and C. 91 with 2 n = 20 chromosomes and meiosis with n = 10 chromosomes in the C. 7 population. In the metaphase plate mitosis, 6 pairs of chromosomes are hetero- brachial, 2 of which are satellite-bearing, and 4 pairs are isobrachial. In the meiosis the chromosomes are paired in diakinesis and in meta¬ phase I forming 10 bivalents normally (Fig. 22). But they sometimes form secondary associations and show chromatid bridges in the first anaphase. We think that it is a stable diploid. It seems that secondary associations are the trace of an aneuploid origin. The sixth population we studied was tetraploid with 2n = 40 chro¬ mosomes amongst which there were four satellite chromosomes: it was the French population of Mont Lachens (1236). From this study we can deduce that the basic chromosome number of these plants is x = 10. In Spain and in Morocco the plant populations are diploid: in these countries different authors named them C. variegata Lmk. or C. lingulata Lag. On the other hand the south of France population is a tetraploid one and Briquet (1931) named it C. triumfetti All. var. seuseana Gugler. But all these six populations are morphologically identical and we can group them under the same name 18 19 Fig. 18. Somatic chromosomes of C. asperah. ssp. stenophylla Wk., population 1309, 2 n = 22. Fig. 19. Somatic chromosomes of C. asperah. ssp. subinermis DC., population 1297 b, 2 n = 22. Fig. 20. Somatic chromosomes of C. asperah. ssp. subinermis DC., population 1452, 2 n = 22+4B (supernumerary chromosomes indicated by an arrow). Fig. 21. Chromosome pairing at first metaphase of meiosis, 11 bivalents, in C. aspera h. ssp. subinermis DC., population Ca., n — 11. Fig. 22. Chromosome pairing at first metaphase of meiosis, 10 bivalents, in C. variegata Lmk., population C. 7, n = 10. M. J. FERNANDEZ-MORALES AND C. GARDOU 73 of Centaurea variegata Lmk., the oldest one. This taxon is represented by diploid and tetraploid populations all along its geographical area. Thus, in the section Cyanus Cass., there are three basic chromosome numbers x = 10, x = 11 and x = 12. (1) The first one with x = 10 is represented by diploid and tetraploid taxa as we have seen before. It is C. variegata Lmk. that we can connect with the oriental species of this group C. fischeri Willd. var. ochroleuca (Willd.) Grossh. with 2 n = 40 according to Podubnaja- Arnoldi (1931) and Tonian (1968) and C. huetii Boiss. with 2 n = 40+2 B according to Tonian (1968). (2) The second one, with x = 11, is also represented by diploid and tetraploid taxa. They can be connected with C. triumfetti All. and C. montana L. according to the results of Guinochet (1957) and Baksay (1957). (3) The third one known, with x = 12, is only composed of annual diploid herbs: it concerns C. cyanus L. s.s. The position of the oriental species C. depressa Bieb. is difficult to solve in this way, because its chromosome number, 2 n = 16, (Chouk- sanova, Sveshnikova & Alexandrova, 1968 and Tonian, 1968) does not correspond to any series amongst the three previously cited. section Seridia juss. In this section we have only studied C. aspera L. from Spain. Most of the plants of this species are remarkably variable herbs. In this way Willkomm (1870) mentioned four subspecies in Spain. The first one, C. aspera L. ssp. aspera, grows widely in south-west Europe: it is the best known one, diploid with 2n = 22 chromosomes according to the results of Maude (1939), Fahmy (1951), Guinochet (1957), Fernandes & Queiros (1971), van Loon (1971) and Gardou (1972). We studied the karyology of five populations belonging to two subspecies in the east and south-east of Spain as follows: (1) C. aspera L. ssp. stenophylla Wk. 1309: El Palmar, province of Valencia, grassland between La Albufera and El Palmar; 2 n = 22 (Fig. 18). 1311: La Albufera, province of Valencia, nearer the ranch of El Saler; 2 n = 22. (2) C. aspera L. ssp. subinermis DC. 1297 b: Arganda, province of Madrid, along the road; 2 n = 22 (Fig. 19). 1452: Venta de las Angustias, province of Granada, 37 km south of Granada, between Beznar and Tablate; 2 n = 22+0-4 B (Fig. 20). Ca.: Lecrin valley, province of Granada, south-west of Sierra Nevada; 2 n = 22, n = 11 (Fig. 21). 74 A STUDY OF THE GENUS CENTAUREA L. The two subspecies studied are diploid as was the case for C. aspera L. ssp. aspera. Only the population 1452 presents 0-4 supernumerary chromosomes. In meiosis, for example in the Ca. population, the chromosomes are regularly associated showing 11 bivalents (Fig. 21). It seems to be a stable diploid. Consequently the endemism and the polymorphism of these Spanish subspecies only occur at the diploid level. Thus, in these taxa, we cannot find a parallelism between the morphological polymorphism and the karyological regularity except for the population showing supernumerary chromosomes. CONCLUSION Some taxa belonging to three sections of the genus Centaurea L. were studied in this report. We found that the relationship between chromo- somic problems and endemism is different from one taxon to another. We also think that the origin of the great number of endemic Centaurea L. species of these sections in south-east Spain can perhaps be found in the reciprocal translocation mechanism. References Baksay, L. (1957). The cytotaxonomy of the species Chrysanthemum maximum Ram., Centaurea montana L., ...in Europe. Annls hist. nat. Mus. natn hung., n.s. 8: 155-68. Braun-Blanquet, J. (1931). Uber die Trockenrasengesellschaften des Hegau und ihre Genese, II. S.I.G.M.A. 7 & Beitr. NatDenkmPflege 14: 3. Braun-Blanquet, J., Pinto da Silva, A. R. & Roseira, A. (1964). Resultats de trois excursions geobotaniques a travers le Portugal septentrional et moyen. Agronomia lusit. 23(4): 229-313. Briquet, J. (1931). Compositae Cynaroidae. In Burnat, E., Briquet, J. and Cavallier, F., Flore des Alpes-Maritimes, 7: 311 pp., Geneve. Chouksanova, N. A., Sveshnikova, L. I. & Alexandrova, T. V. (1968). Data on karyology of the family Compositae Giseke. Citologija 10: 198-206. Essad, S. (1962). Etude genetique et cytogenetique des especes Lolium perenne L., Festuca pratensis Huds. et de leurs hybrides. These, 116pp., Paris. Essad, S., Arnoux, J. &Maia, N. (1966). Controlede validitedes caryogammes. Chromosoma 20: 202-20. Esteve-Chueca, F. (1965). Algunas novedades para la Flora muricana. An. Inst, bot. A.J. Cavanilles 23: 185. Fahmy, T. Y. (1951). Recherches caryologiques sur quelques especes mediter- raneennes. These, 173 pp., Montpellier. Fernandes, A. & Queiros, M. (1971). Contribution a la connaissance cytotax- onomique des Spermatophyta du Portugal, II - Compositae. Bolm Soc. broteriana 45: Ser. 2, 5-121. Gardou, C. (1972). In IOPB Chromosome number reports XXXVII. Taxon 21(4): 495-500. M. J. FERNANDEZ-MORALES AND C. GARDOU 75 Guinochet, N. (1957). Contribution a P etude caryologique du genre Centaurea L. sens. lat. Bull. Soc. Hist. nat. Afr. N. 48: 282-300. Hoffmann, O. (1897). Compositae. In Engler, A. and Prantl, K . , Die natiirlichen Pflanzenfamilien, 4 (4-5): 87-387, Leipzig. Maude, P. F. (1939). The Merton catalogue, a list of the chromosome numbers of species of British flowering plants. New Phytol. 38: 1-31. Podubnaja-Arnoldi, V. (1931). Ein Versuch der Anwendung der embryo- logischen Methode bei der Losung einiger systematischen Fragen. I. Ver- gleichende embryologisch-zytologische Untersuchungen liber die Gruppe Cynareae, Fam. Compositae. Beih. bot. Zbl. II, 48: 141-237. Quezel, P. (1951). Les dolines a neige du Massif du Ghat (Grand Atlas Oriental), quelques aspects de leur peuplement. Bull. Soc. Hist. nat. Afr. N. 42: 121-30. Rivas Goday, S., Rigual, A. & Esteve, F. (1962). Contribution al estudio de la Asplenietea rupestris de la region Sud-Oriental de Espana. An. Inst. bot. A.J. Cavanilles 20. Tonian, Z. R. (1968). The chromosome numbers of some species of the genus Centaurea L. (in Russian). Biol. Zurn. Arm. S.S.R. 21: 86-96. Van Loon, J. C., Gadella, T. W. J. & Kliphuis, E. (1971). Cytological studies in some flowering plants from Southern France. Acta bot. neerl. 20: 157-66. Willkomm, M. & Lange, J. (1870). Prodromus Florae Hispanicae, Stuttgart. STUDIES IN THE GENUS CROCUS B. Mathew and C. A. Brighton ABSTRACT A unique collection of about 90 species of Crocus, mostly of known wild origin, has been assembled at Kew and is being used as the basis for a revision of the genus. The cytology of nearly all the species is being studied comprehensively for the first time using natural source material, and it is hoped that this, together with a critical examination of the morphological details, will assist in a more satisfactory classifi¬ cation than is at present available. Dr E. F. Warburg in 1957 terminated a paper on Crocus species with the statement ‘ It seems impossible with the present information to get any picture of the probable evolution within the group or to devise a really satisfactory classification.’ Unfortunately there has never been a really thorough review of the genus, including disciplines such as morphology and cytology, and although a beautiful monograph by Maw (1886) exists, many new species have been described since then and the genus is in need of careful revision. The first real attempt to study the genus as a whole in recent times was by Mather and Collins at the John Innes Institute in the 1930s. Mather (1932) did publish much cytological data; unfortunately, in many cases it was based on plants of unknown wild origin, and probably no voucher specimens exist, making it impossible to check back on their material. Sadly, Collins died before any of his taxonomic work was published. At Kew we have in cultivation about 900 collections of Crocus, representing 90 of the 100 known species. Many of these have been studied and collected in their wild habitats by us. Originally it was intended to revise the genus on a geographical basis for each of the Floras currently in preparation (i.e. Flora Europaea, Flora of Turkey, Flora Iranica, Flora of Iraq) but it was obvious that such a valuable living collection had a much greater potential for a complete cytotaxo- nomic study, and a revision of the whole genus is now envisaged. Crocus is an Old World genus, distributed from Portugal in the west to Russian Tadzhikistan and Afghanistan in the east, and from Poland in the north as far south as North Africa and Israel. In Europe it is a [76] B. MATHEW AND C. A. BRIGHTON 77 genus mainly of the mountain areas, although in terms of altitude the species on the whole do not reach the alpine zones, preferring the rocky, drier foothills. There are a large number of species in the Balkan region, for example, along the Adriatic coastal mountains of Yugoslavia, the northern Aegean and especially in the Peloponnese and the Greek islands. Also the Mediterranean climate areas of western and southern Turkey contain a large number of species. This great concentration of species in the Balkans has led to the speculation that the origins of the genus lie within this area (e.g. Maw, Feinbrun, Bate-Smith, 1968, 1969). Others (e.g. Bowles, 1924, Greuter, 1968) favour the theory that the origins lie in Africa, since the characters of Crocus , in particular the cormous state and the subterranean ovary, are almost certainly an adaptation to dry arid conditions. The little-known C. boulosii Greuter from Cyrenaica adds weight to this theory, since Greuter considers it to be a primitive species. It seems probable that the origins of Crocus lie in a Romulea- like plant. Romulea is a related genus, the northern hemisphere species of which, unlike Crocus, have an aerial, not subterranean, ovary. Goldblatt (1971) considers that if the origins of Crocus are in a Romulea- type plant, the species with a high number of small chromosomes are the most primitive. C. carpet anus from Spain has a count of 2n = 64 and has a leaf quite unlike the rest of the Crocus species, being semi-terete and more like some Romulea species. It is possible, therefore, that Crocus originated in Africa and migrated to Europe via the Iberian peninsula, C. boulosii and C. carpetanus from Libya and Spain respectively representing the most primitive species now in existence. From the point of view of a practical system of classification for Flora purposes and field identification. Crocus species exhibit a remarkable range of characters. The corm tunics are extremely important and the gross morphological features of these represent some of the major dividing lines between the groups of species. Unfortunately with the corm tunic, as with virtually all other known characters in the genus, no hard and fast dividing lines can be made between the groups of species, and there always remain a few species which do not really fit into any group or section. The tunic may consist of a mass of fine or coarse fibres, these either netted (Reticulati section of Maw), inter¬ woven (Intertexti) or parallel (Fibromembranacei). It may be mem¬ branous or coriaceous without obvious fibres, sometimes the base splitting off as horizontal rings (Annulati), or splitting vertically into strips of tissue. Although such major differences would appear to give an excellent basis for subdividing the genus, it is a very artificial system, for the differences between the tunics of species within, for example, the reticulate group are as striking as those between some of the groups 78 STUDIES IN THE GENUS CROCUS themselves. Perhaps even more fundamental is the absence or presence of a basal spathe subtending the flower. Herbert (1847) used this character to form his divisions Nudiflori and Involucrati. Here again, however, this has been shown to be not entirely reliable, and in populations of some species it would appear that plants with and without this spathe occur. Herbert had trouble with this character for he had a third division, Subnudi, into which he placed the species with an imperfectly formed spathe, regarding these as the transition point between the two divisions. The bract and bracteole which arise at the base of the ovary (the ‘ proper spathe ’ of many authors) also provide a useful method of distinguishing the species. The bracteole can be equal to the bract, much reduced or absent altogether. Although very useful at specific level there is probably no greater significance in this and the absence or presence does not represent any fundamental division within the genus. The degree of division of the style into distinct arms varies from trilobed in its simplest form to much-dissected, and here again this provides good distinguishing features at specific level. Baker (1873) used this to form his subgroups Holostigma, Odontostigma and Schizostigma, but there are so many in-between stages of division that it is considered impossible to use this at a higher rank than that of species. These comments apply to almost any character or group of characters which has so far been chosen, and although some sizeable units made up of closely related taxa exist within the genus, there are always several species which do not readily fit into the groups. The studies at Kew include a comprehensive survey of Crocus seed characters using the scanning electron microscope, and at the present stage the results are promising, for the surface architecture varies enormously throughout the genus. Pollen studies show that the external morphology is not critical from a taxonomic point of view, although Schulze (1970) found that the pollen could be spiraperturate or non- aperturate. A full study of all the species is necessary before it becomes clear if this coincides with any other character. Certainly, of those ten species studied by Schulze (1970), the five with spiraperturate pollen were in section Nudiflori and the five with non-aperturate pollen belonged to section Involucrati. The leaf characters have been studied recently by Culling (1972), and as a result of this comprehensive survey it was suggested that there might be as many as ten groups of related species within the genus. However, the same problem occurs with leaf features as was mentioned above with other characters, that some species do not readily fit into any grouping. It is obviously essential that as much data as possible about all the B. MATHEW AND C. A. BRIGHTON 79 species are acquired before any attempt can be made to make accurate suggestions regarding evolutionary trends, or to construct a funda¬ mental classification. At specific level, however, for the practical purposes of Flora accounts the morphological study at Kew is well advanced. CYTOLOGY A large number of chromosome counts have already been published for the genus Crocus. Mather (1932), Pathak (1940) and Karasawa (1932-56) have contributed the main part of this work, but their material was of unspecified origin and we do not have herbarium material at our disposal. In order to study natural evolution one must look at plants in their natural state, and a knowledge of localities will also make it possible to determine distributions of basic numbers, chromosome races and possible lines of evolutionary development. The earlier workers showed that Crocus is an extremely variable genus, both in chromosome numbers and karyotype morphology. Until recently it was not known whether this reflected the state of the genus in the wild. Feinbrun (1957, 1958) and Sopova (1972) made the first chromosome counts from material from known wild sources and showed that the genus is indeed a very variable one. At Kew we have approximately 900 collections of some 90 species, all from known wild origins extending over the entire range of distribu¬ tion, and this has enabled us to obtain a much more comprehensive picture of Crocus as it occurs in the wild. The initial list of chromosome counts for 88 species has been published (Brighton, Mathew & Mar- chant, 1973), and shows a great range of chromosome numbers including some intraspecific variation. Our results both confirm and extend those of earlier workers. Our investigation has shown that Crocus has chromosome numbers of 2 n = 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 34, 44, 48 and 64. We also found that B-chromosomes occurred in some species, the highest number recorded being 11. This wide range of numbers is unusual, though not unique in the monocotyledons. In the Iridaceae, for example, Iris also has a range of numbers from 2n=\6 to 2n = S4; while in the Liliaceae the chromosome numbers of Ornitho- galum range from 2n = 6 to 2rc = 58, and those of Scilla from 2 n = 10 to 2 n = 54. As well as the numerical variation we also observed substantial j interspecific differences in chromosome morphology. Karyotypes are being prepared and will be published in due course. Some species show a marked stability in their karyotype characteris- 80 STUDIES IN THE GENUS CROCUS tics. For example, C. flavus (2 n = 8) has four pairs of large submedian chromosomes and a distinctive secondary constriction in the shorter arms of one pair which makes the karyotype easily recognizable. C. olivieri (2n = 6) has three pairs of large subterminal chromosomes with a small secondary constriction next to the centromere of one pair of chromosomes. Again it is an immediately recognizable karyotype, completely different from C. flavus , which can provide the taxonomist with important information when classifying his specimens. In addition to interspecific variation we have also found intraspecific differences in some species. In C. cancellatus, for example, we have plants with 2 n = 8, 10, 12 and 16 chromosomes. B-chromosomes also occur in this group. Further variation was found in the karyotype morphology of the plants with 2n = 8 and 10. In both there were two distinct karyotypes. Those of the 2 n = 12 and 2 n = 16 plants were all constant. It is hoped that with the C. cancellatus agg. we will be able to correlate the cytological differences with those of phenotype and geographical distribution which may resolve the group into several taxa. C. heuffelianus presents us with a greater problem, as there are no obvious characters which can be correlated with the differences in chromosome number. In this species somatic numbers of 2 n = 8, 10, 12, 18, 20, 22 and 23 have been found, and mixed populations are known to occur. Population samples have been taken and a more comprehen¬ sive study is in progress. Intraspecific variation of this kind was also found in Ornithogalum by Cullen & Ratter (1967). In a variable genus like Crocus one must consider the possible existence of polyploidy, especially in the higher numbers. From studies of the karyotypes it has been impossible to determine a polyploid series or polyploid relationships within groups, but it is hoped that meiotic studies will throw more light on this problem. A study of this kind illustrates that in looking at material from natural populations one can obtain a more realistic picture of the genus, and that close co-operation between cytological and taxonomic disciplines can be very profitable. We hope that continued studies along these lines will enable us to classify Crocus adequately and provide useful evidence on lines of evolution. References Baker, J. G. (1892). Handbook of Iridaceae: 76-95. Bate-Smith, E. C. (1968). The phenolic constituents of plants and their taxo¬ nomic significance. J. Linn. Soc., Bot. 60: 325-56. Bate-Smith, E. C. (1969). Flavonoid patterns in the monocotyledons. Ch. 8 in Harborne & Swain, Perspectives in Phytochemistry. B. MATHEW AND C. A. BRIGHTON 81 Bowles, E. A. (1924). A handbook of Crocus and Colchicum. Brighton, C. A., Mathew, B. & Marchant, C. J. (1973). Chromosome counts in the genus Crocus (Iridaceae). Kew Bull. 28(3): 451-64. Cullen, J. & Ratter, J. A. (1967). Taxonomic and cytological notes on Turkish Ornithogalum. Notes R. bot. Gdn Edinb. 27: 293-339. Culling, C. (1972). The taxonomy of the genus Crocus with particular reference to leaf and seed structure. M.Phil. Thesis, Queen Elizabeth College, London. Feinbrun, N. (1957). The genus Crocus in Israel and neighbouring countries. Kew Bull. 12(2): 269-85. Feinbrun, N. (1958). Chromosome numbers in Crocus. Genetica 29: 172-92. Goldblatt, P. (1971). Cytological and morphological studies in the S. African Iridaceae. Jl S. Afr. Bot. 37: 317^160. Greuter, W. (1968). Crocus boulosii. Candollea 23: 45-49. Herbert, W. (1847). A history of the species of Crocus. JlR. hort. Soc. 2: 249-93. Karasawa, K..(1956). Karyological studies in Crocus. IV. Genetica 28: 31-4. Mather, K. (1932). Chromosome variation in Crocus. I. J. Genet. 26: 129-42. Maw, G. (1886). A monograph of the genus Crocus. Pathak,G. N. (1940). Studies in the cytology of Crocus. Ann. Bot. Sn.s. 4: 227-56. Schulze, W. (1970). Beitr. zur Pollenmorphologie der Iridaceae. Wiss. Z. Friedrich Schiller-Univ. Jena 19: 437-45. Sopova, M. (1972). The cytology of ten Crocus species from Macedonia. Godisen Zb. filos. Fak. Univ. Skopje prirodno-mat. Oddel. 24: 73. Warburg, E. F. (1957). Crocuses. Endeavour 16(64): 209-16. STUDIES IN THE GENUS MYOSOTIS J. Grau ABSTRACT The paper considers the value of cytological studies in elucidating taxonomic and evolutionary relationships in the genus Myosotis in Europe. Information is given on M. decumbens and its distinction from M. sylvatica ; the former has not previously been recognized as a member of the British flora, and its old herbarium records from Scotland need verification. The genus Myosotis - the forget-me-nots - is so well known that it might be unnecessary to explain what a Myosotis looks like. There is a great resemblance in at least most of the European species, and this causes some problems in determining some of the species. Though the genus also has a considerable part of its area in the southern hemisphere, especially in Australia and New Zealand, I want to restrict my lecture to the European species. To clear up the connections and to give some hints for a better circumscription of the species, cytology was of great help. I want to explain in the two parts of my talk what seems to be behind this great similarity of species. I also want to demonstrate that the Mediterranean area is of great use in solving Central European systematic problems. The genus Myosotis is represented in Europe by three ecological groups. The first of these groups consists of perennial species which grow in woody or alpine areas. A second group prefers very humid areas ; these are the marsh forget-me-nots such as Myosotis scorpioides and M. caespitosa. The only thing I want to say about this group here is that it forms a polyploid series ending with octoploid species. The third group consists of annual herbs which have their centre of distribution certainly in the Mediterranean area but spread into Central Europe too. These species can again be subdivided into three units with clearly different centres of origin. The first subgroup certainly originates in the Aegean area. It has only straight hairs covering the whole plant. The northernmost place where it can be found is southern France. All five species of this subgroup have 24 chromosomes. (I have to digress here to say a few words [82] J. GRAU 83 concerning the chromosome numbers of Myosotis. In the Boraginaceae the basic number x = 12 is very frequent. It is also to be found in many of the species of Myosotis and this number must be without any doubt the primary basic number of this genus. In many cases it is possible to show how from this secondary basic numbers can rise. One example of this I want to show later on. Going back to our first subgroup we can say now that all its species are diploid.) In the second subgroup, the centre of which is a little further to the east and which is distributed more widely, all parts of the plant have hooked hairs. Now this group has a very strange type of distribution. It is to be found in spots and very limited areas towards the west up to the Sierra Nevada in southern Spain. Strangely enough the two most important species of this group have been described from these most remote places of growth. Cytologically this group is more variable, that is to say we can find diploids and polyploids, euploids and aneu- ploids. A third subgroup has its centre on the Iberian peninsula. This group is distinguished by its yellow flowers and very large pollen grains. It consists of a polyploid series up to hexaploid species with 72 chromo¬ somes. This hexaploid species is the only one in Central Europe and is of rare occurrence in northern and eastern Europe. After this subdivision only a few annual taxa are left which do not fit into this system. M. ramosissima is one of these species. It shows a little focus in western Europe, and one of its subspecies reaches in a typical Atlantic distribution from Portugal via France to southern England. This is the subspecies globularioid.es with different calyces and nutlets. In M. ramosissima s.l. only polyploids exist. Perhaps they are allopolyploids connecting the groups mentioned above. The second species which belongs to the annuals and also does not fit into the three subgroups is M. arvensis. But it seems that this species is, unlike the above-mentioned species, a secondary annual and derives from perennials as we shall see later on. A diagram (Fig. 1) gives a summary of the groups mentioned. On the left the euploids - on the right the few aneuploids, at the bottom the diploid Mediterranean species, at the top the mainly northern poly¬ ploids. These show moreover a widespread distribution while the Medi¬ terranean diploids are found to be more or less local without a coherent area. M. arvensis does not really belong to the annuals, as I said before, and therefore it leads to the second part of my talk. This part concerns the perennials of alpine or woody habitats. In contrast to the annuals, which are quite satisfactorily distinguished and classified, in these species the poverty of characters is so great that at 84 STUDIES IN THE GENUS MYOSOTIS euploid aneuploid ©incrassata gr. ©minutifiora-gr. ©ramosissima-gr Fig. 1. Diagram showing cytological relationships and geographical distribution of annual species of Myosotis. least sometimes there has been the tendency to treat them as only one aggregate. Only a cytological investigation, which showed very many different chromosome numbers, brought the possibility of a clear separation of species in quite a number of cases. To explain this I have to detail a little the morphology of chromo¬ somes in Myosotis. We can distinguish two chromosome types within the European species: metacentric ones with more or less equivalent arms, and acrocentric ones with a very reduced second arm. This second type especially is very remarkable and can be used to charac¬ terize the chromosome sets. The perennial species I talk about can roughly be divided into two groups. Within the alpestris group there J. GRAU 85 M.alpestris ■ ■ li ■ ■ M.sylvatica ssp.elongata M.sylvatica ssp.cyanea M.sylvatica ssp.sylvatica !■■... HIM1 Fig. 2. The karyotypes of different perennial species of Myosotis. On the left the acrocentric chromosomes, on the right the metacentric chromosomes. exist diploids, tetraploids and hexaploids, which means that we have a regular polyploid series. This series is founded on the basic number x = 12. The sylvatica group, however, shows the numbers n = 9, 10, 11, 14, 16 and 17 without any representative of the basis n = 12. This seems to be an aneuploid series with some gaps. A detailed investigation, however, showed that this series is not directly connected. The taxa with n = 11, 10 and 9 above all form a separate descending aneuploid series starting with an unknown ancestor with the chromosome number n = 12. This can very easily be explained considering the two chromo¬ some types. We can see this in Fig. 2. I assume that we can start with 86 STUDIES IN THE GENUS MYOSOTIS M.sylvatica M.sylvatica 4n i M. decumbens! I I ■ ■ ill . Fig. 3. The karyotype of M. sylvatica, of its hypothetical tetraploid form, and of M. decumbens. Lost chromosomes open. a karyotype similar to that of M. alpestris , which has 7 metacentric and 5 acrocentric chromosomes in the haploid set. Twice in the course of the development the number of acrocentric chromosomes was reduced by one chromosome. The results of the process are the numbers of n = 1 1 and n — 10. In the last step finally two metacentric chromosomes fused and formed a new one with equal arms. This is the case in the type subspecies M. sylvatica. All taxa with these three derived numbers are very closely related and form the four subspecies of M. sylvatica in Europe. But how could the chromosome number higher than the original basis 12 in the sylvatica group evolve? Even in this case the chromosome types are very helpful. In Fig. 3 we see again the karyotype with n = 9 chromosomes, the development of which we have just seen. When we imagine this chromosome set doubled we obtain the karyotype below. In other words a simple polyploidization would have occurred. Comparing this hypothetical tetraploid set of chromosomes of M. sylvatica with the existing set of M. decumbens we can find an astonishing correspondence. Only the two white marked chromo¬ somes are missing in the lowest row. This correspondence could in my opinion be interpreted in only one way. After the polyploidization of a sylvatica- like plant a continued loss of chromosomes occurred down to a north Italian subspecies of M. decumbens which only has 14 chromosomes left. J. GRAU 87 Fig. 4. Diagram showing cytological and morphological characters of the European species of the M. sylvatica and M. alpestris groups. Outline dia¬ grams represent nutlets and fruiting calyces. We can see the situation just described in a diagram (Fig. 4). Black dots mean existing species or subspecies, white ones hypothetical forms. On the left you will see the polyploid series of M. alpestris with quite a rich development of diploid species. In the middle there is the aneuploid series of M. sylvatica. On the right side the species and subspecies around M. decumbens descend down to n = 14. In addition I have put some of the main morphological characters on the picture. Within M. sylvatica and M. decumbens s.l. the fruiting calyces are round at the base and deciduous, with M. alpestris they are narrowed at the base and persistent. The nutlets of M. alpestris are obtuse with a large lateral folded attachment area, those of M. sylvatica are acute with a small attachment area. Very similar are those of M. decumbens. Combined with the persistence of the fruiting calyces in M. alpestris hooked hairs are quite rare, though always present in the two other 4 UBG 88 STUDIES IN THE GENUS MYOSOTIS groups. These hairs are of different length in M. sylvatica and M. decumbens. Besides this character the calyx teeth in M. decumbens are broadly triangular against narrowly triangular ones in M. sylvatica. Also the attachment area of the nutlets in M. decumbens is a little larger. Also included in the diagram is the more or less ruderal, annual to biennial M. arvensis with its two cytologically different subspecies. This species has closed fruiting calyces which are very rich in hooked hairs, which means a better adaptation to dispersal of the fruit as a whole. We could ask now: why is this great cytological diversity connected with such a morphological uniformity? But I think that this is not the right question to ask. Rather, we should state that it is just this uniformity which allows the cytological processes mentioned above. The first step in the development of this rather young group is a cytological change. The morphological consequences may follow later. We can see this also in the genus Pulmonaria , where the discrepancy of morphological and cytological behaviour is much greater. Some further words concerning the areas of the species dealt with. M. alpestris is to be found, as far as Europe is concerned, in the higher mountains and sometimes uplands of continental Europe with the exception of Scandinavia. In Great Britain it exists only in Scotland and N. England. M. sylvatica grows in wide areas of Central Europe extending to southern Scandinavia and Great Britain. Of its cyto¬ logically more primitive subspecies the ssp. elongata is to be found in Sicily and southern Italy, the ssp. cyanea mainly in the Balkan Peninsula. M. decumbens has a very interesting area, probably influenced by glaciation. It exists in the type subspecies in the southern half of the alpine chain, the uplands of France, the Pyrenees, northern Spain and the Sierra Nevada, and parts of Scandinavia. In Germany it grows only in the south-western part of the German Alps, the Allgau, where we recovered it after nearly 100 years of presumed extinction. And last but not least the same situation may also exist in Great Britain. The only two records of this species originate again from Scotland, but since they also are more than a hundred years old, they need to be verified in the field. Details of the specimens are: ‘Perth, leg. Gardiner, 1846’. (M) ‘Ben Lawers, Perthshire, 18 — ’ (W). J. GRAU 89 References Grau, J. (1964). Die Zytotaxonomie der Myosotis alpestris- und der Myosotis silvatica-G ruppe in Europa. Ost. bot. Z. Ill: 561-617. Grau, J. (1968). Cytotaxonomische Bearbeitung der Gattung Myosotis L. III. ’ Die annuellen Sippen. Mitt. bot. Miinchenl: 17-100. Grau, J. (1970). Cytotaxonomische Bearbeitung der Gattung Myosotis L. IV. Erganzende Studien. Mitt. bot. Miinchen 8: 127-36. CYTOTAXONOMY OF THE GENUS LEU CANTHEMUM IN YUGOSLAVIA D. Papes ABSTRACT The paper reports on the cytotaxonomy of the 1 1 species of Leucan- themum occurring in Yugoslavia, and makes tentative suggestions as to their micro-evolutionary relationships. The morphological variety and differences in chromosome number exhibited by Leucanthemum have attracted the interest of several cytotaxonomists during the last twenty years, yet areas remain where cytological information is lacking. This was especially true of the Yugoslav representatives, the investigations of which started only recently (Mirkovic, 1966, 1969; Papes, 1971 a, 1971 b, 1972a, 1972h, 1972 c and 1973). Leucanthemum is a highly polymorphic and widely spread genus. It grows all over Europe, North Africa and eastern Asia, and is naturalized in North America and Australia. Its polymorphic character is particu¬ larly conspicuous in Yugoslav representatives, within which we dis¬ tinguished 11 species (Horvatic, 1963; Papes, 1972a), which vary also in their chromosome numbers and ploidy levels (2x to 8x) accompanied by aneuploidy and the existence of B chromosomes (Fig. 1) (Papes, 1971 b, 1972 c). The present paper deals with cytotaxonomy and some evolutionary aspects of possible origins and relationships of karyotypes in Leucan¬ themum species. For this purpose morphological features have been considered in conjunction with chromosome numbers, satellite form and frequency. 427 individuals belonging to 99 populations of Yugoslav Leucanthemum species were investigated. Cytotaxonomic methods applied here included karyological, taxonomic, morphological, eco¬ logical and phytogeographical investigations, and some study of the breeding system. Among these 11 species from Yugoslavia only two are diploids: L. rot undifolium (W.K.) DC. and L. praecox Horvatic. L. rotundifolium , 2 n = ( 2x ) = 18, is an old species, which has been completely isolated in almost all aspects from other species for a long time. It has a special morphology of leaves (Fig. 2); its locality on an old mountain - [90] 91 Fig. 1. Diagram of possible origins and relationships of karyotypes in Leucanthemum species. 92 LEUCANTHEMUM IN YUGOSLAVIA Fig. 2. Leaves of stem in Leucanthemum rotundifolium. Vranica in central Bosnia - is the only one in Yugoslavia. L. rotundifolium is the only annual, fully self -fertile species, and is therefore uniform, without any variability. The other 10 species (Fig. 1) are mainly self-sterile, depending exclusively on outcrossing and thus capable of cross-fertilization in suitable combinations. This characteristic is responsible for the great variation of chromosome numbers among the species and even within a population of one species. In this way different levels of polyploidy arising from the basic number 9 have been found, as for instance 2jc, 4x, 5x, 6x, lx and 8x, accompanied by appearance of aneuploidy and the existence of B chromosomes. Among these 10 species only L. praecox Horvatic is a diploid (Fig. 3), while all other species are polyploids of hybrid origin, being either allopolyploids or autoallopolyploids. According to the results of inves¬ tigations, especially the analysis of meiosis and data of the high degree of fertility, it has been stated that autopolyploids do not exist among them. In the meiosis of all polyploids, bivalents predominate over univalents and rare multivalents (which have no more than 6 chromo¬ somes in associations). D. PAPES 93 Fig. 3. Leaves of rosette and stem in Leucanthemum praecox. The representatives of these species show a considerable cytological, morphological, ecological and phytogeographical diversity. However, according to their related characteristics, they may be divided into two groups. The first group includes the following four species: L. praecox Horvatic, 2 n = ( 2x ) = 18; L. vulgare s.s Lam., 2 n = (4x) = 36; L. roh- lenae Horvatic, 2 n = (4x) = 36, (6jc) = 54; and L. leucolepis Briq. et Cav., 2n = (4x) = 36, (6jc) = 54. Certain morphological characteris¬ tics are important for this group as for instance: branched stem, usually hairy stem and leaves; soft, wide leaves with enlarged base, edges deeply toothed in L. praecox (Fig. 3) and L. rohlenae , and slightly toothed in L. vulgare (Fig. 4). The pappus is completely undeveloped. In L. praecox and L. vulgare the involucral bracts have dark edges and in L. rohlenae and L. leucolepis light ones. The representatives of these species are perennials flowering already in the first year. Sexual reproduction is predominant. The populations occur in habitats influenced by man, e.g. abandoned semi-cultivated fields and road edges, and rarely on pastures and meadows. The second group of species (Fig. 1) consists of high polyploids: octoploids (L. heterophyllum , L. liburnicum and L. illyricum ) and hexaploids ( L . croaticum and L. chloroticum ), occurring in the coastal region at a low altitude and on seaside mountains, as well as on inland mountains. All species of this group are perennials flowering not before the second or third year. Apart from sexual reproduction, vegetative 94 LEUCANTHEMUM IN YUGOSLAVIA Fig. 4. Leaves of stem in Leucanthemum vulgare. reproduction is very frequent. The habitats of their populations are mainly untouched: limestone cracks in the alpine zone above the wood border and rocky mountain pastures. Their common morphological characteristics are: unbranched stem, smooth and leathery leaves and stem, narrow leaves, pinnate (Figs. 5 and 6) or narrowly linear; and achenes of ray florets always having the pappus in the form of a membranous rim. The edges of the involucral bracts are darker in L. heterophyllum and to a certain extent in L. illyricum and L. liburnicum, and lighter in L. croaticum and L. chloroticum. L. heterophyllum (Willd.) occurs in high mountains above 1000 m in the north-western part of Yugoslavia, its individuals are octoploids sometimes with 1-2 B chromosomes and very often aneuploids as well. The other 4 species are coastal endemic species: L. liburnicum , L . croaticum , L. illyricum and L. chloroticum (Fig. 7). These were separated from the complex central European species L. atratum Jacq. where they were treated as subspecies. Although having many common characteristics, the representatives of the four species, studied from natural populations, display a considerable cytological, morphological D. PAPES 95 Fig. 6. Leaves of rosette and stem in Leucanthemum illyricum. DISTRIBUTION OF THE ENDEMIC LEUCANTHEMUM SPECIES LIBURNICUM Horvatic 2n (8x) = 72 CROATICUM Horvatic 2n (4x, 5x, 6x) = 36, 45, 54 ILLYRICUM Horvatic 2n (4x, 6x, 8x) = 36, 54, 72 [V] CHLOROTICUM Kerner and Murbeck 2n (6x) = 54 Fig. 7. Distribution of four coastal endemic Leucanthemum species. Fig. 8. Leaves of rosette and stem in Leucanthemum amplifolium. D. PAPES 97 Aneuploids 2n±2— 4 2n+l | Norfnal-2n 1-2 B’s 2-5 B’s 5-10 B’s Fig. 9. Frequency of aneuploids and B chromosomes of the Leucanthemum species in Yugoslavia. and geographical diversity. They are polyploids and often aneuploids, gaining and losing 1-3 chromosomes and some of them having B chromosomes. Their geographical distribution is shown in Fig. 7: L. liburnicum grows in the northern part of the Adriatic, L. croaticum occurs at all altitudes up to 1100 m above sea level, L. illyricum on the high mountains in Hercegovina and L. chloroticum on the mountains of Montenegro. L. liburnicum Horvatic; from 6 populations were octoploids or aneuploids with 2 n = 69, 70, 72, 73 and 74+1-3 B. L. croaticum Horvatic (Fig. 5); in 7 populations the plants were usually hexaploids (2 n = (6x) = 54) but some of them were penta- ploids (2 n = (5jc) = 45) and a few tetraploids (2 n = (4x) = 36). L. illyricum (Horvatic) Papes (Fig. 6) was usually octoploid (2 n = (8x) = 72) and only some of them were tetraploids. L. chloroticum Kerner et Murbeck; the chromosome numbers of its populations are 2 n = (6jc) = 53, 54 and 55. L. amplifolium (Fiori) Papes, 2 n = (6x) = 54 and 2 n = (5jc) = 45 (Fig. 8) is an interesting species, which could represent a typical interspecific hybrid between the species of the above mentioned two groups. The most favourable parental characteristics have been crossed here, so that the newly formed species was able to survive in new habitats such as forest borders and woodland clearings. 98 LEUCANTHEMUM IN YUGOSLAVIA The results of this paper show that aneuploids are very frequent and that they occur only in high polyploids (Fig. 9). It has been stated that the percentage of aneuploids grows with higher levels of polyploidy. There are no deviations in the chromosome numbers in diploids. In tetraploids they only occurred in one sample, whereas more than 30 per cent of hexaploids, heptaploids and octoploids were aneu¬ ploids. The presence of B chromosomes was found in 1 1 populations out of 99 (Fig. 9). Their number varied from 1 to 10, depending on the species, the ploidy level and the method of reproduction in the popula¬ tion. B chromosomes were found in polyploids at all ploidy levels (4x, 5x, 6x, 8x), but they have never been found in diploids. Although in many genera the B chromosomes are a feature of diploids and are thus rarer in polyploids, in Leucanthemum the reverse appears to be the case. Their morphology varies at different ploidy levels, from popula¬ tion to population. There are three types of B chromosomes: small metacentrics in the tetraploid L. vulgare 1-6 Bs, and in the heptaploid (hybrid), very small metacentrics in the octoploid L. illyricum and telocentrics or nearly telocentrics in L. liburnicum (Papes, 1971 b). Now, if we want to say something about the evolutionary aspects of the 11 native Leucanthemum species from Yugoslavia, we need to look again at the diagram of possible origins and relationships of their karyotypes, where morphological features have been considered in conjunction with chromosome numbers, satellite form and frequency (Fig. 1). Only two species are diploids: L. rotundifolium and L. praecox. L. rotundifolium (Fig. 2) is completely isolated from the others, no other species having developed from this. L. praecox ( Fig. 3) could be one of the parental diploid species, since its three pairs of satellited chromosomes occur also in the karyotypes of the species L. vulgare , L. rohlenae and L. leucolepis. However, there should still exist some diploid ancestors with at least two different characteristics, namely four submetacentric chromosomes bearing distal satellites on the short arm, and pale involucral bracts. L. praecox was able to survive as a diploid in seasonal isolation, flowering early in spring before the others. The second group of species consists of high polyploids, so it is possible to assume that the ancestors of these species originated from outside Yugoslavia and should be searched for among the diploids of the Central and Southern Alps and south-western Europe, but they may even have originated from North Africa and western Asia. As for the four endemic species, considering other authors (Contan- driopoulos, 1964; Favarger & Villard, 1965; Guinochet & Logeois, 1962; Polatschek, 1966 and Villard, 1970), we have found that they are D. PAPES 99 very closely related to the Leucanthemum representatives in Europe, especially in the Mediterranean region, as is shown in the following: L. liburnicum Horvatic appears as an octoploid and its allies are L. halleri (Sut.) Polatschek 2x, L. atratum (Jacq.) DC. 2x, 6x and L. coronopifolium Vill. 6x from the Alps. L. croaticum Horvatic is hexaploid and, regarding morphology of leaves, its allied species extend far into western Europe and even into North Africa, these being L. hosmariense Ball. 2x, L. cebennense DC. 4x, L. monspeliense (L.) 4x, L. corsicum DC. 4xand L. ceratophylloides (All.) 6x. L. illyricum (Horvatic) Papes is octoploid and its relatives grow in the south of the Apennine peninsula. L. chloroticum Kerner et Murbeck is hexaploid and its allies are: L. graminifolium Lam. and L. Burnatii Briq. et Cav., which are diploids, and L. tridactylites Kerner et Hunter which could be tetraploid. The fact that so many polyploid-endemic species have been found may indicate that, in the evolution of Leucanthemum species, numerical changes are more important than structural changes. ACKNOWLEDGMENTS Part of this work was prepared in the Jodrell Laboratory, Royal Botanic Gardens, Kew, while the author was the recipient of a British Council Scholarship. I offer my special thanks to Dr Keith Jones and his staff for their help and assistance in the work. My thanks also go to Professor Z. Devide for his comments on the paper. References Contandriopoulos, J. (1964). Recherches sur la flore endemique de la Corse et ses origines. Revue gen. Bot. 71: 361-84. Favarger, C. & Villard, M. (1965). Nouvelles recherches cytotaxonomiques sur Chrysanthemum leucanthemum L. sens. lat. Bull. Soc. Bot. Suisse! 5: 51-19. Guinochet, M. & Logeois, A. (1962). Premieres prospections caryologiques dans la flore des Alpes maritimes. Revue Cytol. Biol. veg. 25: 465-79. Horvatic, S. (1963). Genus Leucanthemum in Flora Jugoslaviae. Acta bot. croat. 22: 203-18. Mirkovic (Papes) D. (1966). Citotaksonomska istrazivanja vrsta roda Leucan¬ themum Adans. em Briq. et Cav. na podrucju Jugoslavije. Acta bot. croat. 25: 137-52. Mirkovic (Papes) D. (1969). Citotaksonomska i citogenetska istrazivanja roda Leucanthemum na podrucju Jugoslavije. Poredbeni studij diploidnih vrsta: L. rotundifolium (W.K.) DC. i L. praecox Horvatic. Acta bot. croat. 28: 245-52. Papes, D. (1971 a). Distribucija triju priobalnih endema: Leucanthemum libur¬ nicum Horvatic, L. croaticum Horvatic i L. chloroticum Kerner et Murbeck. 100 LEUCANTHEMUM IN YUGOSLAVIA I jugoslavenski simpozium biosistematicara. Zbornik referata , 99-108. Sarajevo. Papes, D. (1971 b ). B chromosomes of the genus Leucanthemum in Yugoslavia. Genetika 3: 216-70. Papes, D. (1972 a). Citogenetska istrazivanja roda Leucanthemum Adans. em. Briq. et Cav. Doktorska disertacija. (With English Summary) 1-218. Prirodoslovno-matematicki fakultet Sveucilista Zagreb. Papes, D. (1972 b). Pentaploids in populations of various Leucanthemum species. Acta bot. croat. 31: 71-9. Papes, D. (1972 c). Variation of chromosome numbers in natural Leucan¬ themum heptaploids. Acta bot. croat. 31: 81-5. Papes, D. (1973). Leucanthemum illyricum (Horvatic) Papes, comb. nov. et stat. nov. Acta bot. croat. 32: 243-51. Polatschek, A. (1966). Cytotaxonomische Beitrage zur Flora der Ostalpen- lander, II. Ost. bot. Z. 113: 101-47. Villard, M. (1970). Contribution a l’etude cytotaxonomique et cytogenetique du genre Leucanthemum Adans. em. Briq. et Cav. These. Bull. Soc. Bot. Suisse 75: 96-188. HYBRIDIZATION IN YELLOW-FLOWERED EUROPEAN RORIPPA SPECIES B. Jonsell ABSTRACT The hybridization of three species of Rorippa in Europe is considered from the cytogenetic and the ecological points of view, and the situations seen in wild populations are tentatively explained. The role of man in extending the habitats available for vegetatively vigorous hybrid clones is estimated especially for wild lake-shore populations in Sweden. INTRODUCTION Rorippa, a genus of world-wide distribution, numbers approximately 80 species, most of which belong to the yellow-flowered section Rorippa. Three European species of this section, all long-lived perennials and pronounced outbreeders, contribute in particular to hybridization in the genus, viz. R. amphibia (L.) Bess, and R. sylvestris (L.) Bess., both widespread in much of Europe, and the eastern R. austriaca (Cr.) Bess, (map in Meusel, Jager & Weinert, 1965; cf. also Jonsell, 1973), a rare alien in Britain and Scandinavia. Also R. palustris (L.) Bess., a short-lived perennial and highly autogamous species, may be involved in hybrids, though nearly always in highly sterile Fx stages. In particular cases, however, a concealed, but more far-reaching role of this species cannot be excluded (see below). I will here consider mainly the interactions between R. amphibia and R. sylvestris, to some extent those between the latter and R. austriaca, and only touch on R. palustris in passing. Between the three first-mentioned hybridization has for a long time been observed, and it is well documented in herbaria. Whole series of hybrid derivatives bridging the morphological gaps between these species are present, and there are local areas where considerable fusion between them has occurred. This is particularly evident in some river systems of Central and Eastern Europe, where all three species grow together. My own experiments and field studies were based upon material from Northern and Western Europe, where the hybrid situations are more clear-cut. Fig. 1 surveys a few characters of the species involved and gives examples of some hybrid derivatives. [ 101 ] 102 HYBRIDIZATION IN RORIPPA Fig. 1 . Schematic survey of median cauline leaves and ripening fruits with their pedicels in some Rorippa species and their hybrids. According to the amount of variation one or more examples are illustrated from each species or hybrid. POLYPLOIDY AND CROSSING ABILITY It should first be emphasized that we are dealing with a polyploid complex, and that hybridization of significance takes place only at tetraploid or higher levels. Fig. 2 illustrates the ploidy situation and the ways in which hybrids are known to be formed in nature. For example, R. amphibia is in Britain both diploid (e.g. the Thames population) and tetraploid (Jonsell, 1968, p. 104); in Northern Europe only the latter. B. JONSELL 103 2n 48 40 32 24 16 2n R.au x sy 48 40 32 24 R.au 16 Fig. 2. Survey of polyploidy levels ( x = 8) and spontaneous hybrid formation in hybridizing species of Rorippa sect. Rorippa. Along unbroken lines hybrid swarms occur; bold hatched lines indicate formation of fertile intermediates; thin hatched lines formation of nearly or totally sterile intermediates. See further in text. R. sylvestris occurs frequently both as tetraploids and as hexaploids , both cytotypes apparently being widespread in Europe, at least in the north and west (Jonsell, 1968). The tetra- and hexaploid cytotypes mentioned are all actively hybridizing, which thus partly leads to the formation of pentaploids. R. austriaca is known as a diploid only, which offers a problem as to the way in which tetraploid R. austriacax sylvestris, important in large parts of Europe, is formed. The mathematically possible genesis from R. austriaca (2n = 16 )xR. sylvestris (2n = 48) seems less probable as the exclusive or more important way, both for morphological reasons and from the fact that we have practically unbroken series of highly fertile hybrid derivatives from one species to the other. Moreover, crosses attempted between these cytotypes failed completely (Fig. 3). Unreduced R. austriaca gametes may instead play an important role. Diploid R. amphibia and R. austriaca sometimes form hybrids with tetraploids (cf . also Javurkova-Kratochvilova & Tomsovic, 1972). The resulting triploids might gain local importance, but owing to sterility, both male and female, they seem to be evolutionary ‘dead ends’. Another triploid is the hybrid between R. amphibia (2 n = 16) and R. palustris (2 n = 32), well represented in British herbaria from the 104 HYBRIDIZATION IN RORIPPA Fig. 3. Crossing polygon indicating maximum pollen fertility in Fj-products of each combination. Hatched line = crossing failed; thin line = no progeny reaching flower; double thin line = pollen fertility in no case over 30 per cent; bold line = ditto in no case over 80 per cent; double bold line = ditto largely over 80 per cent. The polygon is mainly based on the experiments reviewed in Jonsell, 1968, pp. 42-3. Thames banks around Kew, where it spread vegetatively. It was examined cytologically by Howard (1947). Let us now turn from the situation in natural populations to the information about their potential to form hybrids provided by crossing experiments. The crossing polygon (Fig. 3) illustrates the pollen fertility of obtained from wild plants; the hybrids included are also of wild origin. The progeny from each combination are, of course, varyingly fertile; in the polygon the maximum fertility observed in any derivative is shown in order to present an idea of the best than can be expected of that combination. The crossing ability is, as far as the outbreeding species are concerned, in general high at the tetraploid level, both B. JONSELL 105 Fig. 4. Survey of two cases with aneuploid progeny raised in cultivation. (Number of individuals is plotted against somatic chromosome number; the black squares indicate chromosome number of parent plants.) A. Progeny raised from spontaneous seeds of a pentaploid R. amphibiaxsylvestris hybrid, apparently back-crossed to tetraploid R. amphibia. B. Progeny raised from reciprocal crossings between pentaploid and hexaploid R. sylvestris. between species and in back-crosses to wild hybrids. Crosses between tetra- and hexaploids were performed with about equal facility, and the pentaploids, too, can show good crossing abilities, yielding vegetatively vigorous progeny with various aneuploid chromosome numbers (Fig. 4B). Crosses involving R. palustris or any of the diploids gave much poorer results, both those between various diploids and those between a diploid and some other level. Remarkable is the facility of tetraploid R. amphibia to cross with others at the same or higher levels, in contrast to the fact that crosses between the two R. amphibia cytotypes failed to give viable progeny. Above the diploid level genetical isolation between the outbreeding species is apparently not very effective. But had these plants to rely totally or essentially upon their seeds for reproduction and dispersal, the isolation would certainly be effective enough to reduce the forma¬ tion of hybrids to short-lived events without major consequences for the population structures. 106 HYBRIDIZATION IN RORIPPA DISPERSAL AND ESTABLISHMENT The genesis of hybrids in nature is greatly favoured by both the self-sterility and the clonal behaviour of the three species, R. amphibia, R. austriaca and R. sylvestris. The self-sterility is very pronounced, as a rule absolute (Jonsell, 1968, p. 35). They have all, in partly different ways, effective means of vegetative propagation and dispersal. R. amphibia spreads by fragments of basal parts of its shoots, R. austriaca and R. sylvestris especially by portions of root-runners able to produce adventitious shoots. Local populations often consist of only a single long-lived clone. The presence in such an area of another of these species implies a very high possibility that hybrid seeds will be formed. Once a hybrid has become established, the chances for back-crossing and resulting hybrid swarms are considerable. The hybrids spread by their seeds and by shoot or root fragments. By the latter a successful hybrid derivative may build up a locally important, dominant or even exclusive population. Except for cases of anthropochory, vegetative dispersal would on the whole imply confinement to one river system. The hybrid seeds, however, may be spread over long distances, perhaps as a rule unintentionally by man, as is remarkably evident in the case of R. austriacax sylvestris (cf. below). Another example is the almost simultaneous discovery of R. amphibia and its hybrid with R. sylvestris in a place on the River Torne in northern Finland, very far from any previously known locality of the species involved (Hylander, 1960, 1961 ; Jonsell, 1968, p. 130). The agency that was here responsible for the probable seed transport can, of course, only be guessed at. Weak genetic barriers in combination with pronounced self-sterility and the ability to build up long-lived clones constitute the basic conditions for successful hybridization. But there would still be no success without suitable habitats for the hybrid products. The eco¬ logical conditions are of utmost importance for the establishment of the hybrids, a crucial point being the chances for seedlings to grow up among competing vegetation. Let us first deal with the R. amphibiax sylvestris hybrids and briefly compare the ecological requirements of the parent species. R. amphibia colonizes open but rather sheltered shores and forms eventually mostly large clumps, pure or mixed up with other species, but at any rate a rather closed community. It nearly always grows in very wet places, most often with the basal parts submerged. R. sylvestris, when a shore plant, is bound to gravelly or sandy shores above the water level, and is much more susceptible to competition, not remaining long where the vegetation becomes dense. These two shore species have in fact very different ecological demands. B. JONSELL 107 To illustrate the establishment of hybrids I will take two examples from Sweden. At Lake Malaren (east central Sweden) the natural localities of R. sylvestris are the shores of a number of large eskers, formed by the retreating land-ice, that cross the lake and supply considerable stretches of rather exposed, open gravelly shores. On such places most hybrids are also to be found (cf. map in Jonsell, 1968, p. 117), sometimes a single clone only, sometimes real hybrid swarms, although the number of biotypes in one locality always seems to be rather restricted. This shows that only a very small proportion of the hybrid seeds produced gives rise to mature plants. Samples of wild hybrid seeds give rise in cultivation to many odd, but vegetatively vigorous, hybrid plants, the counterparts of which are never met with in nature (Jonsell, 1968, fig. 17, p. 119; cf. also Gorenflot, 1964, on Plantago ). The R. amphibiax sylvestris hybrids have, as expected, in varying degree the root-runners of R. sylvestris, and the more they have, the more they are capable of invading the special habitats of that species, while the R. amphibia type of habitat is not suitable for seedlings sensitive to competition. Most hybrids established are close to FiS between the species or to products of back-crosses to R. sylvestris, while those closer to R. amphibia seem to have greater difficulties in obtaining a foothold. Obviously hybrids can become established under the natural condi¬ tions supplied by certain shores at Lake Malaren. Human actions (clearings, etc.) especially formerly favoured the chances of the hybrids (e.g. swarms at small harbours outside the esker areas), but now mostly reduce them by allowing competing vegetation. The present conditions at Lake Malaren have not prevailed for much more than 1,000 years. Earlier that stretch of water was a bay of the Baltic, probably too saline for any of the parent species. It should be emphasized that these hybrids never behave as real weeds, and the same is true for the tetraploid R. sylvestris of the Malar region. Weed clones of that species are always hexaploid there, and do not invade the shores. Hexaploid, and to a large extent also tetraploid, R. sylvestiis consists of numerous clones which have apparently originated in connection with human cultivation, and they are, at least in N.W. Europe, neophytes when they grow on shores. In that position also the hexa- ploids may, when sandy or gravelly open shores are available, hybridize with R. amphibia, with vigorous and often rather fertile pentaploids as the result. An example will be taken from Scania (southernmost Sweden), a lake with sandy shores recently formed by lowering of the water level. One hexaploid R. sylvestris clone invaded the shores and crossed with the indigenous tetraploid R. amphibia forming a commonly occurring pentaploid hybrid, vigorous and highly fertile. In spite of 108 HYBRIDIZATION IN RORIPPA careful investigations no aneuploid back-cross products were found in that place, but spontaneous seeds from the pentaploids gave rise to aneuploid progeny (Fig. 4A), easy to raise and keep in cultivation, but obviously incapable of becoming established in nature. Spontaneous aneuploids are on the whole unknown in this clonal complex and unlike many others (e.g. Cardamine pratensis , Lovkvist, 1956; Achillea, Ehrendorfer, 1959) it seems to show an obstacle to fusion between the polyploidy levels. In this example human action has clearly been the condition for the establishment of the hybrids, but not even here do they spread as weeds. Situations ecologically similar to the Scania example, with rather intermediate pentaploids and or with tetraploid hybrid swarms, seem to prevail in N.W. Europe. The R. austriacax sylvestris hybrids have quite other ecological requirements. Like both parent species they have root-runners and they are as well as any of them able to inhabit dry places in the cultivated landscape (harbours, roadsides, railways, gardens, etc.). They are in N.W. Europe totally dependent on human cultivation and appear everywhere as weeds. They occur far outside the range of native and even introduced R. austriaca , forming independent long-lived clones, which have slight chances of seed-setting and thus contributing to the origin of new genotypes. They might exceptionally meet tetraploid R. sylvestris, but where new genotypes appear, fresh introductions of hybrid seeds as a rule seem to have occurred. In Sweden, seeds formerly arrived with cereals imported from the Black Sea region, where the hybrid is frequent (Jonsell, 1973), but other ways of intro¬ duction have certainly occurred (Jonsell, 1968, p. 143). Perhaps into Sweden, but in any case within the country, the root-runners are transported by soil, just as weed clones of R. sylvestris, and the hybrid is now in some places a prominent garden weed. In British herbarium material not a single specimen of this characteristic hybrid was found. INTROGRESSION The very variable tetraploid R. sylvestris shows in many areas features of R. amphibia in spite of the absence of intermediate plants, sometimes in single individuals, sometimes in whole local populations (e.g. in Britain, along the River Severn and River Wye in Monmouthshire). Introgression has obviously occurred on the tetraploid level via now extinct or very rare hybrids. To return to the area of Lake Malaren in Sweden, it is striking that the comparatively rare R. sylvestris is there very variable with many local, characteristic clones, while the much more numerous and widely distributed R. amphibia is highly uniform. B. JONSELL 109 R. sylvestris seems practically everywhere by this lake to be influenced by hybrid processes, while R. amphibia is largely left untouched. This may be explained by considering its nearly always much larger popula¬ tions, where a few introduced genes will be merely swamped, as well as the difficulties for seedlings to become established in them. R. palustris grows in the area, too, and it cannot be excluded (though it is very difficult to prove) that it has not in some degree contributed to the variability of R. sylvestris. No intermediates between them are known, but it is obvious that only those derivatives which have the effective root-runners of R. sylvestris, and consequently as a rule are close to that species in other respects as well, would have chances in the long run. CONCLUSIONS We have seen here that habitats free from competition are necessary for the establishment of R. amphibiax sylvestris hybrids, and that these conditions can be supplied under natural circumstances, but are more often the consequences of human action. The ecological restrictions of these hybrids, going back to the diversity in life history of the parent species, put limits on the hybrid process, the more so since these hybrids (but not R. austriacax sylvestris) seem incapable of fully utilizing the advantages supplied by man to become weeds. Although the hybrid nature, or at least the intermediate character, of all the actual forms has been evident from their discovery, they have taxonomically often been treated as species. This is still the case in Central Europe for ‘ R. prostrata (Berg.) Schinz & Thell.’, a name that stands for R. amphibiax sylvestris which is probably in the form of both pentaploids, comparatively clear-cut and often independent, and less clear-cut tetraploids. The local clones, similar from one river system to the other, give the impression of constituting a species, but no evidence to hand speaks in favour of such an interpretation. Broadly seen there are many counterparts in other genera to the hybrid situation met with in the yellow-flowered Rorippas, which fit a model with the combination of a clonal and a polyploidal complex as the significant point. The polyploidy has its consequences for the breakdown of genetic isolation, but it is also true that, of the two cytotypes of R. amphibia, the tetraploid has a much wider distribution to the north and east than the diploid. R. sylvestris s.str. is not known from below the tetraploid level, but a closely related form restricted to Hungary, R. kerneri Menyh. (= R. sylvestris ssp. kerneri (Menyh.) Soo) is reported to be diploid (Borsos, 1970), and there are other restricted taxa close to R. sylvestris in S.E. Europe and further 110 HYBRIDIZATION IN RORIPPA eastwards (Jonsell, 1973). The polyploids, which meet and hybridize, have probably to a large extent thanks to civilization reached their present ranges, at least in N. Europe. This means that Man’s influence may be behind it all, even where the hybrid process now seems to be going on under natural conditions. References Borsos, O. (1970). Contributions to the knowledge on the chromosome numbers of phanerogams growing in Hungary and South-Eastern Europe. Acta bot. hung. 16: 255-65. Ehrendorfer, F. (1959). Differentiation-hybridization cycles and polyploidy in Achillea. Cold Spring Harb. Symp. quant. Biol. 24: 141-52. Gorenflot, R. (1964). Introgression, polymorphisme et taxonomie chez les Plantaginacees. Adansonia (nouv. ser.) 4: 393^117. Howard, H. W. (1947). Chromosome Numbers of the British Species of the Genus Rorippa Scop, (part of the Genus Nasturtium R. Br.). Nature 159: 66-7. Hylander, N. (1960). Rorippa amphibia funnen vildvaxande i norra Oster- botten. Svensk bot. Tidskr. 54: 273-5. Hylander, N. (1961). Rorippa amphibiaxislandica funnen i norra Finland. Mem. Soc. Fauna Flora fenn. 37: 253-5. Javurkova-Kratochvilova, V. & Tomsovic, P. (1972). Chromosome study of the genus Rorippa Scop. em. Reichenbach in Czechoslovakia. Preslia 44: 140-56. Jonsell, B. (1968). Studies in the North-West European species of Rorippas.str. Symb. bot. upsal. 19 (2): 1-222. Jonsell, B. (1973). Taxonomy and Distribution of Rorippa (Cruciferae) in the Southern U.S.S.R. Svensk bot. Tidskr. 67: 281-302. Lovkvist, B. (1956). The Cardamine pratensis complex. Symb. bot. upsal. 14 (2). Meusel, H., Jager, E. & Weinert, E. (1965). Vergleichende Chorologie der zentral-europaischen Flora. Jena. WILD HYBRIDS IN THE BRITISH FLORA C. A. Stace ABSTRACT Hybrids have not received from British field-botanists enough of the precise and detailed attention they deserve. This is at least partly due to uncertainties regarding the number, characteristics and abundance of hybrid combinations which exist. Such information, as far as it is known, has now been made generally available in a book, Hybridization and the Flora of the British Isles, published in 1975. The text was prepared by 86 experts over a total period (from inception to publica¬ tion) of more than 6 years. This paper presents some statistics from the book relating to the number of interspecific hybrid combinations found in the British Isles. A total of 975 such hybrids are included, of which 626 are reckoned to be well substantiated, 122 possibly correctly recorded, and 227 erroneously recorded or probably so. Possible reasons for the varying abundance or patchy distribution of many hybrids are discussed, with particular reference to those found on the Continent of Europe but not in the British Isles. INTRODUCTION The study of hybrids and hybridization between distinct species of plants has a long, involved and often controversial history, the earlier phases of which are documented and discussed in two excellent textbooks (Roberts, 1929; Zirkle, 1935). Aspects which are even today still in contention include factors involved in the formation and estab¬ lishment of hybrids; the abundance, distribution and ecological impor¬ tance of hybrids in the wild; and the overall evolutionary significance of hybridization. In this paper I hope to present some facts which may go a little way towards answering just two of these points: how common are hybrids in the wild, and what factors determine their frequency? It is important to be able to answer these questions because they are basic to all the others; it is clearly not possible to estimate the importance or significance of a phenomenon if one does not know how common it is. At a more practical level the accurate recognition and [ ill 1 112 WILD HYBRIDS IN THE BRITISH FLORA recording of hybrids is essential to field-botanists, not only because of the intrinsic interest of hybrids, but also because such information will reveal the exact limits of the parents in terms of both structural variation and distribution. This is obviously a two-way process, because it is inevitably the field-botanists who are going to produce most of the primary data on variation and distribution. The absence of sufficient primary data has in the past often led to a subjective view of hybridization being adopted by botanists, who have in many cases become either ‘believers’ or ‘non-believers’. But, just as with those other two opposing groups, the ‘ splitters ’ and ‘lumpers ’, constant adherence to one dogma will prove to be as often misguided as judicious. It is perhaps not too obvious to emphasize that each case must be judged on its own particular evidence, and that the gathering of such evidence is an essential preliminary to sensible judgement. Nevertheless, the possibility that hybrids exist has often provided a sort of escape clause in taxonomically difficult groups. If two taxa are separable only with difficulty the presence of troublesome intermediates can all too easily be ascribed to hybridization. If little is known of the group, and experimental studies have not been carried out, it might be impossible to know whether or not this is the correct interpretation. There is a great number of records of hybrids in this category. Many of them have since been confirmed or refuted, but others remain uncertain. The existence of so much uncertainty, and the widely scattered sources of reliable information, much of which is unpublished, promp¬ ted the proposal by Professor D. H. Valentine and Dr S. M. Walters in 1968 (following a suggestion by the former in 1950), that a book should be compiled to act as a source of reference concerning all that is known about British wild hybrids. The scope of the book (Stace, 1975) is limited to vascular plant hybrids reported from the wild in the British Isles, and involving two or more taxa generally differing at the species level or above. These limitations are somewhat arbitrary, but a course was chosen as close to completeness as possible without setting an unrealistic target. The limitations also set the book apart from that all-embracing classic monograph on hybrids, Die Pflanzenmisch- linge (Focke, 1881), but of course the information now available for each hybrid is vastly more detailed and diverse than in 1881, and far more hybrids are now known. Hybridization and the Flora of the British Isles was published in 1975 (more than six years from its inception) under my editorship, but involving 86 different authors, each with specialist knowledge in one or more genera. The aim of the book is to lay out the evidence, both circumstantial and experimental, concerning every hybrid combination claimed to C. A. STACE 113 have been found in the British Isles. This, it is hoped, will prove of value to both amateur and professional botanists in a number of ways. In particular it should provide teachers with ideas for examples and projects, enable research workers to pinpoint queries and problems, and aid field-workers in identifying hybrids more accurately and in showing up gaps in our information which they might be able to fill. Perhaps the first purpose that the book has served has been to provide some of the basic data presented in the next few paragraphs. SOME STATISTICS Before analysing the number of hybrids which occur or have been recorded it is important to consider the raw material available for hybridization, i.e. the number of species in the British Isles. One can never obtain an exact number because no two authors are of precisely the same opinion concerning species limits, or concerning the criteria for the inclusion of alien species, but the figures given in a modern authoritative work (Dandy, 1958) give a good idea of the situation. Dandy listed 2, 179 native species, of which 610 were ‘ additional microspecies ’ in the genera Rubus and Hieracium, plus 643 naturalized alien species. One should probably add about 180 native microspecies to account for species of Rubus and Hieracium since recognized, and for the micro¬ species of Taraxacum, which were not listed by Dandy. But the total figure of 3,002 (which many workers would wish to increase substan¬ tially by the addition of further naturalized species) means little by itself, because the majority of the microspecies are agamospecies unable to hybridize, and even the figure of 2,212 for the remainder does not take into account their degree of relationship. The data in Table 1, which lists the number of species in the larger genera, are therefore relevant. They show that 35 genera contain ten or more species, but that only seven contain more than 21 species and that three of these are the large agamospermous genera previously mentioned. It should be emphasized that these figures are not taken directly from Dandy (1958), but are updated and include native species only. Table 2 indicates that of 720 British genera almost exactly half are monotypic in the British Isles, and that almost exactly half of the remainder form no hybrids solely involving British species either in the British Isles or elsewhere. Of the remaining genera which do form hybrids, 143 form hybrids in the British Isles and a further 38 form hybrids solely involving British species but only abroad. Of course, many of the monotypic genera and of the polytypic genera which are listed as forming no hybrids do form hybrids abroad involving one or more non-British species. Good examples are Pulsatilla, Elymus and 114 WILD HYBRIDS IN THE BRITISH FLORA Table 1. Numbers of native British species in the larger genera of the British flora Rubus c. 400 Juncus 26 Hieracium c. 250 Euphrasia 26 Taraxacum c. 140 Ranunculus 24 Car ex 80 28 others 10-21 Table 2. Numbers of genera in the British flora analysed according to their propensity for hybridization of which 720 366 genera are monotypic of which 354 173 form no hybrids of which 181 38 form hybrids abroad leaving 143 which form hybrids in the British Isles Pedicularis. The figures 720 and 366 are taken direct from Dandy (1958), but the other numbers have been gathered from Hybridization and the Flora of the British Isles. In addition there are 14 confirmed intergeneric hybrid combinations in the British Isles, involving both monotypic and polytypic genera, and in the latter case both those forming other hybrids and those not. These include five in the Orchidaceae and five in the Gramineae, and are listed in Table 3. Apart from many other erroneous or doubtful claims of intergeneric hybrids in the British Isles there are records of 22 additional intergeneric hybrid combinations abroad solely involving British spe¬ cies (e.g. in DactylorhizaxPseudorchis ), and a further 11 combinations of British genera where one or both of the species involved are non-British (e.g. in CotoneasterxSorbus ). Naturally, whether a hybrid is intergeneric or not depends entirely upon one’s concept of the limits of the genera concerned, and it is true that several of the combinations in Table 3 would not be considered intergeneric by some workers. The generic limits used by Dandy (1958) have been employed for the present purpose. Table 4 gives the total number (975) of interspecific (including intergeneric) hybrids recorded from the British Isles, and attempts to C. A. STACE 115 Table 3. Intergeneric hybrid combinations in the British flora and the number of specific combinations in each. Data taken from Stace (1975) Nos. of hybrids 3 1 1 1 1 3 1 4 1 6 4 1 1 1 29 Table 4. Numbers of interspecific hybrids recorded in the British flora. Data taken from Stace ( 1975) 626 confirmed or probably correct 122 possibly correct 227 erroneously recorded or probably so 975 464 recorded only abroad 1439 Asplenium x Phyllitis Crataegus x Mespilus Conyza x Erigeron Anthemisx Tripleurospermum Coeloglossumx Gymnadenia Coeloglossum x Dactylorhiza Gymnadenia x Pseudorchis Dactylorhizax Gymnadenia Anacamptisx Gymnadenia Festuca x Lolium Festucax Vulpia Agropyron x Hordeum AmmophilaxCalamagrostis Agrostis x Polypogon classify them into three groups according to the degree of certainty of the record. It seems likely that at least 25 per cent of the combinations are erroneously recorded. In addition there are 464 hybrids between British species which have been found abroad but not in the British Isles. These figures are taken from the new hybrid book (Stace, 1975), and easily exceed the number of hybrids mentioned in any other British work (see Stace, 1975, p. 22). However, the figure of 464 for the ‘foreign hybrids’ should be used with much caution, for it does not pretend to be complete, and it includes a proportion of records which will almost certainly turn out to be ill-based. Moreover the above figures do not include hybrids between the microspecies of Rubus. In this genus 5 UBG 116 WILD HYBRIDS IN THE BRITISH FLORA Table 5. Numbers of hybrids in the British flora in the eight most hybridogenous genera. Data taken from Stace ( 1975) Salix 59 Carex 34 Euphrasia 54 Rumex 29 Epilobium 48 Potamogeton 25 Rosa 37 Dactylorhiza 19 there are many more or less sterile hybrids known in the British Isles and elsewhere, but in addition it is likely that a very high (and quite unknown) proportion of the generally recognized microspecies are in fact hybridogenous, having become stabilized by agamospermy. Finally, the book also includes 23 intraspecific hybrids which for various reasons are given as full a treatment as the interspecific ones. Table 5 shows the distribution of hybrids in eight genera. These possess from 19 to 59 confirmed hybrids ; no other genus possesses more than 13, apart from the special case of Rubus mentioned above. Of the eight, Epilobium and Dactylorhiza have by far the largest number of hybrids in relation to the number of species, for almost every parental combination has been found. At the other end of the spectrum one can list ten rather large genera which form no hybrids at all: Trifolium, Vicia, Lathyrus, Alchemilla, Sedum , Oenanthe, Orobanche, Cam¬ panula, Hieracium sensu stricto and Allium. The absence of hybrids in Alchemilla and Hieracium can of course be explained on the basis of agamospermy, but in the other cases it is probable that strong interspecific incompatibility barriers do really exist, and in several of them (e.g. the three Leguminosae and Allium ) this has been demon¬ strated experimentally. It is possible that Polygonum should be added to the above list, for none of its many reported hybrids has been satisfactorily proven. IDENTIFICATION AND DISTRIBUTION OF HYBRIDS It is clear from the above data that, even if the erroneous or doubtful records are ignored, a great many hybrid combinations are to be found in the British Isles. The distribution and frequency of these hybrids are, however, far less easily ascertained. Undoubtedly many hybrids are rare, but many are not and a good number are greatly under-recorded; it is to be hoped that more intensive work by field-botanists will help to fill the many gaps in our information. Perring & Sell (1968) mapped about 50 hybrids in their Critical Supplement to the Atlas of the British Flora, including a number of more frequent ones and others which are C. A. STACE 117 easily recognized or for which reliable information had already been accumulated by specialists. That this is only about 8 per cent or perhaps less of the total number is a good illustration of the imperfect state of our knowledge. Detailed recording of hybrids in the field naturally relies upon accurate identification, the lack of which in the past has resulted in a very large number of erroneous records. Most of these should have been referred to extreme or abnormal states of one (or both) putative parents, or to a third species. Species which have at one time been wrongly attributed to a hybrid status often bear the specific name of hybridus or intermedius , etc. It is not possible here to provide detailed hints on the determination of hybrids, but three general points are perhaps not out of place. (a) While most hybrids are sterile to a lesser or greater degree, sterility by itself is not a very good criterion. Many hybrids are highly or even completely fertile, and in addition hybrids which are in reality largely sterile may appear to have perfectly viable pollen (e.g. most Juncus hybrids). This may be due to the facts that sterility can exist mainly on the female side, or can take effect only after pollination. Moreover, the lack of good pollen and/or seed can be caused by many factors other than hybridity , for instance the lack of a compatible pollen source (e.g. Calystegia spp.), or environmental effects (as is common in Juncus inflexus ). ( b ) Hybrids are not always exactly or even roughly intermediate in appearance between their parents. Often some of the characters are intermediate, while others (perhaps the most conspicuous) are close to one or other parent. In other cases a hybrid may much more closely resemble one parent than the other in all characters (e.g. Geranium purpureumxG. robertianum), or may exhibit some features not found in either parent (e.g. Papaver dubiumxP. rhoeas). (c) It is dangerous to extrapolate evidence obtained from the study of one hybrid to other closely similar or related hybrids. Often, of course, a series of hybrids among a well-defined group of species has many features in common, but it is not rare to find the opposite situation. For example, the three hybrids known to involve Juncus effusus are, respectively, fully fertile (xj. conglomeratus ), slightly fertile (xj. inflexus ), or completely sterile (xj. balticus ). There are also often many surprises concerning the distribution of hybrids. In particular there are numerous instances where hybrids do not occur wherever the two parents co-exist, even though the latter might be highly compatible. The reasons for this are very various and in general well documented by several reviewers, and need not be gone into here. They can be placed into two main categories: 5-2 118 WILD HYBRIDS IN THE BRITISH FLORA (a) The isolating barriers normally separating the two species break down in some places only. This is an obvious possibility in the case of external barriers, since many aspects of the environment vary from place to place, but is no less true of internal barriers. It is a fairly common experience to find that a hybrid not previously synthesized despite numerous attempts has been readily obtained with a new, different genetic stock of one or both parents, or vice versa. ( b ) Hybrids can become established in certain areas, but not in others. This is often the result of disturbance of the habitat by man, who unwittingly creates new niches suitable for hybrids. Of greatest interest to British botanists are those hybrids which have been found in the British Isles but not elsewhere (e.g. Daphne laureola xD. mezereum , various Festucax Vulpia combinations), or vice versa; 464 of the latter category are mentioned in the new hybrid book. Some of these records are probably erroneous, and others represent combinations unlikely to occur in the British Isles because the parents do not come into contact (e.g. Juncus balticusxj. fili- formis ) or because of the rarity of one or both parents (e.g. Aceras anthr op ophorumx Orchis simia , Woodsia alpinaxW. ilvensis). But a detailed search in this country for the others could well be rewarding; it would determine the precise pattern of hybrid distribution compared to that of the overlap of the parents, and also help to assess the factors isolating the parents as well as those causing breakdown of isolation. The discovery of several hybrids new to the British Isles (or to science) in the past few years shows that the British list is by no means complete. Among the many hybrids known abroad and which might be found in the British Isles the following 15 have been selected from genera (or generic combinations) which have so far no known hybrids in this country: Asplenium ruta-murariaxCeterach officinarum Cystopteris fragilisxC. montana Gymnocarpium dryopterisxG. robertianum Dianthus armeriaxD. deltoides Scleranthus annuusxS. perennis Pimpinella majorxP. saxifraga Pyrola minorxP. rotundifolia Cynoglossum germanicumxC. officinale Sambucus nigraxS. racemosa Aceras anthr op ophorumx Herminium monorchis Dactylorhiza fuchsiix Pseudorchis albida Aceras anthrop ophorumx Orchis purpurea Dactylis glomerataxD. polygama C. A. STACE 119 Melica nutansxM. uniflora Phleum bertoloniixP. pratense JUNCUS BALTICUS HYBRIDS It is likely, however, that even after quite detailed analyses the explanations of many odd patterns of distribution will remain unan¬ swered. It is often very difficult to pinpoint a particular factor involved in species isolation, and frequently it is a complex interaction of a number of isolating factors which is operating. I want to illustrate such difficulties by describing a situation in Juncus subgenus Genuini con¬ cerning certain hybrids involving Juncus balticus. Juncus balticus hybridizes with J. arcticus in arctic and subarctic areas where the two species meet. The two species are closely related and several authors consider them to represent subspecies of a single species. In much of north-western Europe, particularly around the coasts of the Baltic, J. balticus meets J. filiformis , and these two species frequently hybridize in such situations. As mentioned above, their ranges do not overlap in the British Isles. The other two hybrids of J. balticus are far less widespread, there being three valid records of each. J. balticusxj. effusus has been found in two localities on the Lancashire coastal dune-systems, and one in a similar situation on the Baltic coast of East Germany, near Ribnitz (type locality for J.xobotritorum Rothm.). Both of the Lancashire localities for this hybrid have been eradicated by building develop¬ ments, one in 1968 and the other in 1974. The few other records for this hybrid are errors for J. balticusxj. filiformis , including the type of J.xscalovicus Aschers. & Graebn., which was originally attributed by Ascherson & Graebner (1893) to J. balticusxj. effusus. Finally, J. balticusxj. inflexus occurs in three localities on the Lancashire coastal dune-systems, two of them fortunately within the boundaries of nature reserves. There are morphological and/or anatomical differences between each of these six colonies of hybrid Juncus, all of which are completely sterile, and it seems likely that each of them is the result of a separate hybridization, five of them within a 27 km stretch of the coast of Lancashire (see map in Stace, 1972). Fig. 1 shows the extent of the overlap of the ranges of J. balticus, J. effusus and J. inflexus, covering a considerable band of terrain mainly on or near Baltic and North Sea coasts, as well as the two areas in which hybrids are known to occur. The British area (Lancashire) coincides with an outlying locality for J. balticus (separated from others by c. 300 km), but the German area is in a part of the range where J. balticus is very common, albeit near its southern limit. 120 WILD HYBRIDS IN THE BRITISH FLORA Fig. 1. Approximate northern limits of Juncus effusus and Juncus inflexus and approximate southern limit of Juncus balticus in Europe. The dots indicate the two localities of the six hybrid populations mentioned in the text. Table 6. Chromosome numbers of J. balticus, J. effusus, J. inflexus and their hybrids determined in the present study J. balticus J. effusus J. inflexus J. balticusxj. effusus ‘Ainsdale’ J. balticusxj. effusus ‘Hightown’ J. balticusxj. inflexus ‘Fylde’ J. balticusxj. inflexus ‘Birkdale’ J. balticusxj. inflexus ‘Freshfield’ J. effusus xj. inflexus 2 n = 84 2 n = 40 2 n = 42 2 n = 82 2 n = 80-82 2 n = 82-84 2 n = 84 In = 84 2 n = 42 C. A. STACE 121 The five British colonies have been examined cytologically (Table 6), and show that the J. effusus or J. inflexus parent contributed in each case an unreduced gamete. The diffuse-centric, very small (mostly 0-3-0-5 /im long) chromosomes of Juncus subgenus Genuini are very difficult to count with complete accuracy, and it is still not certain whether any species possesses more than one chromosome number, but the results obtained clearly indicate that J. balticus and both its hybrids have roughly twice as many chromosomes as the other two species and the hybrid between them. Many attempts have been made to re¬ synthesize these hybrids, using several different parental strains, includ¬ ing those from the Lancashire coast, but no hybrids have so far been produced. All three species are self-compatible and wind-pollinated, and usually set seed freely. One not infrequently encounters sterile colonies of /. inflexus (which in the following year may produce plenty of good seed), and the same has been reported for J. balticus, but in the years I have been studying them the Lancashire clones of J. balticus have always formed abundant viable seed both in situ and in the botanic garden. DISCUSSION The occurrence of five of the six known colonies of these hybrids within 27 km of each other, despite the abundant cohabitation of the parents over a wide area of northern Europe, is both intriguing and perplexing, and leads one to question what factors so conducive to hybridization might exist in Lancashire yet not elsewhere. There are many other areas known where hybridization is a no¬ toriously frequent phenomenon, but these generally involve hybrids over a wide taxonomic spectrum and can be related to disturbance of the habitat and to the influx of new (often weedy) species. New Zealand and Hawaii are two well-known examples. In New Zealand forest fires and clearance by man have been suggested as the major factors (see Cockayne & Allan, 1934), while in Hawaii hybridization has followed the colonization of open volcanic terrain by immigrant species (see Gillett, 1972). Briggs & Walters (1969, p. 197) have suggested that introgression, first detected and most studied in North America, where much of the present-day vegetation is the result of man’s very recent activities, might in fact be more prevalent there than in Europe, where the vegetation has had a greater chance to reach an equilibrium with man’s requirements over several millennia. In the Orkneys Miss E. R. Bullard (the B.S.B.I. recorder for those islands) has informed me (in litt. 1971) that the numbers of hybrid plants and the ‘ number of species 122 WILD HYBRIDS IN THE BRITISH FLORA involved seems to be exceptional so much so that in some genera, e.g. Senecio pro parte and Euphrasia, hybrids are more frequently encoun¬ tered than species. This is probably once again a measure of the large amount of open ground and high proportion of weedy species in the Orkneys. There are many genera in the British Isles where hybridization has occurred between native and alien species, e.g. Heracleum , Calys- tegia, Linaria, Senecio, Tragopogon, Endymion, Juncus and Spartina, besides others where it is also widespread among native species, and, if in these genera species have evolved allopatrically without the formation of sterility barriers, this is not surprising. But none of these situations seems to resemble that concerning the Lancashire Juncus hybrids, and indeed the only apparently close parallel I can find is that described in Equisetum subgenus Equisetum by Page (1973). There are four hybrid combinations in this subgenus in the British Isles, one of which ( E . arvensexE. fluviatile ) is widespread. The other three are all confined in the British Isles to the Hebrides (where E. arvensexE. fluviatile also occurs): E. arvensex E. palustre occurs on the Isle of Skye and is not known elsewhere; E. palustrexE. telmateia occurs on Skye and in two Mediterranean localities (Gerona, Spain and Alpes Maritimes, France); and E. fluviatile xE. palustre occurs on the Isle of Harris and is again not known elsewhere. Page considers that all the hybrids have arisen indepen¬ dently in the Hebrides, and suggests that the damp, oceanic climate and the exposure of bare mud by digging of ditches (both of which would favour the growth of prothalli) have been instrumental in their formation. In Juncus there is a good range of factors tending to prevent hybridization. Besides the slightly different geographical distributions the ecological preferences of all the species differ, although J. balticus, J. effusus and J. inflexus are all quite common in coastal dune-slack communities, in which the hybrids occur. There are also differences in the flowering time, although these are not absolute, and in chromo¬ some number. There is evidently a fairly strong interspecific incompati¬ bility system as well, for no artificial hybrids have been produced; foreign pollen germinates successfully but the pollen tubes cease growth before they reach the ovules. It is very difficult to imagine special features of the Lancashire coastal climate which might affect hybridi¬ zation potential. There are perhaps four suggestions that one might make in trying to throw some light on the situation: (a) The chromosome numbers of the five Lancashire hybrids, each indicating non-reduction in the parent other than J. balticus, are remarkable. But if this is other than a coincidence one still needs to explain the cause of non-reduction. Juncus pollen is dispersed in C. A. STACE 123 tetrads, so that non-reduced pollen (diads) should be easily recognized. None has been seen in any species, although it is possible that non-reduction takes place on the female side, when /. balticus would have been the male parent of the hybrids. {b) The Lancashire /. balticus , discovered in 1913, is 300 km re¬ moved from the next nearest colony (in eastern Scotland), and its origin is obscure. No differences in morphology, anatomy, chromosome number or compatibility with other species between Lancashire, Scot¬ tish and Scandinavian material of J. balticus have been detected, but the possibility remains that physiological differences do exist which have contributed to the formation of hybrids in Lancashire. Although there is no evidence that J. balticus was ever really common even locally in Lancashire, one colony is known to have been eradicated within the last 30 years (Stace, 1972), and it is thus possible that a distinctive Lancashire strain of the species has become extinct. It is quite conceivable that the Lancashire J. balticus has a very distant origin, for in recent years two other species of Juncus have become well established in the British Isles far from their native areas of distribution: J. subulatus from southern Europe, in the Bristol Channel (Willis & Davies, 1960), and J. planifolius from Australasia and South America, in W. Galway, Eire (Scanned, 1973). (c) It is possible that the scarcity of hybrids is due to the lack of conditions suitable for hybrid establishment, but that such conditions are or were in some way provided on the Lancashire coast. Seedling establishment in subgenus Genuini is almost certainly dependent upon open conditions (Lazenby, 1955), and is then often very successful. In closed vegetation the plants probably increase solely by vegetative spread, and all the three species considered here can be strongly invasive. Of the five Lancashire hybrid colonies four are (or were) of considerable size and occur among dense dune-slack vegetation, into which they are spreading, but the fifth was only about 6 in. in diameter when it was discovered in 1966 in bare sand in the damp hollow of a sand dune. It would be expected, however, that many other areas in the band of overlap between the three species would offer similar opportunities for seedling establishment. (d) There remains the possibility that the intensive field-work carried out by many botanists in Lancashire has led to the discovery of all the hybrid sites, whereas those elsewhere remain undetected. I do not consider this very likely, as there are many diligent and experienced Scottish and Scandinavian botanists who are very active field-workers, and German botanists have been aware of the possibility of hybrids involving J. balticus since the last century. /. balticusxj. effusus resembles and could be mistaken for J. balticusxj. filiformis, which 124 WILD HYBRIDS IN THE BRITISH FLORA is common around the Baltic Sea coasts, but J. balticusxj. inflexus is a very distinctive plant less likely to have been overlooked. One is therefore forced to the conclusion that no single known factor is implicated in the uneven and unexpected distribution of these Juncus hybrids. It is more likely that an interaction of a variety of factors is responsible for the present situation, and probably the balance is quite fine. In such cases it is usually impossible to predict the abundance of hybrids, even when experiments have provided a good deal of basic information, and it is surely likely that the Juncus example can be paralleled in many other groups. For this reason the continued intensive and extensive observation and recording by field-workers, coupled with experimentation by biosystematists, is essential. It is certain that close co-operation by amateur and professional plant taxonomists, so long a notable feature of the Botanical Society of the British Isles, will continue to be as fruitful as ever. ACKNOWLEDGMENTS I am grateful to Mr J. W. Grimes for much assistance in the investiga¬ tions of hybridization in Juncus, and to the Science Research Council for providing a Research Grant. References Ascherson, P. & Graebner, P. (1893). Beitrage zur Kenntniss der norddeutschen Flora. Ber. dt bot. Ges. 11: 516-30. Briggs, D. & Walters, S. M. (1969). Plant Variation and Evolution. London: Weidenfeld and Nicolson. Cockayne, L. & Allan, H. H. (1934). An annotated list of groups of wild hybrids in the New Zealand flora. Ann. Bot. 48: 1-55. Dandy, J. E. (1958). List of British Vascular Plants. London: Botanical Society of the British Isles. Focke, W. O. (1881). Die Pflanzenmischlinge. Ein Beitrag zur Biologie der Gewachse. Berlin: Borntraeger. Gillett, G. W. (1972). The role of hybridization in the evolution of the Hawaiian flora. In D. H. Valentine, ed., Taxonomy, Phytogeography and Evolution, pp. 205-19. London, New York: Academic Press. Lazenby, A. (1955). Germination and establishment of Juncus effusus L. The effect of different companion species and of variation in soil and fertility conditions. J. Ecol. 48: 103-19. Page, C. N. (1973). Two hybrids in Equisetum new to the British flora. Wat sonia 9: 229-37. Perring, F. H. & Sell, P. D., eds. (1968). Critical Supplement to the Atlas of the British Flora. London: Nelson. Roberts, H. F. (1929). Plant Hybridization before Mendel. Princeton: Princeton University Press. Scannell, M. J. P. (1973). Juncus planifolius R. Br. in Ireland. Ir. Nat. J. 17: 308-9. C. A. STACE 125 Stace, C. A. (1972). The history and occurrence in Britain of hybrids in Juncus subgenus Genuini. Watsonia9 : 1-11. Stace, C. A., ed. (1975). Hybridization and the Flora of the British Isles. London, New York: Academic Press. Willis, A. J. & Davies, E. W. (1960). Juncus subulatus Forsk. in the British Isles. Watsonia 4: 211-17. Zirkle, C. (1935). The Beginnings of Plant Hybridization. Philadelphia: University of Pennsylvania Press. TAXONOMIC PROBLEMS IN THE FERN GENUS POLYSTICHUM CAUSED BY HYBRIDIZATION G. Vida and T. Reichstein ABSTRACT The genus Polystichum in Europe consists of two diploid species, P. lonchitis (LL) and P. setiferum (SS). A third allotetraploid species, P. aculeatum (LLSS) has long been formed between these diploids. The meiotic division, spore fertility and the progenies of the triploid P. x illyricum (LLS) were studied in order to demonstrate the degree of genetic isolation between P. lonchitis and P. aculeatum. P.x illyricum exhibits at meiosis c. 41 bivalents and c. 41 univalents. Difficulties with the distribution of the univalents cause a high degree of spore abortion. A limited number of progeny, however, could be obtained by dense spore sowings. These plants were triploid or hexa- ploid (autoallohexaploid: LLLLSS). Analyses of meiosis and spore- formation in the triploid F2 demonstrate the probable way of origin of the F2 plants. The types of spore-formation in these triploid hybrids are discussed in connection with the possibility of gene exchange among the allotetraploid and its two diploid progenitors via triploid hybrids. INTRODUCTION The genus Polystichum in Europe is represented by two diploid and two tetraploid species as follows: P. lonchitis (L.) Roth, 2x; P. setiferum (Forskal) Woynar, 2x ; P. aculeatum (L.) Roth, 4x; P. braunii (Spenner) Fee, 4x (x = 41). The first three are native to the British Isles. Hybrids in every combination have already been found (Meyer, 1960; Sleep & Reich¬ stein, 1967; Vida, 1966) and synthesized (Sleep, 1966). According to the results of genome-analysis of these species (Manton, 1950; Manton & Reichstein, 1961; Sleep, 1966; Vida, 1966, 1972) the origin of the tetraploid P. aculeatum could be explained by the following scheme: [ 126] G. VIDA ANDT. REICHSTEIN 127 P. lonch LL P. x lonchitiforme ^ LS P. setiferum SS > 1 P. x illyricum Chromosome doubling T P. x bicknellii LSS LLS Y P. aculeatum LLSS L = P. lonchitis genome (41 chromosomes) S = P. setiferum genome (41 chromosomes) P.x illyricum (Borbas) Hahne and P.x bicknellii (Christ) Hahne, both triploid hybrids, have been reported in many places where P. aculeatum grows together with P. lonchitis or P. setiferum (also in the British Isles: see Sleep, 1971; Sleep & Synnott, 1972). They can be regarded as back-crosses to one of the two diploid progenitors. At meiosis, therefore, they usually form bivalents between the chromo¬ somes of the similar genomes (41 bivalents), while the remaining third genome is represented by unpaired chromosomes (41 univalents). Morphologically (as well as ecologically) the two diploid species P. lonchitis and P. setiferum are very easy to distinguish. The alpine-boreal P. lonchitis has persistent, coriaceous, linear-lanceolate fronds with undivided pinnae. The chiefly Mediterranean P. setiferum , on the other hand, develops soft, sometimes not persistent, lanceolate fronds with clearly divided pinnae. Keeping in mind the above-illustrated origin of the tetraploid P. aculeatum , it is not surprising that this species is intermediate between its two diploid progenitors in almost every respect. This fact is reflected in some earlier taxonomic treatments of the group, giving subspecific rank to P. setiferum under P. aculeatum. Nevertheless, P. lonchitis , P. aculeatum and P. setiferum are distinct species which are not too difficult to identify provided the plant is not juvenile or stunted. The triploid hybrids, however, make this morpho¬ logical distinctiveness almost completely continuous, since the range of variation of each taxon (incl. hybrids) overlaps to a large extent (Fig. 1). The taxonomic difficulty is further increased by an interesting obser¬ vation, namely, that at some places, where two Polystichum species (P. lonchitis and aculeatum or P. setiferum and aculeatum) have been growing together, presumably for many generations, there is an in¬ creased morphological similarity between the two cytologically different 128 TAXONOMIC PROBLEMS IN POLYSTICHUM species, which could also be expressed in numerical terms (Vida ined.). There are two alternative explanations for this phenomenon. One can argue that the morphological difference between two Polystichum species is a product of the very complex natural selection acting differently under dissimilar ecological conditions of these species. Accordingly, natural selection in the ecologically intermediate mixed populations favours genotypes approaching an intermediate morpho¬ logy in both species. The other alternative possibility would be genetic introgression via hybridization and back-crosses, which is much more effective in achieving rapid genetic alterations, provided there is no strong reproductive barrier between the species concerned. In order to test this second hypothesis we started to study the meiosis, spore formation, fertility and progenies of the two triploid hybrids ( P.xillyricum and P.xbicknellii). MATERIALS AND METHODS (a) Source of the wild hybrids Wild triploid hybrids were cultivated in Basel or Budapest of the following origin: P. x illyricum TR-255 Switzerland, Kanton Schwyz, Scharsack, SE from Hag- genegg, Mythen. Limestone, c. 1390 m. leg. TR. 27 July 1960. Cultivated in Basel (TR). TR-1767 Switzerland, Herrenrliti near Engelberg, leg. Dr W. Gatzi 1959. Cultivated in Basel (TR). TR-1768 Switzerland, Untertrlibsee near Engelberg. Along the way to the Gentialp, leg. Dr W. Gatzi 1959. Cultivated in Basel (TR). GV-Pol.~8 Poland, Tatry: Giewont: limestone scree under Picea forest. Altitude 1400 m, leg. GV. 14 September 1962. Cultivated in Budapest (GV). P. x bicknellii GV-Pol. la Hungary: Mecsek-Hills: Szuado Valley near Orfii, leg. GV. August 1961. Cultivated in Budapest (GV). GV-Pos. 2 b Hungary: Bakony Hills: Toth-arok Valley near Fenyofo, leg. GV. August 1961. Cultivated in Budapest (GV). G. VIDA AND T. REICHSTEIN 129 (b) Cytology Cytological investigations were carried out using the standard carmine squash method (Manton, 1950) for meiosis and sporangial development. Root trips were previously treated with colchicine (01 per cent for 3-5 h at 4 °C) followed by fixation in acetic-alcohol (3:1) and 5 h soften¬ ing in a snail enzyme solution according to Roy & Manton (1965). (c) Progeny yield In order to avoid foreign spores in testing spore viability of the triploid hybrids, great care was taken to preserve the purity of spores and cultures. In Basel fronds with ripe spores (TR-1767, 1768, 2051) were collected from plants grown in the experimental garden, washed under running tap water and pressed to release the spores between two sheets of clean white paper. The spores were sown on a mineral solution solidified with 1 per cent agar and the prothalli when c. 1*5 mm long transplated on sterilized soil in pots covered with transparent plastic cups. In Budapest the experiments started much later, taking into account the previous results at Basel. Pots with large hybrid plants (GV-Pol. 8, Pol. 1 a, Pos. 2 b) have been kept separately in a greenhouse where there was no other Polystichum around. In spring as the new fronds started to unroll, they were covered with a paper bag, firmly sealed at the petiole. We fixed the bag in a position to allow the frond to develop inside the bag. One side of the paper bag was replaced by cellophane in order to make photosynthesis (and observations) possible. When the spores were ripe, the whole frond with the bag was cut and the sowing was subsequently made on agar just as in Basel. RESULTS (a) Cytology of the wild hybrids The natural hybrids Rxillyricum (TR-255, 1767, 1768; GV-Pol. 8 ) and P.xbicknellii (GV-Pol. la, Pos. 2b) were all studied cytologically. Each one was found to be triploid with 2 n = 123 and n = c. 41n + 41j at meiosis. Representative stages of the meiosis and spore formation are illustrated with P.xillyricum in Figs. 2-7 (Plate I). Large numbers of univalents disturb the meiotic division from the anaphase I onwards. These chromosomes seem to hesitate at metaphase I in the equatorial plate and either move to one of the poles at random or divide by distributing chromatids to the opposite poles. In this latter case the second anaphase exhibits further irregularities, because these chroma- 130 TAXONOMIC PROBLEMS IN POLYSTICHUM Fig. 1. Pinnae of Polystichum taxa taken from the middle of the frond. Genomic symbols: LL = P. lonchitis; LLS = P.xillyricum; LLSS = P. aculeatum; LSS = P.xbicknellii; S S = P. setiferum. Each taxon is illustrated by two extreme forms to show the overlapping characters. Ploidy (2 n) Spore Number of Number of - Culture Data of sowing germination prothalli sporophytes 2x 3x 4x 6x Unknown 131 * (N r}- ’-h m ON 3 O C/3 3 PQ r- NO ON > ON 0) C/3 3 CQ to £ D ON a ^ 3 J 73 *2 3 PQ m TD C3 C" Z* O £ O £ o a u a s-h X 0) X t"- 73 > 5 ^ >o PQ tN CQ n oo NO t" <3 00 • »««* X > > O x o a' O - o o a * f The diploid and tetraploid plants were probably contaminants of nearby Polystichum setiferum and aculeatum cultures. 132 TAXONOMIC PROBLEMS IN POLYSTICHUM tids cannot split again. In any case, the resulting tetrad consists of four nuclei (sometimes also a few micronuclei) with unbalanced chromo¬ some sets, which causes an early abortion of the spores (Fig. 7). In some sporangia, however, sporadic diad formation was also observed among the tetrads (Figs. 5,6). These are of great importance, for potentially they can develop into viable spores, provided that all the 41 univalents behaved uniformly prior to diad formation. ( b ) Fertility and progeny Dense spore sowings were made from each hybrid plant with the results given in Table 1. It is rather surprising that each plant produced an unexpectedly large number of prothalli. These prothalli, however, developed very slowly and many of them showed abnormal morphology. So far, only the earlier spore sowings (Basel) were able to develop a few sporophytes, but these were very interesting. Two cultures of P. x illyricum (TR-1767 and 1768) produced only 6x plants (Figs. 9, 13, 14) growing very slowly but eventually giving rise to a more or less fertile autoallohexaploid (genome symbol: LLLLSS) with morphology of P.x illyricum. On the other hand, TR-255B gave mainly triploids (Figs. 8, 10-12) showing vigorous growth but a striking variation in morphology when mature, two plants imitating P. aculeatum (except for bad spores), while two others looked like normal big P.x illyricum. The meiosis in these triploid progenies seems to be similar to the wild mother plant (pre¬ sumedly Fj), but their fertility appears to be slightly increased, which is illustrated by the greater number of healthy spores in the developing sporangia (Figs. 11, 14). Here again, diad formation was observed in many sporangia. The result of P.xbicknellii progeny (GV-Pol. la, Pos. 2b) at this stage merely confirms the fact shown by P.x illyricum that triploid hybrids in this combination are also able to produce F2 progenies. Nothing is known, however, about the chromosome number of the prothalli, although they show markedly different cell sizes, which could be connected with different ploidy levels. DISCUSSION The ability of interspecific triploid hybrids to produce functioning gametophytes (and consequently sporophytes) has not been seriously considered in ferns.* This is not infrequent, however, in flowering * Intraspecific triploid hybrids of Osmunda regalis have been studied in this respect by Manton (1950). Figs. 2-7 (Plate I). Cytology of wild P.xillyricum (TR-1767 and 1768a). Fig. 2. Root tip mitosis with 2 n = 123 chromosomes (TR-1768a). xlOOO. Fig. 3. Meiosis (dikinesis) showing 41 bivalents and 41 univalents (TR-1768a). x 1000. Fig. 4. Meiotic telophase I (TR-1767) showing lagging univalents partly forming micronuclei between the opposite poles, x 1000. Fig. 5. The results of meiosis in TR-1767. Tetrads, diads and intermediate forms, x 600. Fig. 6. Content of a young sporangium after meiosis (TR-1768a) with some diads. x 300. Fig. 7. Entirely aborted spores in a sporangium of TR-1768a. x 300. Figs. 8-14 (Plate II). Cytology of the progenies of wild P.xillyricum. Fig. 8. Root tip mitosis in a triploid progeny (TR-255B-pr.4). In = c. 123. x 1000. Fig. 9. Root tip mitosis in a hexaploid progeny (TR-1767-pr.l). 2 n = 246. x 1000. Fig. 10. Spore mother cells at the first meiotic metaphase in a sporangium of a triploid progeny (TR-255B-pr.2). x 300. Fig. 11. ‘Good looking’ and aborted spores in a sporangium of the same plant (as Fig. 10). x 300. Fig. 12. Aborted tetrad-spores and apparently viable diad-spores in a triploid progeny (TR-255B-pr.6). x 300. Fig. 13. Normal, viable spores in a hexaploid progeny (TR-255B-pr.7). x 300. Fig. 14. Viable tetrad-spores of a hexaploid progeny (TR-1767-pr.3). x 300. G. VIDA AND T. REICHSTEIN 133 plants, where in some cases it may give rise to back-cfosses as well. This can cause unilateral intr ogres sion from the diploid to the tetraploid species (e.g. in Dactylis - Zohary & Nur, 1959). The degree of fertility of triploid flowering plants is usually markedly different on the male and female sites. Sometimes unidirectional movement of the univalent chromosomes at the female meiosis includes all the chromosomes of the odd genome(s) into the functioning mega¬ spore, thus giving rise to a viable embryo-sac (cf. Lewis & John, 1963, p. 337). At the same time the distribution of univalents in the pollen mother cells is much more random, resulting in only a low percentage of germinable pollen grains. Basically the latter is true for the isosporous ferns too. Here the probability of a balanced chromosome set after the meiotic divisions is further decreased by the high base number (jc = 41 in Polystichum). Supposing a random distribution of the 41 univalents, the occurrence of the extreme situation 0 and 41 at the opposite poles is rather unlikely (1/241), even if we consider the large number of trials (c. 1 05— 1 07 per frond). We cannot be sure, however, that the univalent distribution is really a random process. It is also disputable that only spores with complete genomes are viable. Furthermore, the observed occurrence of diads instead of tetrads in some spore formation tells us that this distributional difficulty can probably be overcome by suppressing the meiotic process after the first division. In any case, the meiosis determines the fate and genetic character of the resulting spores. Theoretically, a triploid fern hybrid (HHN) with two homologous (HH) genomes and a single unrelated non-homologous (N) genome can result in the following types of spore formation: Type 1 Meiosis I. Equal distribution of chromosomes of the even genomes; unequal distribution and often ‘micro-nuclear’ separation of the univalents (odd genome). Meiosis II. Separation of chromatids. Result: 4 abortive spores -l-a few very small shrunken extra-spores from the ‘micronuclei ’. Sporophyte: none. Type 2 Meiosis I. Equal distribution of chromosomes (paired genome) ; equal distribution of chromatids (univalents). Meiosis II. Blocked. Result: 2 large viable spores with unreduced (3x) chromosome sets. Sporophyte: via selfing, 6x; via apogamy, 3x (as the plant). Type 3 Meiosis I. As in Type 2. 134 TAXONOMIC PROBLEMS IN POLYSTICHUM Meiosis II. Equal distribution of chromatids of the even genome; unequal distribution of chromatids of the odd genome. Result: same as in Type 1. Sporophyte: none. Type 4 Meiosis I. As in Type 2. Meiosis II. Equal distribution of chromatids of the even genome; inclusion of all the chromatids of the odd genome into one of the opposite poles. Result: larger diploid and smaller haploid spores genetically equivalent to the two parental species of the Fx hybrid. Sporophyte: via selling, 2x and 4x parental species (P1? P2); via hybridization, 3x hybrid as the Fj. Type 5 Meiosis I and II. Same as in Type 4, but instead of inclusion of the odd genome into one of the opposite poles, it remains alone and forms an extra nucleus between them. Result: 5-6 small spores, each consisting of a haploid genome of either the ‘odd’ or the ‘even’ genomes. Sporophyte: diploid species, one identical with the 2x parent of the Fx, and the other one identical with a diploid species apparently not involved directly in the Fx (being the other 2x progenitor to the tetraploid parent). Type 6. Chromosomes are doubled prior to meiosis, consequently there is a normal meiosis (as a 6x plant). Result: 4 large triploid spores. Sporophyte: same as in Type 2. Out of these theoretical alternatives only Types 1, 2 and 3 were observed in meiosis of both P.xillyricum and P.xbicknellii. From these only Type 2 can yield progeny. Unfortunately we do not know whether they formed sporophytes sexually or asexually. The almost equal number of 3x and 6x F2 sporophytes supports the assumption that both sexual and apogamic reproduction can occur in the case of a triploid gametophyte. The fact that the frequency of Type 2 spore formation (diads) has been increased in F2 plants can be explained by its high selective advantage. After many generations this could probably lead to an established special form of apogamic reproduction. Although Types 4 and 5 have not been unequivocally observed, their possibility cannot be ruled out. Intermediates between Types 3 and 4 have actually been seen in some cases. One of the reasons for setting up a new, large-scale experiment in Budapest was to demonstrate the occurrence of these types in Polystichum hybrids. If they really exist, G. VIDA AND T. REICHSTEIN 135 they could explain the putative introgression between Polystichum aculeatum and the related diploid species. On the other hand back- crosses of a hexaploid P.xillyricum or P.xbicknellii would almost surely be completely sterile because of the further difficulties with multivalent chromosomes, and introgression in this way is hardly possible. References Lewis, K. R. & John, B. (1963). Chromosome marker. London: Churchill. Manton, I. (1950). Problems of cytology and evolution in the Pteridophyta, pp. 316. London: Cambridge University Press. Manton, I. & Reichstein, T. (1961). Zur Cytologie von Polystichum braunii (Spenner) Fee und seiner Hybriden. Ber. schweiz. bot. Ges. 71: 370-83. Meyer, D. E. (1960). Zur Gattung Polystichum in Mitteleuropa. Wildenowia 2: 336-42. Roy, S. K. & Manton, I. (1965). A new base number in the genus Lygodium. New Phytol. 64: 286-92. Sleep, A. (1966). Some cytotaxonomic problems in the fern genera Asplenium and Polystichum. Ph.D. thesis (unpubl.). University of Leeds. Sleep, A. (1971). Polystichum hybrids in Britain. Br. Fern. Gaz. 10: 208-9. Sleep, A. & Reichstein, T. (1967). Der Farnbastard Polystichum xmeyeri hybr. nov. = Polystichum braunii (Spenner) Fee x P. lonchitis (L.) Roth und seine Cytologie. Bauhinia ( Basel) 3: 299-374. Sleep, A. & Synnott, D. (1972). Poly stichumx illy ricum: hybrid new to the British Isles. Br. Fern Gaz. 10: 281-2. Vida, G. (1966). Cytology of Polystichum in Hungary. Bot. Kozl. 53: 137-44. Vida, G. (1972). Cytotaxonomy and genome analysis of the European ferns. In G. Vida (ed.), Evolution in Plants. Symp. Biol. Hung. 12: 51-60. Akad. Kiado, Budapest. APPENDIX I Ten special demonstrations were mounted for the Conference. Summaries or short papers based on four of these demonstrations are given below. The others, not separately discussed here, were as follows. B.S.B.I. and Council for Nature: ‘Save these Flowers’ - a conservation exhibit. *Mr B. Mathew and Miss C. A. Brighton: Cytotaxonomy of Crocus. Mr and Mrs D. Parish: Colour photographs of flowers of Europe. *Dr C. A. Stace: Hybrids in Juncus section Genuini. Dr S. M. Walters: Index to the rare endemic plants of Europe. Dr S. M. Walters and others: Problems of local floristic conservation. The two exhibits marked * illustrated papers given to the Conference, and therefore printed in the main text of this volume. [ 136] RANGE EXTENSIONS IN THE HYDROCHARITACEAE C. D. K. Cook Of the 12 fresh-water genera of the Hydrocharitaceae only 2 ( Maidenia and Nechamandra - both possibly congeneric with Vallisneria ) show any inclina¬ tion to ‘stay at home’. It is difficult to say why this family is particularly well-equipped for extending its range. Many species are attractive or interesting aquarium plants and many introductions are a direct result of the aquarium trade. However, some species such as Blyxa japonica and Ottelia alismoides are ricefield weeds and were probably introduced with imported rice seed. In their native ranges many Hydrocharitaceae are found in eutrophic or brackish conditions, and may therefore be pre-adapted for various kinds of chemical pollution. Another factor probably encouraging their establishment in Europe is thermal pollution, and it is interesting that most of the recent intro¬ ductions are plants of warmer regions. The following is a list of genera and recent range extensions. The pondweed genera ( Egeria , Elodea, Hydrilla and Lagarosiphon) are dealt with at the end. Blyxa: c. 10 species, native in the warmer regions of Asia. B. aubertii has become naturalized in N. America, and B. japonica has become naturalized in Europe (northern Italy). Hydrocharis : 3-6 species, native in the Old World. H. morsus-ranae, native in Europe, has become established in N. America. Limnobium: c. 3 species, native in the warmer parts of S. America. L. laevigatum has become established in S. Europe and Java. Maidenia: 1 species, M. rubra, native in N.W. Australia. It is a poorly studied plant and is possibly congeneric with Vallisneria. Nechamandra: 1 species, N. altemifolia, native in India and S.E. Asia. It is possibly congeneric with Vallisneria although recent workers have combined it with Lagarosiphon and Hydrilla. Ottelia: c. 40 species, native in the warmer parts of the world, 1 species in Brazil and the rest in the Old World with centres of speciation in central Africa and southern China. This genus is much in need of revision. O. alismoides [ O . japonica ] is naturalized in Europe (northern Italy). Stratiotes: 1 species, S. aloides, native in Europe and northern Asia. It is apparently extending its range in Europe and has been reported as introduced in N. America. Vallisneria: 6 to 10 or more species, cosmopolitan but absent from cold regions. This genus is much in need of revision. V. spiralis is extending its range in Europe ; this and other species are apparently extending their ranges in many other parts of the world. The pondweeds ( Egeria , Elodea, Hydrilla and Lagarosiphon ) in the vegetative phase are rather alike. For absolute certainty in determination flowers are necessary. The following is an attempt at a key, based on vegetative characters, [ 137] 138 APPENDIX I to the pondweeds introduced into Europe. It is possible that additional species may exist in Europe. Flowering material should be collected and sent to a botanical centre, or reference should be made to the cited works of Obermeyer and St John. 1A Leaves spirally arranged (at least at base of stem) 2 A Leaves stiffly recurved, usually densely packed, up to 2-5 cm long Lagarosiphon major 2B Leaves flaccid, spreading, not densely packed, very rarely more than 1-5 cm long Lagarosiphon muscoides 1 B Leaves in whorls or opposite pairs 3 A Leaves usually reflexed, flaccid, densely packed together, usually exceeding 2 cm in length Egeria densa 3B Leaves spreading, not reflexed, flaccid, rarely densely packed, rarely exceeding 2 cm in length 4 A Leaves usually in whorls of more than 3 (often 6-8) 5 A Teeth on margins of leaves distinctly visible to naked eye; whorls of fewer than 8 leaves not frequent Hydrilla verticillata 5B Teeth on margins of leaves barely visible to naked eye; whorls of fewer than 8 leaves very frequent 6 A Teeth on margins of leaves just visible to naked eye; female flowers with large showy petals Egeria najas 6B Teeth on margins of leaves not visible to naked eye; female flowers without petals Elodea nuttallii 4B Leaves rarely in whorls of more than 3 7 A Leaves elliptic to ovate-lanceolate, with rounded apex, rarely more 7B Leaves gradually tapering into a long, narrow, pointed tip, up to 2 cm or more long 8 A Leaves in whorls of 2 to 6; middle part of leaf up to 2 mm wide; leaf margins curving to a pointed tip; sepals of female flowers rarely more than 2 mm long; styles forked at tip Elodea nuttallii 8 B Leaves opposite or in whorls of 3 ; middle part of leaf rarely more than 1 mm wide; leaf margins straight at apex, gradually tapering to a very fine pointed tip; sepals of female flowers up to 3 mm or more long; styles deeply forked Elodea emstae Egeria : like Elodea but male spathe 2- to 4-flowered; female spathe split halfway down one side; petals showy, about 3 times larger than sepals; stamens 9(-10). Occasionally included within Elodea : see St John, H. Mono¬ graph of the genus Egeria Planchon. Darwiniana 12: 293-307, 523 (1961). 2 species, native in warm temperate S. America. E. densa has become naturalized in Europe, Japan, Africa, N. and Central America, Australia and New Zealand. E. najas is sporadic in northern Europe. It is a popular aquarium plant but apparently does not persist in Europe. Elodea [ Anacharis ]: c. 17 species in N. and S. (but not Central) America. Three species are found in Europe, but other species might be expected. Introduced plants of Elodea are nearly always female. St John, H. Monograph of the genus Elodea. I. The species found in the Great Plains, the Rocky Mountains, and ERRATUM Page 138 after 7 A in key remainder of sentence is missing; this should read: than 1.5cm long; plants usually robust Elodea canadensis APPENDIX I 139 the Pacific States and Provinces of North America. Res. Stud. Wash. State Univ. 30: 19^44 (1964). II. The species found in the Andes and western South America. Caldasia 9: 95-113 (196n). III. The species found in northern and eastern South America. Darwiniana 12: 639-52 (1963). IV. The species of eastern and central North America, and Summary. Rhodora 67: 1-35, 155-80 (1965). Elodea ernstae (often incorrectly determined as E. callitrichoides ) is native in S. America and naturalized in S.E. England, France and perhaps elsewhere. It is very like E. nuttallii in the vegetative stage. Elodea canadensis is native in N. America but has become established in Europe, Asia, Australia, New Zealand and Africa. It was introduced into the British Isles in 1836 and spread rapidly causing great alarm. Today it is widespread but has settled down and is no longer considered to be troublesome. Elodea nuttallii [= E. occidentalis ] is native in N. America but has become established in Holland, Belgium, ?France, northern Germany, northern Swit¬ zerland and ? British Isles. The so-called E. nuttallii from Esthwaite Water and western Galway is much smaller and more delicate than the E. nuttallii actively spreading in Continental Europe. Flowering material of the British plants is needed before one can be sure of their identity.* Hydrilla: 1 species, H. verticillata , native in tropical and temperate Asia and northern Europe (Upper Dnieper, Lithuania, Latvia and northern Poland). It has not been found in W.C. or S. Europe but has become naturalized in S.E. U.S.A. It can be distinguished from Elodea and Egeria by the larger teeth on the leaf margin and the regularly high number of leaves (c. 8) in each whorl. Lagaro siphon: c. 16 species, native in Africa and Malagasy: see Obermeyer, A. The African species of Lagarosiphon. Bothalia 8: 139-46 (1964). It can be distinguished from Elodea, Egeria and Hydrilla by the spirally arranged leaves. L. major is established in Europe and New Zealand. L. muscoides is sporadic in Europe. It is a popular aquarium plant but apparently does not persist in nature. Editor's note The exhibit by Professor Cook showed living material of a number of fresh- water aquatic plants of taxonomic interest to European botanists. In this short account the author has concentrated on the Hydrocharitaceae, species of which in recent years have significantly extended their range in parts of Europe. Professor Cook’s new book, Water Plants of the World (Junk, The Hague, Netherlands), the publication of which is expected as this Conference report goes to press, will undoubtedly make a very valuable contribution to the taxonomic and ecological understanding of the fresh- water flora of the world. * Since this paper was given, Professor Cook has received good material of E. nuttallii collected near Oxford, by Mr R. Palmer. IDENTIFICATION BY POLYCLAVE R. J. Pankhurst and R. R. Aitchison A method of identification by using punched cards (a polyclave) was demon¬ strated. This works by putting together cards in the hand for each character of the specimen, and eliminating any taxon which does not agree. At the end the position of the clear hole(s) in the pack of cards indicates which taxon (taxa) is the answer. The method is more convenient than a conventional key because any characters can be used in any sequence. The novelty shown here is that the cards are derived and manufactured by computer. This means that such keys are easy to create and to revise, and cheap to produce. Keys were shown for the British microspecies of Rubus (400) and Taraxacum (133). The computer program (Pankhurst & Aitchison, 1975) and documentation are available on request (RJP). The Rubus polyclave and descriptions of how to use it are also available from RJP. Reference Pankhurst, R. J. & Aitchison, R. R. (1975). ‘ A computer program to construct polyclaves. ’ In Proceedings of meeting on ‘ Automatic Identification of Biological Specimens' , ed. R. J. Pankhurst. London: Academic Press. [ 140] MAPPING THE EUROPEAN VASCULAR PLANT FLORA J. Suominen (1) Atlas Florae Europaeae (AFE; vols. 1 and 2 published) consists of maps showing the European distribution of species and subspecies. Established introductions and extinctions are indicated by special symbols. Textual com¬ ments on the maps give important synonyms, notes on taxonomy and nomen¬ clature and important new or omitted records (mainly complementing or correcting Flora Europaea ) as well as references to total range maps. A map in vol. 1 (p. 11), which gives the numbers of species of Pteridophyta in the 50-km squares, reveals the differences in the richness of the flora, and also reflects variation in the intensity of floristic research. (2) Goal of AFE. The atlas is independent of Flora Europaea , although it has close connections with that work, as it complements the distributional information given there. The basic aim is to proceed in systematic order and include all the taxa that can reasonably be mapped. This is a most fundamental point and distinguishes AFE from all the other European mapping projects undertaken so far, which either do not cover the whole of Europe or do not include all the European species. In view of the high standard of many of these atlases, the fact that all the European species will be included in taxonomical order seems to be the main reason and justification for the preparation and publication of AFE. It is hoped that the regrettable necessity of publishing many incomplete or even provisional maps may serve as a stimulus to further botanical research, both taxonomical and chorological. (3) Method of producing AFE. Each European country is represented by one (sometimes more) member of the Committee for Mapping the Flora of Europe (CMFE). He, with assistant botanists, is responsible for collecting floristic data for the 50-km UTM grid squares in his country. These national data, usually entered in a section of the base map, are submitted to the secretariat of the CMFE in Helsinki. The secretariat uses the maps from the different countries to compile the final European maps and prepares the textual comments on the maps. The map and text drafts are then sent for checking to the Committee members and some specialists in taxonomy before they are printed. The individual countries meet the cost of collecting their own records. The secretariat receives a yearly grant from the Finnish Ministry of Education to cover technical and material expenses. AFE is published by Societas Biologica Fennica Vanamo and distributed by Ticto Ltd., 5 Elton Road, Clevedon, Avon BS21 7RA. (4) Historical background of the CMFE and AFE. Two important events in European botany preceded the launching of the scheme to map the distribution of the European vascular plant flora These were the publication of the Atlas of the British Flora in 1962 and the completion of vol. 1 of Flora Europaea in 1964. The commencement of the mapping scheme was British as well. At the [ 141 ] 142 APPENDIX I Tenth International Botanical Congress in Edinburgh in 1964 Dr F. H. Perring presented a map showing the 50-km-square distribution of Silene acaulis. He then organized a network of European botanists to take part in a mapping experiment with ten species, the results of which were displayed at the Fourth Flora Europaea Symposium in Arhus, Denmark, where the CMFE was founded in August 1965. The secretarial duties were offered to Helsinki and the secretariat was installed there in December 1965. During the subsequent years further experiments were undertaken and a mapping team was organized to cover all the European countries. The results of these experiments and questions relating to mapping techniques and the organization of the scheme were discussed by the CMFE at meetings held in 1966 in Cracow, Poland, and in 1968 in Halle, DDR. By 1973, when a meeting took place in Varna, Bulgaria, the two first volumes of AFE had been published, and the main topic of the meeting was what steps should be taken to ensure that the following volumes of AFE were completed at a reasonably rapid rate. . (5) The production of AFE - present situation and future prospects. The use of a network of national representatives has many valuable advantages, making it possible to utilize local sources of information and the most recent records, to avoid errors in the literature and those caused by misleading place-names, and to reach satisfactory decisions as to the native or introduced status of plants and extinctions. However, such a team is really effective only when each country is able to keep up with the others. Here lies the weak point of the mapping scheme. Occasional difficulties and other duties may prevent commit¬ tee members from submitting data in time, and a delay in any single European country can easily hold up the entire mapping scheme and retard the publication of AFE. In view of the fact that only a small percentage of the European vascular plant species has so far been mapped in AFE, it is absolutely essential for the continuation of the mapping scheme and the completion of the atlas to find some means of working to a reasonable schedule. 04/01 SOME EUROPEAN GERANIUMS WITH ASIATIC CONNECTIONS P. F. Yeo and H. Kiefer The work in progress consists of (a) a biosystematic investigation into various groups of species within which there are known or suspected micro-evolutionary affinities, ( b ) an attempt to arrive at an improved classification of the genus, including the use of computer-generated groupings. The exhibit showed two species-groups, (1) the G. pratense group, showing that various Central Asian and Himalayan forms can be considered as either falling within G. pratense or being closely related to it; (2) the G. cinereum group, consisting of a series of geographical vicariants extending from N.W. Africa through Spain, the Alps, Italy, the Balkans and eastern Turkey; specific separation of G. argenteum from the rest of the group appears questionable. [ 143] APPENDIX II ADDRESSES OF CONTRIBUTORS Mrs R. R. Aitchison, Botany School, Cambridge. Miss C. A. Brighton, Jodrell Laboratory, Royal Botanic Gardens, Kew, Rich¬ mond, Surrey. Professor C. D. K. Cook, Botanischer Garten der Universitat, Pelikanstrasse 40, 8001 Zurich, Switzerland. Dr R. Czapik, Institute of Plant Anatomy and Cytology, The University, Grodzka 52, 31-044 Cracow, Poland. Dr M. J. Fernandez-Morales, Departamento de Biologia, Colegio Universi- tario de Ciencias, Malaga, Spain. Dr M. Fischer, Botanisches Institut und Botanischer Garten der Universitat, Rennweg 14, 1030 Vienna, Austria. Dr C. Gardou, Laboratoire de Taxonomie Vegetale, Universite de Paris Sud, Batiment 362, 91405 Orsay, France. Professor J. Grau, Institut fur Systematische Botanik, Menzinger Strasse 67, 8000 Munich 19, Germany. Dr W. Greuter, Goulandris Natural History Museum, 13 Levidou Street, Kifisia, Greece (now at Conservatoire et Jardin Botaniques, 192 Route de Lausanne, 1202 Geneva, Switzerland). Dr B. Jonsell, Institute of Systematic Botany, P.O. Box 123, 751-04 Uppsala, Sweden. Miss H. E. M. Kiefer, University Botanic Garden, Cambridge CB2 1JF. Mr B. Mathew, Royal Botanic Gardens, Kew, Richmond, Surrey. Mr R. J. Pankhurst, Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD. Dr D. Papes, Department of Botany, University of Zagreb, 41001 Zagreb, Yugoslavia. Mr D. and Mrs M. Parish, 5 Venator Place, Minster Park, Wimborne Minster, Dorset BH21 1DQ. Professor T. Reichstein, Institut fur Organische Chemie der Universitat, St Johanns-Ring 19, 4056 Basel, Switzerland. Dr C. A. Stace, Department of Botany, The University, Manchester M13 9PL (now at Department of Botany, The University, Leicester LEI 7RH). Dr W. T. Stearn, Department of Botany, British Museum (Natural History), Cromwell Road, London SW7 5BD. Dr J. Suominen, Department of Botany, University of Helsinki, Unioninkatu 44, Helsinki 17, Finland. Dr G. Vida, Department of Genetics, Eotvos Lorand University, Miizeum krt 4/a, 1088 Budapest, Hungary. Dr P. F. Yeo, University Botanic Garden, Cambridge CB2 1JF. [ 144] , BOTANICAL SOCIETY OF THE BRITIS1 Conference reportsslono C001 ih PUBLICATIONS Prices, including postage, revised 1974 CONFERENCE REPORTS 3 0112 009164309 1 British Flowering Plants and Modern Systematic Methods. Ed. A. J. Wilmott, 1948. Wrappers £1.75 2 The Study of the Distribution of British Plants. Ed. J. E. Lousley, 1951. Wrappers £1.75 3 The Changing Flora of Britain. Ed. J. E. Lousley, 1953. Bound £2.35 Out of print 4 Species Studies in the British Flora. Ed. J. E. Lousley, 1955. Bound £2.30 5 Progress in the Study of the British Flora. Ed. J. E. Lousley, 1957. Bound £2.25 6 A Darwin Centenary. Ed. P. J. Wanstall, 1961. Bound £2.25 7 Local Floras. Ed. P. J. Wanstall, 1963. Bound £2.25 8 The Conservation of the British Flora. Ed. E. Milne- Redhead, 1963. Bound £2.00 9 Reproductive Biology and Taxonomy of Vascular Plants. Ed. J. G. Hawkes, 1966. Bound £2.20 Out of print 10 Modern Methods in Plant Taxonomy. Ed. V. H. Heywood, 1968. Bound £6.10 11 Flora of a Changing Britain. Ed. F. H. Perring, 1970. Paper¬ back reprint £1.35 12 Taxonomy and Phytogeography of Higher Plants in Relation to Evolution. Ed. D. H. Valentine, 1972. Bound £7.60 13 Plants Wild and Cultivated. Ed. P. S. Green, 1973. Bound £2.80 14 The British Oak. Eds. M. G. Morris & F. H. Perring, 1974. Bound £6.35 MISCELLANEOUS PUBLICATIONS British Herbaria. D. H. Kent, 1958. Bound £2.55 Flora of Islay and Jura (v.c. 102). J. K. Morton, 1959. Wrappers 95p West Norfolk Plants Today. C. P. Petch & E. L. Swann, 1962. Wrappers £1.40 British Sedges. A. C. Jermy & T. G. Tutin, 1968. Wrappers £1.45 Index to Botanical Monographs. D. H. Kent, 1967. Wrappers £3.05 The Taraxacum Flora of the British Isles. A. J. Richards, 1972. Wrappers £2.25 All orders should be sent to E. W. Classey Ltd., Park Road, Faring- don, Oxon. SN7 7DR.