INTRODUCTION
|
|
INTRODUCTION |
This work covers the whole of Turkey as politically constituted today. We have also included most of the eastern islands of the Aegean (part of Greece) for two reasons: because they are floristically closer to Anatolia than to the rest of Greece, and because they are excluded from Flora Europaea; these are the islands shown to the east of the broken line in Map 2 (p. 10).
The Flora attempts to include all those species native to Turkey, or which are established aliens; in practice it is often impossible to distinguish between these two groups (for example, with many weeds). Cultivated plants (mostly crop plants or trees) are included only when they have become naturalised, or are so common as to be a feature of the landscape and provide a habitat for the native flora.
On the whole we have adopted a fairly broad generic concept for the Flora, although some allowance has had to be made for the preferences of specialists invited to revise particular groups. The species concept adopted is also fairly broad; but as we frequently have only very limited material, many taxa recognised as species may have to be reduced in rank (or included in synonymy) when they become better known. For specific rank we have usually required correlated discontinuous variation in at least two taxonomic characters, although we have sometimes accepted less when traditional usage is strong.
We have only recognised two infra-specific categories - subspecies and variety - and these are used somewhat sparingly in accordance with our state of knowledge. Subspecific rank is used for those cases in which variation is only partly discontinuous, or when only a single character is usually coupled with a distinctive distribution (geographical or ecological). Varietal rank is used for two situations: (1) when the taxon is believed to represent a local (rather than a regional) variant; (2) when a striking variant is insufficiently known, so that its status is uncertain. Single-character variants, apparently scattered at random and not forming stable populations (formae), are not formally recognised (unless by accident) in the Flora. It is only in a very few, well-studied groups that cytogenetic information on Turkish material has been sufficient to influence taxonomic treatment.
It was decided to use a grid system, based on two degrees of latitude and longitude, as the primary division for the citation of specimens. The rectangular shape of Turkey means that a grid is particularly convenient, and can be readily memorised. This grid results in Turkey being divided into twenty-nine squares, as shown in Map 1. As can be seen on the same map, we have also used the modern provinces (vilayets) for the citation of specimens. It should be added that some of these vilayets have suffered boundary changes as a result of political vicissitudes, making a grid system particularly useful as a primary reference system.
In addition to the squares and vilayets, however, we have recognised certain other, larger divisions that are used for conveying the general distribution of a taxon in Turkey before the citation of specimens. This has been largely based on the editor's field knowledge and our acquaintance with distributional patterns in[1]
|
map 1 Turkish Vilayets and Grid system (cf. p. 1). |
Anatolia. Six major divisions arc recognised in Turkey: Turkey-in-Europe, North, West, South, Central, and East Anatolia, as shown on Map 2; the East Aegean Islands form an additional division.
As discussed more fully below, the Anatolian divisions approximate to phytogeographical areas, although we have often followed vilayet boundaries when these do not do violence to our uneven knowledge of the flora and vegetation.
Within the five major Anatolian divisions, we have recognised several subdivisions, mainly formed by grouping the vilayets together into what are believed to be fairly well-marked or convenient areas (cf. p. 11). Most of these have been named according to the classical provinces used by Boissier, although the boundaries are our own. These subdivisions are shown on Map 2 (p. 10) and are tabulated below.[ The terms Armenia and Kurdistan are used in a purely geographical sense and have no political implications. ] When convenient we have grouped some of the provinces together under the titles of North-West Turkey, South-West, North-East and South-East Anatolia, as shown by the stippled areas on Map 2. The term Outer Anatolia has been used to include North, West and South Anatolia, and Inner Anatolia to include Central and East Anatolia. Although by means of the grid the Flora can be used without reference to the divisions we have described, we hope they will be found useful by those interested in plant distribution. These divisions are intimately linked with the attempt to assign the species to phytogeographical elements (cf. p. 24); they also serve to link the distributional data with those given in earlier literature.
Asiatic Turkey (Anatolia) covers about 290,000, and Turkey-in-Europe 9250, square miles - a total area that is nearly a third larger than that of France or the Iberian peninsula. Most of Anatolia consists of a plateau, rising steadily towards the east and bounded on the north and south by steep mountain ranges - part of the Alpine-Himalayan system. East of the Anti-Taurus, the southern bounding range curves round in an arc to the mountain mass of Kurdistan in South-East Anatolia, embracing the outer plateau of Mesopotamia which is really the northern extension of the Syrian Desert. In West Anatolia the plateau falls gradually to sea level, and terminates in a series of promontories which face the islands of the Dodecanese (Sporades) and Khios.
North Anatolia. Throughout most of this area the coast plunges steeply into the Black Sea which, like the south coast of Anatolia, is almost devoid of islands. In the east (Lazistan), the range is high, continuous and close to the sea, rising at many points to over 3000 m, the highest peaks being in the Tatos Daglari north-east of Rize. Most of the rock in this area is basic igneous, though many of the jagged highest peaks are of granite. The rivers cleave their way through deep gorges to the sea. On the south side the range falls steeply to the plateau, the Kelkit and Coruh rivers following the fault lines parallel with the main range till they cut through the mountains near Samsun and Batumi (USSR) respectively.
From Ordu westwards the range spreads out; the general level of the northern range falls off and the highest peaks are set well back from the coast. Though most of this area consists of basic igneous rocks and flysch, the summits of many of the highest peaks are of crystalline limestone or marble. Several rivers cut their [3] way on to the plateau, which in this part of Anatolia rises to meet the northern range.
From east to west the major North Anatolian rivers are the Coruh, Yesilirmak (of which the Kelkit is a tributary), Kizilirmak, Filyos or Yenice Cay (which passes through magnificent limestone gorges) and Sakarya; of these, the Yesilirmak and Kizilirmak have built up alluvial fans. These five rivers appear to have played an important part in the infiltration of Mediterranean floral elements from the Black Sea coast into the interior of North Anatolia. The oldest rocks (Devonian and Silurian) are found in the neighbourhood of the Bosporus.
West Anatolia. Many of the mountain ranges run east to west, and are divided by broad meandering river valleys: for example, the Gediz and Büyuk Menderes rivers. Except south of Denizli, few of the mountains rise above 2000 m. The geology is very varied; chalk at lower altitudes, basic igneous rocks, limestone, mica schists (for example, mountains north and south of Odemis) and granite making up various parts of the country. A considerable area of serpentinous rocks, centred on Sandras Dagi and Marmaris, occurs in the south-west (Caria). The coastline is greatly indented by gulfs and promontories, many of which are still little known botanically.
South Anatolia. In this Flora, the term South Anatolia is used to include the whole of the Taurus range, Amanus, Ala Dag and Anti-Taurus. As in North Anatolia, the coastline is very steep, coastal plains of any size being confined to the neighbourhood of Antalya (Pamphylia) and Adana (Cilician plain). There is a break in the high mountains between the massif of the Lycian Taurus in the west and the Isaurian Taurus, which is separated from the main block of the Cilician Taurus (Bolkar Daglari) by the Goksu valley. Similarly, deep gorges separate Bolkar Daglari from Ala Dag to the north-east. This massif continues north-eastwards as the Anti-Taurus, which encloses another Goksu river between its two main ranges; the eastern arm continues southwards to form the Amanus range (Gavur Daglari), which runs close to the Gulfoflskenderun until it reaches the Syrian frontier at Akra Dag (Cassius). [ For convenience, the whole of Akra Dag, whose southern foothills are in Syria, is included in the Flora. ]
The dominant rock of South Anatolia is hard limestone, but considerable areas of basic igneous rocks, shales and slates occur from the Isaurian Taurus eastwards. There are extensive areas of soft chalky rocks near Ermenek and Mut in the Goksu valley of Cilicia; diorite is found above Bulgar Maaden in the Cilician Taurus. Limestone is scarce in the Amanus, where basic igneous rocks predominate.
Central Anatolia. Much of the plateau of Central Anatolia lies between 800 and 1000 m, although various mountains rise much higher. In the centre, the plateau falls gradually to the treeless depression containing the large salt lake, Tuz Golu. Much of this central plateau is covered by very soft rocks - chalk, clay and marl; in the vicinity of the Cankiri and Sivas there are areas of gypsaceous chalk hills.
Several volcanoes - now extinct - rise from the plateau of Inner Anatolia; some of the more prominent in this area are Kara Dag near Karaman, Hasan Dagi and Melendiz Daglari near Nigde, and Erciyas Dagi (2916 m) near Kayseri. Most of these, like their counterparts further east, have been formed in Quaternary times. A large area of volcanic tuff occurs near Nevsehir.
East Anatolia. Apart from Mesopotamia, East Anatolia is higher and much [4] more mountainous than Central Anatolia, the level of the plateau itself reaching over 1700 m in Kurdistan and North Armenia (Armenian Highlands). A mountainous belt, which we may call the 'Anatolian Diagonal', extends from the Anti-Taurus north-eastwards to Susehri. Drainage of the East Anatolian area is into the Black Sea through the Coruh and Kelkit rivers in the north; into the Caspian by the Aras river in the east; and into the Persian Gulf in the South by the Dicle (Tigris), largely fed by the snows of Kurdistan, and the numerous tributaries of the Firat (Euphrates).
Much of East Anatolia is covered with basic igneous rocks of various kinds (between Malatya and Kangal, for instance, and between Elazig and Maden, one sees little else), but Munzur Daglari (south of Erzincan) is of hard limestone.
Many of the high mountains of Kurdistan are of limestone, Cilo Dag (4168 m) being largely dolomitic. High mountains of Quaternary volcanic formation are Süphan Dagi (4434 m) and Nemrut Dagi near lake Van; the highest peak in Turkey, Agri Dagi (Ararat, 5165 m), has a similar origin. Extensive areas in North-East Anatolia (for example, between lake Van and Ardahan in the Armenian Highlands) are covered by basaltic lava flows.
Mesopotamia, in its topography, climate, and flora, stands somewhat apart from the rest of East Anatolia. Though often strictly limited to the plains between the Tigris and Euphrates, we have used the term Mesopotamia in a somewhat wider sense, extending it westwards to include everything between the Amanus range and the mountains of Kurdistan. This outer plateau is considerably lower and flatter than the inner plateau of East Anatolia, falling gently from 800 m in the north to about 400 m near the Syrian frontier. It consists of rolling hills and vast plains, the wide basaltic hump of Karaca Dag (1919 m) providing the major relief. Mardin stands near the western rim of a limestone escarpment which juts out from the mountains of Kurdistan east of the Tigris. Looking southwards from Mardin, one sees the vast expanse of the Syrian Desert lying below.
Turkey-in-Europe. This political unit is topographically, climatically, and biologically heterogeneous. The Istranca Daglari that borders the Black Sea is (so far as the flora is concerned) a low continuation of the northern range of Anatolia and is composed largely of schists. Most of the country is occupied by undulating plains drained by the Ergene river. On the north-west side of the Sea of Marmara (which divides Europe from Asia) is the low sandstone range of Tekir Daglari which continues southwards into the Geibolu (Gallipoli) peninsula.
For an account of Turkish geography, see Birinci Cografya Kongresi, published in Turkey (Ministry of Education, 1941). A geological map, Turkiye Jeolojik Hartasi (in eight sheets at 1:800,000) is published by the Maden Tetkik ve Arama Enstitiisu (Ankara, 1942).
North Anatolia. In contrast to the rest of Turkey, the most striking climatic feature of this part of Turkey is the heavy rainfall, which, particularly in the east (Lazistan), continues all the year round; when it is not raining, the hills are frequently swathed in mist. So wet is the climate at Rize and Hopa that tea has become a successful crop. West of Ordu the rainfall lessens, only to increase again in the neighbourhood of Zonguldak. Throughout this area west of Lazistan there is sufficient summer drought near the coast to hamper the establishment (or reestablishment) of Euxine forest which predominates inland and to favour a [5] coastal strip of Mediterranean vegetation. The coastal climate of Lazistan is mild, allowing the cultivation of Citrus crops; west of Cape Sinop cold winds bring lower winter temperatures, which probably account (as they do in the Crimea) for the absence of several common macchie shrubs from the coastal strip of Mediterranean vegetation in North Anatolia. In Lazistan luxuriant Euxine forest almost excludes the development of Mediterranean vegetation. Although precipitation in the area is often believed to increase with altitude, anyone camping out in summer on the peaks of Lazistan will find that skies are usually clearer above the tree line than they are below it; indeed, the alpine pasture may be quite burnt up while the valleys are filled with cloud.
West and South Anatolia. This area has a typical Mediterranean climate near the coast - mild, wet winters, and long, hot, arid summers with almost constant drought (broken by occasional thunderstorms in the mountains) from May to September. Although summer precipitation is negligible, humidity is not so low as it is on the plateau of Inner Anatolia. Rainfall in the Mediterranean belt of Turkey, however, varies considerably from one year to another, and in unfavourable years one can even see evergreen macchie shrubs with their leaves withering in mid-summer. Throughout much of the area snow lies in winter above 1000 m, so that the increased cold, low temperatures and drier air lead to the penetration of steppic elements from Inner Anatolia into the high mountains, where such vegetation usually predominates above the tree line.
Temperatures tend to be higher in South Anatolia than in the West, where in Mysia the Mediterranean climate is modified by the lower temperatures and higher rainfall of North Anatolia. Cold winds blowing outwards from the plateau retard the development of the spring flora on the Mediterranean coast, where it flowers considerably later than it does in southern Greece or the Lebanon.
Inner Anatolia. In so far as precipitation occurs predominantly in winter and spring, the climate of Inner Anatolia resembles that of the Mediterranean climate of West and South Anatolia. In most areas, however, precipitation is less and much of it falls as snow. Winter temperatures are very much lower, particularly in the highlands of East Anatolia, where nearly the whole area is under snow from November to March or April.
In summer, temperatures soar during the day and drop suddenly at night. A striking feature of the climate, which must do much to control the vegetation, is the very low summer humidity - much lower than in the Mediterranean region - and correspondingly high saturation deficit. This favours a predominantly herbaceous and suffruticose flora, and (with the exception of some conifers) precludes the growth of evergreen trees and shrubs.
The plateau tends to be very windy and suffers devastating hail storms. Precipitation is even more erratic than in the Mediterranean belt; in some years spring rains are so light that the flora is very poorly developed, while in a favourable year almost daily thunderstorms continue throughout May or even into June. This seems to be a characteristic feature of the climate of Kars. In general, however, the vegetation of the plateau (except for a few late-flowering perennials and halophytes) is desiccated by mid-summer; the leaves of most herbaceous species wither after flowering.
On the lower, 'outer' plateau of Mesopotamia temperatures are higher than in the rest of Inner Anatolia, thus accounting for Mesopotamia's floristic affinities with the Syrian Desert, of which it is but the northern extension. [6] In the highest mountains of Turkey, above 3000 m, snow lies throughout the summer. A glacier still exists on the north side of Cilo Dag in Hakkari, and a smaller one on Ararat. Elsewhere U-shaped valleys, hanging valleys, associated small lakes and moraines are the only traces that are left of Pleistocene glaciation.
This seems to have been fairly widespread in the high mountains, and evidence of it can be seen in many parts of North Anatolia, the Taurus and Kurdistan. Traces of local glaciation are often found as low as 2000 m, but there is certainly no evidence of an extensive ice sheet - the plateau remained unglaciated, although it was probably cooler and possibly wetter than it is today. It seems very likely that the Turkish flora suffered little extinction during the Pleistocene period, but a considerable shifting about of its vegetational belts must have occurred, of which the survival of numerous Euro-Siberian species in the Amanus (and further south in the Lebanon) is the most striking manifestation. At the present day the xerophytes seem to be spreading at the expense of the mesophytes, but this is no doubt partly the result of man's interference with the natural vegetation.
For a meaningful picture of Turkish climates, reference should be made to H. Walter's 'Climate-Diagrams' (Die Klima-Diagramme der Türkei. Karte 62 x 88 cm, Verlag Ulmer, Stuttgart) which incorporate graphs of monthly precipitation and temperature for most of the vilayet capitals. An account of Turkish climates according to Thomthwaite's classification is given by S. Erinc (Annals Assoc. Amer. Geographers 39: 26-46, 1949).
Boissier's Flora Orientails still provides the main frame of reference in which the Flora of Turkey is set. The sequence of families therefore follows, with a few alterations to bring it more into line with modem opinion, the order adopted by Boissier.
Genera, and the species within them, are arranged so that those most nearly related come as near to one another as a linear sequence allows. Subgenera, sections, and informal groups have only been used when their interpolation may be considered as a helpful guide in large genera.
Authorship of Accounts. This is given below the name of each genus or higher group.
Keys and descriptions. Except in the Filicales, indented dichotomous keys are provided for all taxa up to the rank of family. Descriptions of families and genera refer mainly to the groups as they are represented in the Near East. Descriptions of genera and species are designed as a check on the keys, and are therefore relatively short and diagnostic. Terminology is generally in accordance with G. H. M. Lawrence, Taxonomy of Vascular Plants (1951). In the case of some species that closely resemble other, usually commoner, species described in the Rora, differential diagnoses have often been given instead of descriptions. In such cases it should not be assumed that the taxa differ from one another only in those characters given in the diagnosis. Unqualified measurements refer to length.
The flowering time (or the time when spores are ripe) follows the description or differential diagnosis (for example, Fl. 5-6 = flowering May-June). Infra-specific taxa are keyed out under the species but are not usually otherwise described.
Descriptions of species mainly cover the taxa as they occur in Turkey; when two or more subspecies or varieties are present, the general description of the species [7] covers all of them; when only one infra-specific taxon is present, the description refers to that taxon only.
Cultivated plants included in the Flora are placed near their wild allies (if they have any) in the enumeration and briefly described, but are not usually included in the keys.
Monographs and revisions. Helpful works of this type are cited after the generic (or family) description.
Synonymy and references. Although each accepted species is cited with authority, reference and date, additional references have been used very sparingly, and are cited only when they add significantly to our knowledge of the species. No attempt has been made to include a full synonymy. This has been largely restricted to names accepted in Boissier's Flora Orientalis and in standard Floras of neighbouring areas. Basionyms, however, have always been cited. In general, synonyms cited in Boissier's Flora Orientalis or a later revision or monograph have been accepted unless the contrary is indicated by exclusion or citation under another taxon. An exclamation mark after a synonym means that type material (in the broad sense) has been examined. Due to lack of material, we have often been unable to decide whether extra-Turkish taxa described in Floras of neigh-bouring areas (particularly the Caucasus and Crimea) are synonymous with Turkish taxa accepted here.
Illustration and maps. Reliable illustrations are cited after the synonymy, preceded by 'Ic.:' (Icon); we have not repeated here illustrations cited in the original reference of the taxon. Our own illustration of critical parts are cited last, preceded by 'Figure'. References to our figures of subspecies and varieties follow the name of the taxon.
Various species have been mapped in the Flora, the choice often depending on the phytosociological (or phytogeographical) importance of the species. Dot maps have been used, each symbol representing a locality from which a specimen has been examined or cited. In the case of the Gymnosperms, which (with other forest trees) have been mapped in detail by Ziraat Vekaleti Orman Umum Müdürlügü, Güzel Sanatlar Matbaasi-Ankara, the general distribution of the species is indicated by a stippled area.
Ecological data. Ecological information and flowering time have been synthesised and condensed for each taxon, and are printed in italics. This information refers to the habitat of the species (or lower unit) as it occurs in Turkey. In many cases information is still very incomplete or lacking. So far as possible we have avoided the use of a zonation terminology, preferring to give the altitudinal range instead.
Typification. With the exception of Linnaean species, type localities and material of the majority of species and taxa of lower rank have been cited as an aid to further research. Type localities are given as originally published, allowing for some contraction when necessary, and the translation of feet (altitude) into metres. The whereabouts of holotypes and isotypes have only been cited when the precise nature and location of the type is certain. As a guide, we may say that the types of the following authors are most likely to be found in the institutions listed below (indicated by the contractions used in Index Herbariorum and in the main body of the Flora): Boissier and DeCandolle at Geneva (G); Willdenow at Berlin (B); Bieberstein, Schischkin and Grossheim at Leningrad (LE); Stapf (Luschan's collection) and Handel-Mazzetti at Vienna University (WU); [8]Labillardiere at Firenze (FI); Bornmüller at Berlin (B) or Jena (JE); Rechinger at Vienna (W); many of these are, of course, duplicated in other herbaria. Tournefort's collection, on which many species are based, is at Paris (P).
When holotypes or isotypes are not specified, no inference should be drawn as to the status of the 'type' in the herbarium cited. An exclamation mark after the herbarium contraction shows that the specimen has been examined; an exclamation mark after 'photo.' means that only a photograph of it has been seen. In general, lectotypes have not been chosen except where nomenclature has necessitated it; this choice is better left to monographers.
Although Linnaean species have usually not been typified, we have generally indicated from where they were described and frequently cited specimens in the Linnaean Herbarium* [* The numbers cited are those given in S. Savage's Catalogue of the Linnaean Herbarium (1945).] (Linnean Society, London) or Herbarium Cliffortianum (British Museum) which agree with our use of the name. This by no means implies that such a specimen should be accepted as the type.
General distribution in Turkey. For ease of reference, a summary of the general distribution in Turkey generally precedes the citation of (non-type) specimens.
This follows the scheme shown on Map 2 and in Table 1.
The term widespread has been used for plants distributed throughout most of our area, irrespective of whether such species are common or local. Scattered is used for species which have a very interrupted distribution, often cutting across phytogeographical regions; very scattered implies even more widely dispersed than 'scattered'.
Citation of specimens. In view of the revisional nature of the Flora and the attention we have given to distribution, specimens have been cited rather fully.
They are cited according to the square of the grid in which they occur (cf. p. 2), and within their vilayet (cf. Map 1). Two squares, Al and A2, are divided between Europe and Asia; A1(E) and A2(E) refer to the parts of these squares in Europe, and A1(A) and A2(A) the parts in Asia. Specimens from both parts of these divided squares are cited whenever they occur. For the rest, we have cited one specimen from each province in each square unless more than eight squares are represented; in that case only one specimen in each square is given. Plants from the East Aegean islands (Is.) are cited last, the majority of records being based on Rechinger's Flora Aegaea (1943).
Preference has been given to specimens seen (indicated by an exclamation mark) and to those which have been recently collected or are readily available. [ We hope to include a comprehensive list of 'post-Boissier' collectors and their itineraries in Volume 2.]
Records have been accepted from the literature with considerable caution. What we have included depends very largely on the difficulty of identification and on the expertise of the taxonomist who named the plant. Although we have tried to include in the Flora all species that grow as natives or established aliens in Turkey, we have certainly missed many grid or vilayet records. Though this incompleteness may seem unfortunate, it is preferable to the inclusion of numerous wrong records. Throughout the work we have tried to be responsible for the information provided, and not to rely heavily on the unchecked work of others.
A list of some of the leading floristic works on the Turkish flora, which have been most useful in providing records, is given on p. 15. Other works may be [9] cited under the genus or species. Where we have cited Grossheim's maps, we are referring to his Flora Kavkaza (ed. 2) still in progress; in this critical work Turkish records are largely based on collections in the Leningrad and Tbilisi herbaria made by Russian botanists in North-East Turkey before the First World War.
For records from Turkey-in-Europe, we are much indebted to the researches of Professor D. A. Webb, whose synthesis of the records from this area is due to be published in 1965.
Citation of specimens is usually restricted to locality, altitude, collector's name and number (or date). Place names are a perennial problem to students of the Turkish flora, since classical names, modem Greek, Russian, and old and modern Turkish names have all been used and spelling is often diverse; the fact that many different places have the same name makes their location a difficult (and sometimes insoluble) problem.[* Modern collectors should cite the vilayet on their herbarium labels.] In general we have modernised the old spelling if the name is an orthographic variant (for example, we have spelled 'Gumuschkhane' as 'Gümusane'). When the modem name is manifestly different from that on the label, it is cited as on the latter (as an aid in identifying the sheet) and the modern Turkish name is given in brackets. In using modem Turkish spelling, however, we have not used the circumflex or the undotted i. Apart from this, we have [11] map 3 & table 2 [13] adopted the spelling used in the Times Atlas (mid-century edition). A paper on some old and new place names has already been published by Davis (Notes R.B.G. Edinb. 22: 587-591, 1958); the localities of Sintenis are the subject of a paper by Cullen (pp. cit. 25: 31-39, 1963).
Distribution outside Turkey. This is summarised immediately after the citation of Turkish specimens, and is given only in general terms, becoming less precise the greater the distance from Turkey. So far as possible, distributions have been checked by reference to specimens. The terms used are a compromise between phytogeographical areas and (especially in south and east Europe) the limits adopted by the relevant standard Floras. The usage of the terms is shown on Map 3 and in Table 2 (p. 13).
Endemism. When a taxon is believed to be endemic to our area, this is clearly stated. So far as possible, an indication of its relationships (as assessed by overall resemblances) is given; if its allies do not grow in Turkey, their distribution is summarised.
Phytogeographical elements. The phytogeographical element to which a species is referred is cited wherever the editor feels that it can be given with confidence.
The concept of the element, and a brief account of the phytogeographical regions in Turkey, is given on pp. 16-26.
Observations. As far as space allows, we have not tried to hide lack of knowledge but to draw attention to it, thereby indicating taxonomic problems that can only be resolved by further collecting or biosystematic study. The existence of intermediates, exceptional variation, and other features are mentioned in observations under the relevant species, particularly when the taxonomic status of a species is doubtful. In critical genera the description is often followed by general remarks on the difficulties involved, and suggestions are given as to which parts of the plant are particularly useful for identification, and what special field notes would be helpful for the same purpose. By this means it is hoped that the standard of collecting may be improved and the group thereby become better understood. In this stage of our knowledge of the Turkish flora, it is not so much biosystematic information that is required for a sound basic classification (though in many groups such information would be welcome), but more and better collected herbarium material. Far too many species are still known only from the type specimen, or from scrappy, poorly labelled material.
Species imperfectly known or doubtfully recorded. Names that come in this
category are listed at the end of the genus unless they have already been referred to in observations under the accepted species. No attempt has been made to cite all those names which have been recorded from Turkey. Many records are based upon misidentifications, and the specimens concerned (when seen) have often been cited under the correct binomial in the enumeration. Turkish records which may be considered, on distributional grounds, to be highly unlikely, have usually been omitted.