The Society's Fieldwork in Turkey, 1997

by Brian Mathew

On 14 March 1996, the Secretary of the Royal Horticultural Society's Joint Rock Garden Plant Committee, Dr A C Lesley, wrote to Peter Moore about variation in C. trochopteranthum:

"At recent meetings of the Joint Rock Garden Plant Committee there has been some discussion relating to the forms of this species in cultivation. In particular the apparently distinctive character of what are known to be higher altitude collections compared with those from lower elevations: in large part a question of size (many other characters being variable within each group)". He continues: "Dr Chris Grey-Wilson has drawn attention to the fact that the type material is probably from a high altitude collection and represents the smaller plants in cultivation. what seems to be unknown is the degree of intergradation between the two extremes and whether there is much if any variation in size within populations if there is relatively little intergradation and little intra-population variation in size it might be useful to produce suitable names for the two groups (if only for horticultural use).The Joint Committee requested that I contact you to see if the Cyclamen Society would be prepared to add the field work necessary to answer these questions to their programme of research."

This, then, provides the background to the current series of field studies in south-west Turkey, the first of which took place one year later, in April 1997. The following article must be regarded as an interim progress report, since no conclusions can be reached on the basis of one set of observations.

First, it is useful to go briefly over some of the recorded information about C. trochopteranthum, although one has to admit that hard facts about it in the wild are surprisingly few. This early spring-flowering species, with its distinctive, fragrant, ship's propeller-shaped flowers, is now well known to enthusiasts, although in terms of discovery and introduction to cultivation it is one of the more recent species to come to our notice. Living specimens were introduced in 1956 by Dr Peter Davis and Oleg Polunin (nos 25368 and 25579) and these were at first regarded as the puzzling C. alpinum. However, Otto Schwarz eventually decided that this was not the case and in 1975 described the Davis/Polunin plants as C. trochopteranthum. The complete story is related by Kit Grey-Wilson in Cyclamen (13atsford, 1997) and need not be repeated in detail here. At first, it appears that the species is one of the more restricted in its range, at least compared with C. coum, C. cilicium, C. graecum, C. hederifolium, C. repandum and several others. The stated distribution is "confined to south-western Turkey, mainly between Mugla, Denizli and Egirdir, but also reported from one or two localities somewhat to the north of this general area" (Grey-Wilson, 1997).

Obviously, before a field study could take place it was necessary to find out as many definite localities as possible, otherwise one could spend two weeks actually hunting for populations on which to carry out surveys. An immediate problem presented itself, in that a check through herbarium specimens and literature records showed that there were only 19 recorded localities, and some of these were so close together that they could almost be disregarded as representing separate locations. The extremities of the known distribution showed that C. trochopteranthum occurs within an area some 180 miles long by 50 miles wide, making an area of exploration of some 9,000 square miles, in country which is mostly mountainous and hence rather difficult to cover in any great detail. A further complication is that the species has a fairly wide altitude range, from just above sea level to 1,650 m (5,400 ft)' and the flowering time is thus correspondingly wide, from February to late April. In fact it may start flowering even earlier than this: Trevor Wiltshire revisited one of the sites (Baba Dag in October 1997 and found that flower buds were already showing colour at that time. By its very nature, this field study requires that plants from as many altitude levels as possible are investigated, so the choice of time is a tricky one when, for various reasons, visits of only two weeks at a time are possible. It is quite clear that any survey under these circumstances will have shortcomings, but unless efforts are made there will be no progress at all in the study of cyclamen species.

In view of these facts, it was decided to travel late in the C. trochopteranthum season (mid to late April) to check some of the higher altitude populations, concentrating on the known sites but also trying to locate other, unrecorded, populations to build up a knowledge of its total distribution. Habitat details were also high on the agenda, since accurate information is lacking on the range of conditions under which the species may be found. The area selected was to the eastern end of the range, in the region to the south of Lake Egirdir, one of the higher altitude and less well known areas for the species.

In the event nature took a hand in the matter and, on arrival in Turkey, we found that it had been a mild winter, encouraging early flowering, followed by late snowfall just before our departure which covered the plants while still in bloom. The result was that, not only would they be hidden from view or at best very bedraggled on emerging from snow (which did prove to be the case!), it would also be difficult to travel in the mountains. We decided to change plans and head for the extreme south-west of its area in the Mugla and Marmaris areas, and work from lower altitudes upwards. Even if the plants were not in flower k would be of value for future surveys to have precise locations, and observations of leaf characteristics and habitat notes could be made.

Team and itinerary

The Cyclamen Society members taking part were Brian Mathew, Vic Aspland and Trevor Wiltshire. In Turkey it is against the national laws to collect any plants (including Seeds), and to undertake any form of research it is necessary to have a Research Visa, granted in advance from the Ministry of External Affairs in Ankara and arranged through the Turkish Embassy in London. One condition of this is that there should be a Turkish member in the appropriate field (in this case botany) attached to the group. Additionally, since cyclamen are listed on CITES, two further documents are required if living specimens are to be collected, one from Turkey to export the tubers and one from the UK to import them into the EC region. We are fortunate in that we have good relations with botanists in Istanbul and in Ankara and, not only were they very helpful in facilitating the research permits, it was agreed that local botanists from Istanbul would accompany us. These were Andy Byfield and Sema Atay, both of DHKD), the Turkish Society for the Protection of Nature, and they proved to be excellent and most helpful companions.

The overall itinerary, by vehicle from Istanbul and returning by air two weeks later from Antalya was: Istanbul-Bilecik-Kutahya-Denizli-Mugla-Antalya-Egirdir-Beysehir-Manavgat-Antalya.

Field studies

To gather data on the question of overall size, the main aim of the survey, leaf length and width and the diameter of the flowers were measured, taking as random a sample as possible in the given circumstances. The actual method for doing this had to be worked out on the spot since populations differed so much in their constitution, and on only one occasion was it possible to find the edges of a discrete population and take anything like a random line across it In most instances the populations were either so sparse that in order to acquire 50 samples of leaf and flower one had to measure nearly all individuals found, or else the terrain was such that a straight, random line could not be achieved. On several occasions the populations were so broken into an irregular pattern of clusters of "mini-populations" coupled with scattered individuals, that, rather ironically, one had to choose a random line in order to make sure of encountering enough specimens for the survey.

In addition to the actual measurements, notes were made on leaf pattern type, following a chart devised by Vic Aspland in advance, the flower colour, the width of the petals, their degree of twist and the colour of the "eye" at the base of the petals. Supplementary notes included such matters as fragrance, fimbriated petals, and any particularly striking or abnormal characteristics. The localities were noted to latitude and longitude using a GPS instrument, altitudes were recorded, and pH readings were taken using an accurate device giving a digital readout Habitat notes included aspect and steepness of slopes, amount of shade, overall soil type, rock formations, dominant vegetation type and any closely associated plant species.

Trevor Wiltshire, as the Society's photographer on this occasion, recorded habitats, plant associations, populations of cyclamen with their leaf and flower variations, and any individual plants of C. trochopteranthum which demonstrated a particular aspect of the work Herbarium specimens were collected from sites where the cyclamen were in flower, and a set of these will be deposited in Istanbul Herbarium ('SIX). Permission was obtained to collect up to 50 tubers with a view to comparing the behaviour of plants from low to high altitudes when brought together under standard conditions. In fact, only 23 plants were collected, largely a reflection of the poor weather conditions and consequential paucity of flowering colonies encountered. Photographs were taken in situ of any living plants collected, before their removal. For the surveys, using 50 individuals from each population, one flower and the largest leaf were measured on each individual.


On the question of overall size, on the basis of one season's survey, it is too early to say if there is any correlation between altitude and size of plant; these results can be presented only after assessment, at the end of the programme. Impressions are dangerous, but nevertheless interesting and one can make a few casual observations. Certainly the populations at the higher altitudes were often small-leaved, and some populations Seen at low altitudes had huge leaves, larger than any Seen in cultivation, but there was much variation at all sites and often considerable overlap in measurements between populations. Extreme examples are 97007T (from the Marmaris Peninsula at 75m altitude, 250 ft) where the leaves measured from 47mm long x 55mm wide up to 106mm long x 104mm wide. On the other hand, 97012T (from Camkyüsü at 1,650m, 5,400 ft) had leaves in the range 8mm long x 10mm wide to 38mm long x 38mm wide. Interesting as these at first appear, it must be pointed out that the habitat conditions were quite different and the plants were at a different stage of growth. The former population had long finished flowering, whereas the latter population was still in flower, and the plants of the former were generally in deeper shade. For the final assessment, only those populations which are at a comparable state of growth (ie in flower) will be considered.

Perhaps much more interesting at this stage were the observations on the habitat preference of C. trochopteranthum. Turkey can be divided, for phytogeographical purposes, into three broad vegetation types: the moist Euro Siberian region in the north, along the Black Sea and its adjacent mountains, the central plateau area (the Anatolian Plateau), with its extremely cold winters and steppe-like vegetation (known as the Irano-Turanian region), and the Mediterranean, characterised by autumn/winter/spring precipitation and warm, dry summers. This is a fairly crude, but useful, view of the situation; there are, for example small "enclaves" of Mediterranean vegetation on the Black Sea coast and, vice-versa, small areas of Euro-Siberian farther south in what is otherwise Mediterranean. Botanists familiar with the flora of Turkey are able, by and large, to place individual species into these phytogeographical groups and it soon became clear to us, by looking at plant associations, that C. trochopteranthum is a "Mediterranean" species; on the other hand, C. coum, which is generally considered to be its nearest relative, is clearly a "Euro-Siberian" species.

Travelling in C. trochopteranthum country, and this was particularly noticeable in the mountains to the south of Lake Egirdir, one could be amid various "Mediterranean element" plants, only to turn a corner and be confronted by a piece of "Euro-Siberia", with a quite different range of plants including violets, primroses and coltsfoot, all familiar European species. Even the atmosphere in these areas had a different feel, colder and damper, and the soil a heavy sticky greyish clay. In these habitats, C. trochopteranthum was never encountered. Around another corner and one could revert to the warmer Mediterranean climate with a change of plants again, and the team soon found that it was possible to predict where C. trochopteranthum would be likely to appear, and, similarly, where it would not Unfortunately, C. coum was not found in the areas of "Euro-Siberia", although that is not to say that it is not there!

The rock types and soil pH in which the species was found growing were strikingly variable, from limestone to serpentine and (possibly) sandstone, and from pH 6.8 to 10.1. In view of these somewhat unexpected findings it is proposed that more sampling of rock types will take place in future. One sample from the 1997 season, which seemed to be a particularly unusual rock type for the species, was sent to a geologist who prepared a microscope slide section and identified it as an oolitic limestone. The survey also extended the known range of the species in a north-easterly direction, between Lake Egirdir and Lake Beysehir, and put many more dots on the distribution map It appears that the easternmost boundary for C. trochopteranthum is the mountain range between these two lakes; it was found on the western slopes of this range, which runs roughly in a north-south direction, but on the eastern slopes was replaced by C. cilicium. Another useful aspect of the visit was that it proved to be a tidying up operation on some of the records. For example, an outlying locality cited for the species between Bilecik and Bozüyük, well to the north of the general area for C trochopteranthum, was visited and was found to contain (not unexpectedly) C. coum. Other records on the outer fringes of the distribution appear to represent, in the east, C. cilicium and, in the west, C. mirabile; such records must have been based on non-flowering dried material in herbaria.

Although the main aim of the survey has not yet been achieved, some very useful observations have been made and our overall knowledge of this attractive and very popular species has been enhanced.

Acknowledgements. We wish to express our thanks to the Turkish Government for granting us research permits for this project, and to Prof. Nerirnan Ozhatay (Istanbul University) and Prof. Tuna Ekirn (Gazi University, Ankara) for their invaluable assistance in the preparations. Special thanks go to Andy Byfield (Plant Conservation Officer Turkey, of Fauna & Flora International) and Sema Atay (Plant Section Co-ordinator of DHKD, the Turkish Society for the Protection of Nature). We are also grateful to several friends and colleagues who gave us advance information about sites in which they had seen the species during vacation visits. The 1997 survey was financed by The Cyclamen Society, together with grants for individual members of the team from the Royal Horticultural Society and the Alpine Garden Society.