Table of Contents

Since genetic and other kinds of diversity are not cvenly distributed within taxa, ecogeographical surveys need to be undertaken so as 1o ensure representative sampling for germplasm collections. Such surveys permit a synthesis of of the available information on the distribution and variation within a species or related group of species. The various types of information that can be of use in an ecogeographical survey are described, as well as the ways this information can be used in developing effective strategies for the conservation and ‘management of the wild genepools of the cultivated plants of Europe.

The evolution and speciati on within secl. Brassica can be eharae terized as allopatrie with weakly developed crossing barri ers. Crossing experimenls show Ihat the ehromosomal pairing in spee ies hybrids is high, indiealing that ali the speeies of the cytodcme have basicall y the same C genome. The relali ve ly wide intraspecifie geneli cal and morphological variati on is probably a result of local and regional differentiation. In most regions of the distribution area the specics consisl of spati ally well isolated populations and local adaptation has often given rise lo morphologica ll y dislinet populations as in, for instance, B. Cretica , B. rupestris, and B. villosa. Most populations are small, the reproductive capacity may fluctuate from one year to another and 'bottle-neck' situations may occur frequently. Results of crossing experiments and molecular in vesti galions lead to the eonelusion thal a li culti vatcd forms together with wild west European material belong to the same biological species, B. oleracea. Prospects for in situ conservation of wild populations are discussed.

Using enzyme polymorphism at 10 loci, genetic variation was analysed in both wild and cultivated olive growing around the Mediterranean Basin. Twenty five and 21 alleles were observed in the 333 wild olive trees analysed and in the cultived material (418 trees) respectively. Twenty one of these alleles were common to the two forms. In addition, in most the situations of sympatry between wild and cultivated olive, the two forms possessed the same alleles which differed from one area to the other, suggesting the occurrence of close genetic relationships between these two partners of the same species.

The paper reviews the present state of knowledge of genetic variation in the main native European forest trees species. Isozyrne tests have already shown that as a group these forest trees are extremely variable both in terrns of number of alleles per locus and the level of heterozygosity in populations. The maintenance of high levels of intrapopulation genetic variation in many of these trees is interpreted as a necessary strategy to survive in the heterogeneous environmental conditions that these long-lived plants fac e. Suggestions are offered how to preserve such vital genetic variation in forest tree populations in face of increasing environrnental stresses.

Gene flow is understood here as the movement of genes among sympatric or neighbouring populations of crops and related taxa or the introduction of genes from related taxa into the recipient genepool of a crop, ie. it describes the processes also known as introgression. We must differ between artificial or induced and spontaneously occuring gene flow. Atificial or human-caused gene flow has a rather long history in plant breeding and applied genetics and is documented by the efforts to use wild or weedy relatives of crops as donors of economically important characters. The successful breeding e.g. of disease resistant cultivars in many crop species by the transfer of resistance genes from wild taxa arc well-known examples. The spontaneous occurence of gene flow between crops and relatives however is a poorly understood phenomenon and has been described only anecdotically in most cases. There were gene flow processes between wild taxa and crops in the past which led to obvious evolutionary changes, ¢.g. to the formation of new species (hexaploid wheats). Actual gene flow between crops and relatives can be detected only by careful field studics. In general such gene flow is a rare event and can happen only in crops if their primary genepools includes related taxa and if crop and relatives are more or less sympatric. In Europe the possibility of gene flow between crops and relatives is greatly restricted, because European crops are mostly exotic plants from other regions of the w...

Wild stands of Populus nigra L. can be considered on the verge or extinction in a large part of western Europe. Its natural habitat is being gradually reduced by human activity and the gene pool is threatened by the large-scale presence of c ulti vated hybrids and of one ornamental .' cultivar, the fastigiate ' Lombardy poplar', that spontaneously hybridize with P. nigra. Although cultivars of P. nigra are cultivated in some areas, especially in the Middle East, the main use of this spec ies is as a parent of interspecifìc hybrids in breeding programmes. From an economic point or view the most important hybrid combination is produced by using P. nigra as a pollen parent and with the North American P. deltoides Bartr. as the seed parent, resulting in the common 'euramerican' hybrids (P. euramericana (Dode) Guinier). Some varieties of black poplar are al so widely used for their ornamental qualities. This represents an originaI ex ampIe of a wildlcultivated complex pool in rorestry. The conservation of in Situ genetic resources is limited to some restricted areas, such as the riparian fore sts, typical habitat or P. nigra, which have suffered considerably from urbanization, industri ali zation and competition with human productive activities. Ex situ conservation activities, on the other hand, have been carried out in several European countries. A project for the coordination of efforts airned at the conservation of P. nigra has been recently...

The importance of the study of demography and population genetic in the conservation of wild relatives of cultivated plants is stressed. The following aspeets are analyzed: variation or heterozigosity in small populations; genetic population dynamics and the importance of natural and artifieial seleetinn; life form and plant population structure; structuring in plant populations resulting from pollination and seed dispersal mechanisms; breeding systems and polyploidy and thier effects on heterozygosity variation.

Patterns of population differentiation in response 1o high selection pressures in adjacent environments are examined. In such situations distinct genetically based adapted populations can oceur only metres apart. Population structure in small scale spatially fluctuating, and temporally heterogeneous environments are also considered. Evidence suggests that in such environments adaptation is largely due to phenotypic plasticity, rather than (o fixed genotyopic differentiation.

Brassica olerocea L. belongs to the B. oleracea cytodeme with 2_n_ = 18. Biosystematic studi es show that the cylOdeme is composed of distinct species or groups of regionally distributed species. Wild populations of B. olerocea are distributed along the Atlantic coasts of Spain , France, and Great Britain , where they usually grow in steep parts of maritime c1iffs consisting of limestone or chalk. The size of the populations varies from one locality to another and a considerable intrapopulational vari ation is present in most of the populations regardless of size. Isozyme analysis indicates that even adjacent populations, geographically separated by a few kilometres only, may show quite different isozyme pattems. Thus, in most parts of the distribution area the species is represented by spalially well isolated populations and local adaptation has often given rise lo morphologically dislincl populations. The gene tlow between adjacenl populalions seems lO be smal.

Within the Brassica oleracea complex , nine species are recogni zed for Italy, ali represented in Sicily, seven of which are endemie to this Mediterranean island, Pantelleria and the Egades. Comments are made on the habitats oeeupied in Sicily by these species, their population size, the existence of natural hybridi zation between some pairs of species and the need for in_situ eonservation.

Over the last few years the Plant Breeding Unit of the J. N. J. A. in Madrid has centred its attention on collecting, maintaining and stud ying the genetic di versity or wild species of the tribe Triticeae (genera Aegilops, Agropyron, Elymus, Elytrigia, Hordelymus, Hordeum, Taeniatherum and Thinopyrum) growing in Spain. The pl ant material collec ted inc ludes samples or 609 natural populations belonging to 25 spec ies ()l' subspec ies, both annual and perennial. These reflect the presence and distribution or the different genera and specics in Spain. The present report covers aspects related to the distriblltion and collection of samplcs, and the analysis of lhe genelic variabilily of natural popllialions or lhese plant materials.

Marginai plant populations have low adaptative capacity. They show, however, a higher degree of morphological and genetic variation. This may be due to the fact that their genetic structure is under strong selection pressure.

This paper stresses the fact that different and contrasting genetic systems have evolved in the flowering plants, it also foc uses on some of the main elements that build these systems, particularly the level of ploidy and the various modes of reproduction (c ross-pollination , selfpollination and asex ual reproduction). Since each genetic system moulds the genetic architecture of populations in a different way, each poses specific demands when the planning of conservation is considered.

Many cultivated plants are polyploids and possess both polyploid and diploid wild relatives. Using as an example the polyploid complex of Dactylis glomerata L. (a grass species used partially as a fodder crop), it is shown that ploidy level should be taken into account in formulating strategies for the conservation of the wild relatives of cultivated plants. In association with other biological characteristics, more particularly the reproductive system, the “ploidy level will determine the extent of seed stocks, the minimum size of population for in situ conservation and the nature and degree of disturbance which can be tolerated by the wild populations.

The genetic structure of plant poplliations is deterrnined by severa I factors, from selection processes to histori ca l events (i.e. bottlenecks). Mating system is one of the most inportant factors due to its direct inOuence on thc way progcny genotypes are generated and on the reco!l1binational potential. In self-pollinating species, indi viduals are predo!l1inantly hO!l1ozygous and consequentl y effecti ve recombination is very low. As a consequence strong allelic disequilibri a arise among linked and lInlinked loci , in particular if se1ec tion is in volved, and hence mllitilocus assoc iations. The distribution of multilocus associations suggests that natural selec tion could be involved in their maintenance in some habitats, so they could represent coadaptated gene compl exes. Another criticai point for genetic variability and population structure is the polyploidy Ievel, because in self-pollinating polyploids the loci can be maintained throllghout generations in an heteroallelic state, which is phenotypycally equivalent to the heterozygote. ll1e ex ampie of Avena is taken to illustrate these points, and the consequences that the genetic structure ha ve when planning the collection and conservation of genet ic resources in Ihis species are discusscd.

After a presentation of recent studies on gene tlow and on pattems of genetie variation in pl ants, a brief evaluation is given of the effects of fragmentation of the landseape on the genetic variation in pianI species and the consequential effects or in breeding depression. Data show that reduced fitness in small and isolated populations oceurs, owing to reduced reproduetive success (Iower fruiI or seed set) and beeause of inbreeding. It is also suggested that the stili poorly studied mechanism of outbreedi ng depression may also have negati ve effects. In partieular, outbreeding speeies that used to be widespread and common, suffer more from fragmentation and isolation than endemie and naturally rare species or predominant selfers. More emphasis should be plaeed on the eonservation of new rare species, beeause their ecotypic variation and manifold , but often undiscovered local adaptations may get lost, and eonsequently also their long-term evolutionary perspectives.

Adaptive responses to environmental stresses at family, genus, species, and population levels are examined, with emphasis upon the consequences of natural selection effected by grazing, temperature, and drought.

The evidence obtained from the study of two polyploid complexes namely Dactylis and Lolium/Festuca shows that it is possible to enrich the gene pools of polyploids in these grasses by exchange of genes through crossing (even by intergeneric crosses), and to extend their adaptability to grow over a wider range of climatic conditions.

Under natural conditions, plant growth and spec ies composition of the vegetation are often limited by the availability of soiI nutrients rather than by li ght or water. This paper tries to answer the questione Are Ihere differences in the interactive effects of plant indi viduals, physico-chemical factors and the community of soil mi croorganisms that indi cate the development of different plant strategies for survival in nutrient-poor habitats? After a de finition of the effect plant roots may have on the surrounding soil (the rhizosphere effect), the paper reviews the ev idence for its mie on soil organi c matter mineralisation and therefore, of nutri ent ava ilability on different speeies and on soils of different fertility. C011lparison be tween annual and perennial species of Bromus, indicates a better efficiency of the former to thrive in nutrient-poor habitats, both in relati ve quanti ti es and in temlS of the cost of mining soil nutrients.

The management of the wild relatives of crop plants can only be undertaken in a limited number of cases. Clear objectives therefore have to be set when identifying priority species or conservation action. Numbers, sizes and proportions of populations to be conserved and their gerographic distribution are key questions when preparing management plans. Also important is an analysis of the threats to which the species are subjected. Conservation options include both in situ and ex situ and an integrated multidisciplinary approach is recommended.

Although in a rather primitive phase of modem plant breeding some breeders behaved like 'stamp collectors', the necessary variability for continuous progressin plant breeding has forced the search for new genes to solve new problems. Even though artificial mutation offers possibilities to find suitable variation, wild relatives of cultivated plants offer a different approach as they can be tested for specific traits in order to see whether nature had already created and tested the genes sought for. Gene manipulation by genetic engineering has emphasized even more the need for wild resource conservation, as any species can be the source of important genes for use in any other species. Besides, for modem breeders agroecological data are very important, and are nowincluded in germplasm collection studies. Amongst breeders there is agreement on the increasing need to preserve natural stands of wild species as it is impossible to collect and preserve ali the wild relatives required in plant breeding today. This leads to according an increasing importance toin situ conservation of natura I resources in plant breeding practice. Thus, if only because of professional needs, plant breeders have became strong advocates of natural resou rces conservation that is thus transformed into an economic activity.

The conservation of wild rel ati ves of crops is basically similar to that of wild species although there are important differences in detail. For wild species, conservation within their native habitats i.e. in situ is normally to be preferred, with ex situ approaches playing a complementary mie, for example to provide material for research or for raising plants for reiontmduction into the wild. An account is given of methods of coll ec ting, sampling, storing and characterizing germpl asm of wild plants.

Since the 1960's, genetic conservation of major crops has been based mainly on stati c ex si/u conservation. Whatever their efficiency, genebanks cannot solve the problem of the ' evolutionary freezing' of stored genetic resources. This paper presents one answer, based on the conservation of artificial genetically heterogeneous populations. Some results are presented on the adaptation of populations and the conservation of genetic variability obtained in a pilot programme of dynamic conservation of genetic resources of wheat. Although this experiment concerns a cultivated species, lessons are drawn that could be of interest for conservation experiments on wild species.

populations of rare and endangered plant species in Switzerland. Nature conservation falls principally within the responsil lity of the cantons. The legal measures of biodiversity preservation in Switzerland are good. However, the execution is often unsatisfactory. Offices for Nature Conservancy at federal, cantonal and communal levels try to perform and supervise the legal requirements, but are often overloaded both in terms of personnel and finance. Private institutions that own many conservation areas play an important part in the management of endangered species’ populations. The creation of ‘ccological compensation arcas’ is financially promoted by the Confederation. These agricultural areas are managed without pesticides and fertilizers in order to obtain vegetation with higher diversity. The Swiss Committee for the Preservation of Wild Plants is a private institution supported by the Confederation to initiate, coordinate and survey all actions in Switzerland dealing with the preservation of rare and endangered plants. It publishes guidelines and information about the optimal management of the various species’ populations. Extensive inventories of the most endangered plant species are performed and the results, together with instructions for successful management of the populations, are forwarded to the Conservancy’s institutions.

Wild emmer wheat (Triticum turgidum var. dicoccoides) in Israel is used as an examp1e of how the various roles of research and conservation and management practice may be detemlined in the conservation and management of wild relatives of cultivated plants. An account is given of the govemment-commissioned Ammiad project for the 'Dynamic conservation of the Wild Wheat in Israel ', its prospects and the Iessons that can be drawn from il.

An evaluation of the role played by protected areas in the conservation of the wild relatives of cultivated plants in Ttaly and Sicily has been carried out, based mainly on bibliographic sources. Of the 163 species and subspecies represented in the whole Italian territory, 147 oceur inside protected areas. The figures for Sicily are 113 and 104, respectively. These data arc evaluated in relation to the present number and extension of protected arcas and with their possible increase.

The traditional mIe of managers has been to restore or maintain a certain state or the protected ecosystem, implying that the conservation goal of ali protected areas is to secure the 'balance of nature'. It is now recognized that ecosystems undergo non-equilibrium dynamics. Rather than strive to restore 'balance' , conservation goals should be attainable, and specifically moulded for each protected area. Often areas are not allocated for allaining predetemlined conservation goals, but goals are 'filled ' lo already selected areas. The first role of the manager is to set attainable goals to the protected area. To do this, it is necessary to survey the current and the 'potential' inventory of species and processes, and to evaluate the findings by checking them against existing conservation programmes and their state of implementation. Natural ' assets' of the protected area are then identified, and become the objects of the conservation goals. The subsequent role is to prepare a management programme, that by executing it, conservation goals will be allained. Preparation includes tapping current relevant scientific knowledge, formulating a management hypothesis, outlining an action pIan that comprises conservation activities optimized to accommodate a public relations component, a monitoring protocol for periodically evaluating the success of the programme, a managementoriented research programme, a timetable, and a budget. The third role of the manager is to exe...

Large colleclions of wild relatives of crop plant have been established al the Gatersleben genebank as a consequence or the growing demands of brced ing research and plant hreed ing. A spec ific approach to such collec tions was necessary, reviewing and developing the possibil ities of genebanks. Thc first 'monographi c trea tment of wild plant collec ti ons' desc ribing the genus Aegilops from the perspective or genebank aspecls appea red in 1980. Other examples that mceti breeders' demands are Hordeum and Secale. Two special cases are deseribed for Datura, with a collection stemming from biochemical and taxonomic-,orphological research, and Agrostemma, a weed showing convergent evolution and with economic possibilities, but endangered by developing agricultural techniques and methods. Once in the genebank the material has lo be evaluated to allow more effect ve use in plant breeding.

The conservation work on wild species by the Royal Botanic Gardens, Kew, Seed Bank is described with particular emphasis on the differences between seed bank procedures for wild species and those for crops.

‘The conservation of plant genetic resources in Switzerland is carried out by public and private organizations. The most important genebank is located at the Federal Agricultural Research Station of Changins, where more than 10 000 accessions of different species are conserved. A commission has been created with representatives of public and private organisations concemed with plant genetic resources in order to stimulate collaboration and coordination as e.g. making inventories and the development of common projects. Switzerland has intriduced a number of legal instruments for the protection of biological diversity, not only in the framewaork of the protection of nature, but also in agriculture.

Conservation of wild relatives of cultivated plants refers to complementary strategies termed in situ and ex situ. The implementation of in si tu conservation programmes for these species requires a range of technical and administrative interventions that can be grouped under four main points: Collection and interpretation of data on the biology and the ecological requirements of the species coneemed. Definition of the purpose of the conservation action, chosing between the followingdifferent methods: I) maintain the species in the long run and abandon, if neccessary, the populations which are the most marginai and the most complex or the most expensive to preserve; 2) maintain particular populations that are representative of the genetic variability and, notably, tlìose that are representative of very typical, indeed even extreme situations; or 3) maintain a specific loeal population for local human uses. • Implementation of measures for the protection and management of the habitat of the above mentioned species. Evaluation of the cost of conservation. To implement such effective programmes of in situ conservation, it is necessary to bring together partners (scientists, breeders area managers) who are not accustomed to working together. The existence or intermediate structurcs capablc to discuss with ali these partners Illay certainly help to initiate such projects and bring them IO successo In France, the National Botanical Conservatories (Conserva...

Amongst the aims of the Alpine National Botanical Conservatory of Gap-Charance is the protection of the wild flora of the Alps. Il faced several problems when trying to elaborate its strategies for management and conservation. The example of Rosa gallica L., a poorly known protected spec ies, is quite significant. Further studies are underway to gain a better understanding or its seed dispersal, the genetic diversity within it and its population strcuture.

Wild relatives of our cultivated crops are an important source for the genetic improvement of these crops This paper describes the presence of wild relatives in the collections of Centre for Genetic Resources, the Netherlands (CGN). It shows that the collections differ substantiall y in the presence of wild relatives. It is demonstrated that former working collections of breeders are often dominated by cultivated materia!. This paper brieny reviews the utili zation of wild relatives of Solanum and Lactuca in breeding. Il is emphasized thal the evaluation of wild relatives for certain properties is a prerequisite for the utili zation of this materia!. Evaluation of wild relatives will also reveal the geographical distribution of specific characters in the originai collecting sites. Examples of the geographical distribution of wild potato species with resistance to economie pests and diseases are presented.

The multidisciplinary nature of the conservation of the wild relatives of crops requires considerable efforts of organization so that the different partners in volved can work effeclively logelher. Slake holders need to be identified and some body or organization is mandaled lo ensure coordinalion. Il is also important in view of lhe enduring nature of conservation is to ensure a long-term commilment from those responsible.

Forest ecosystems are important components of the European natural environment and provide substantial services and benefits to human society. They are under severe threat due to pollution and human-induced effects; and these pose severe challenges for the adaptation and survival of tree populations. The special nature of tree poplations is considered, and the need to protect adequate amounts of genetic variability in them is discussed. Both in situ and ex situ approaches play a role in conserving genetic variation of forest tree populations, with the former given high priority.

The manifold and increasing human influences on the genetic structures of tree populations make it necessary to devise measures for conservation of forest genetic resources. After presenting a definition of gene conservation, the various objectives of forest tree conservation are considered. Because forest trees are long-lived plants and still at a relatively low level of domestication, the conservation of the genetic adaptability of their populations to environmental changes should be emphasized. This implies that the choice among candidate populations for conservation should be based on the results of surveys of genotypes at marker gene loci. In view of the expense involved, and the long-term importance of this step, the quality of gene markers and the information on their mode of inheritance becomes critical. Besides marker gene loci, the expression of genetically controlled adaptive phenotypic traits deserves proper weight. The selection criteria are then estimates of certain parameters, such as the adaptive potential, the genetic meaning of which must be made transparent. Methods of dynamic conservation should be given preference because of the environmental situation of tree populations. Conservation in situ requires high sylvicultural skills. The combination of these prodecures with conventional static conservation ex situ induces flexibility and is very promising. Problems exist as to how to sample populations and how to measure their variation. Current research n...

Since a genetic resource is a col1ection of biological material containing either specific or particularly variable genetic variation, severa l cases can be found in European and Mediten'anean pine species that meet these requirements and might be included in gene conservation programmes. On the basis of their geographic distribution, European forest trees may show different pattems of variation among and within populations, as is il)ferred from isozyme markers: they may present several examples of populations carrying specific genetic variation that can therefore be looked upon as genetic resources. Other genetic markers (morphological, physiological and biochemical) are al so useful to indicate existing differentiation among populations and provenances. In this paper, some significant examples are given for the following species: Pinus sylveslris L., Pinus leucodermis Ant., Pinus nigra Am., Pinus cembra L., Pinus halepensis Mill. For each of them the proposed form of preservation is mainly the dynamic in situ conservation, which makes it possible to preserve the genetic adaptability of populations in changing environments.

Adaptability of forest ecosystems is a common goal of forest management and nature protection. The adaptive potenti al of long-lived forest tree poplilations depends on the amount of their genetic variability. The management of populations under threat (e.g. by air pollution) has to integrate strategies for the protection or if necessary, for the increase of their genetic variability. A case study on severely threatened silver fir (Abies alba Mill.) in the Ore Mountains illustrates thc research ac ti vities which are necessary for a geneti c management of the residual populations. Within a joint project, a comprehensive investigation on abundance and vitality as well as on the genetic constitution of residual silver fir provided basic results for decisionmaking in the fields or fore st management and nature protection. The resulls of the genetic investigations are the basis for in situ and ex situ conservation strategies.

The European beech is one of the most important broadleaved forest lree species in Europe occun'ing in vanous ecological conditions. After the establishment of numerolls provenance experirnents in this species (the last one containing 188 provenances and 23 trials), extensive genetic inventories in most part of ElIrope has been carri ed out. In situ and ex situ gene conservation in European beech is outlined.

Allozyme, DNA markers, sequence data and monoterpenes have been used to clarify the genetic and taxonomic relationships between different populations of Abies alba Mill. and the only existing population of Abies nebrodensis (Lojac.) Mattei , a relict fir restricted to a single locality in Sicily. A high level of differentiation was found using ali the different approaches. The genetic distance between A. nebrodensis and the group of A. alba populations was found to be much greater than that among A. alba populations located more than 1000 km apart. The results provide support for the considering A. nebrodensis and A. alba as distinct taxonomic entities.

The paper surveys the needs for protection of forest trees native to Europe, and the recommendations made and the actions taken by the European countries both in regard to in situ and ex situ conservation of genelic resources of forest trees. Populus nigra L. is presented as a model and a test case for orientation.

A survey is given of the European wild species of Prunus s. lato, their classification and relationship to the fruit crops of the genus. Most important in this respect are the subgenera Prunus and Cerasus. A lack of knowledge of the infraspecific diversity of the more widespread and variable wild taxa is emphasized and consequently our lack of knowledge with any degree of certainty, of the evolutionary pathways from the wild relatives to the cultivated taxa of the genus. The wild relatives are valuable genetic resources for future breeding programmes, and their natural populations need in many cases to be protected by in situ conservation projects.

The principal wild progenitors of the cultivated apples and pears grown in Europe and west Asia are the crab appIe (Malus sylveslris (L.) Miller, and wild Pyrus communis L. subsp. communis = P. pyraster Burgst.). Both are widely distributed over Europe and southwest Asia. The wild primary gene pools of these fruit crops are, however, much wider. They include most (and perhaps ali) of the congeneric wild relatives of these crops.

After a brief account of the taxonomy, ecology and present-day distribution of the native population of Abies nebrodensis (Lojac.) Mattei, the authors suggest a protection pIan comprising both in sifu and ex SifU actions.

This paper presents the results of a study on the reproductive cycle of Abies pinsapo. The main reproductive aspects which affect the quantity and quality of the seeds, such as the low pollen dispersal capacity, the subdioecious condition and the combination growth forrn -tree architecture are diseussed. The requirements f(lr seed germination, seedling reerui tment and some aspects of the species are analysed using all the available data.

The taxonolllY, distribution and ecology of the polYlllorphic Capparis spinosa L. in Sicily are discussed. C. spinosa subsp. rupeslris (Sibth: & Sm.) Nylllan , a leathery-Ieaved, pendulousbranched shrub is widespread on carbonate, volcanic and gypsaceous outcrops. Subsp: spinosa, a thomy chamaephyte or hemicryptophyte with creeping branches, is mainly distributed in south -central parts of Sicily, on clay and natric soils under xerophilous conditions. Attention is drawn to a remarkable environment-induced variation within the latter subspecies. The heterogeneity of the cultivated forms of subsp. rupestris on the island of Pantelleria (Strait of Sicily) appears to be linked to crossing and introgression with wild populations.