INDEX
Introduction ......................................................................................................................................................... pag. 1
Late glacial and Holocene history of Buxus sempervirens L. in the Italian Peninsula ....................................... pag. 11
Buxus in Europe: Late Quaternary dynamics and modern vulnerability ............................................................ pag. 25
Genetic structure and taxonomical boundaries in the Western Palaearctic Buxus species ................................. pag. 34
Chloroplast variation in two closely related woody taxa showing contrasting histories and distributions ........ pag. 46
Conclusions ......................................................................................................................................................... pag. 66
INTRODUCTION
Buxus L. is a genus of woody species which belongs to the family Buxacae. This genus has two
centres of diversification (Fig. 1), one in the Caribbean-Latin American area (sect. Tricera) and one
in Eurasia (sect. Buxus), with approximately 50 and 40 species, respectively. Furthermore, a
comparatively small number of 7-8 relictual species, occurs in sub-equatorial and southern tropical
Africa, including Madagascar (sect. Probuxus; Khöler and Brückner 1989). This genus is well
known throughout the world for its ornamental value, with some 200 cultivars and hybrids being
used for topiary, hedging and landscaping.
Fig. 1 Geographic distribution and leaf venation pattern of the genus Buxus (from Köhler and Brückner
1989). 1 B. portoricensis, 2 B. citrifolia, 3 B. acuminata, 4 B. macrophylla, 5 B. mexicana, 6 B. pubescens,
7 B. crassifolia, 8 B. rotundifolia, 9 B. revoluta, 10 B. foliosa, 11 B. vaccinioides, 12 B. cubana,
13 B. hildebrandtii, 14 B. madagascarica, 15 B. macowani, 16 B. balearica, 17 B. sempervirens,
18 B. hyrcana, 19 B. papillosa, 20 B. wallichiana, 21 B. rugulosa, 22 B. microphylla subsp. sinica,
23 B. harlandi, 24 B. microphylla subsp. sinica var. aemulans, 25 B. megistophylla, 26 B. rivularis,
27 B. rolfei, 28 B. cochinchinensis, 29 B. rupicola.
The Eurasiatic sect. Buxus is represented in the Western Palaearctic area with a group of closely
related species distributed in temperate and Mediterranean Europe, north-western Africa, Middle
East and Caucasia. Fossil record of Buxus in these areas show a history of changes in diversity,
with extinction of tropical and subtropical elements, and evolution of temperate and ultimately
Mediterranean taxa (Fig. 2; Kvaček et al. 1982, Bessedik 1983). Heritage of a long history, the
ecological plasticity of Western Palaearctic Buxus has played a major role in the persistence of the
genus during complex geological and climatic events that affected the area. Through adaptations,
two major lineages coped with the Western Palaearctic palaeogeographic and climatic changes.
This lineages include the Mediterranean, more conservative species, and the temperate, more
advanced taxa (Kvaček et al. 1982). These two lineages are quite distinct from a morphologic and
genetic point of view (von Balthazar et al. 2000, von Balthazar and Endress 2002). However, the
large range of variation characters that discriminate the species within each group created some
1
taxonomical problems (Davis 1982, Guseinova 1996, Sonboli et al. 2004). The taxonomic issue is
also complicated by the disjunct phytogeographic pattern, (Fig. 2) with one lineage showing EuroSiberian and the other Mediterranean disjunct ranges (Davis and Hedge 1971).
Fig. 2 Distribution of the genus Buxus L. in the Western Palaearctic area and a fruit remain of an extinct
tropical taxa from the Miocene of central Europe (Kvaček et al. 1982)
During the last decades, the temperate woody flora of the Western Palaearctic area has been the
subject of great phytogeographic and phylogenetic interest. Enormous efforts were devoted to
detect the effects of the Quaternary climatic oscillations on the distribution of temperate trees and
their genealogical lineages, with special consideration for the imprint of the last glacial maximum.
The location and survival of tree taxa in the coldest stages of the last full glacial period has
traditionally favoured a southerly refugial model, that implies survival in the Mediterranean
Peninsulas and re-colonization of central and northern ranges during the Holocene
(Bennett et al. 1991). This view is based on the interpretation of palaeobotanical records, with fossil
data suggesting that areas where each tree species first appear during the post-glacial should broadly
correspond to the glacial refugium for that taxon (Huntley and Birks 1983). Comparing the fossil
records of woody species across wide areas showed that most temperate trees were located in the
southern Peninsulas of the Western Palaearctic area during the last glacial period and achieved their
modern ranges by dispersal from these long-term refuge areas. Nevertheless, recent investigations
on macrofossil material provided strong evidence that coniferous as well as some broadleaved trees
were continuously present in central Europe throughout the last glacial period (see Willis and van
Andel 2004 and references therein). A preamble of this scenario was evidenced already by pollen
analyses, with the understanding that the progressive appearance of tree species in fossil records
may not indicate the movement from a site to another, but rather a population increase to a critical
local density above which the taxon is detectable by pollen-counts (Birks 1989).
Together with increasing fossil findings, genetic studies provided independent and
complementary evidence of Quaternary climatic oscillations effects in shaping the distribution of
woody taxa and their genetic diversity over the Western Palaearctic area. Molecular markers
became a fundamental tool to detect long-term refuge areas, infer colonization histories, understand
2
the role of hybridization and to study the history of formation of species. Phylogeography is a
discipline based on the geographic distribution of genealogical lineages whose phylogenetic
relationships are know or can be estimated. It forms the bridge within several micro- and macroevolutionary disciplines (Avise et al. 1987). The development of novel markers targeting
mitochondrial (mtDNA) or plastidial (cpDNA) genomes allowed tracing matrilineal and patrilinear
phylogenies, as these markers are generally uniparentally inherited. In angiosperm trees, plastids are
generally maternally inherited and therefore moved by seeds only. Associated with the concept that
tree taxa were located in the southern Peninsulas during the last glacial maximum and ‘moved’
northward during the Holocene, this characteristic triggered a large number of phylogeographic
studies on woody species. The availability of several intraspecific studies allowed comparisons of
tree species phylogeographic patterns over the same areas. Comparative phylogeographic studies
evidenced some common genetic patterns that endorsed the predominant role of Mediterranean
Peninsulas as long-term refuge areas for woody taxa (Taberlet et al. 1998, Petit et al. 2003). They
also documented the major contribution in the re-colonization of Central and Northern Europe of
the tree populations located in the southern European territories (Fig. 3). Like fossil studies
previously, a close scrutiny of some genetic patterns indicate that some tree populations may have
persisted the last glacial period outside the southern refugia (see Bhagwat and Willis and references
therein).
Fig. 1 Left: Examples of Mediterranean refugia, main post-glacial colonisation routes and subsequent
suture zones in Europe (Taberlet et al. 1998). Right: Multispecies genetic divergence for 25 European
forests. Higher than average values are in black circles, lower than average are in white circles, and circle
diameter is proportional to the difference from the mean value. The level of divergence with each of the
five nearest forests was represented by connecting lines, with continuous black lines indicating
comparatively high divergence, dotted lines, intermediate divergence and continuous gray lines, low
divergence (Petit et al. 2003).
Increasing fossil and genetic evidence for northern survival stimulated the study of mechanisms
that allowed persistence (Bhagwat and Willis 2008) and initiated innovative investigations on past
population dynamics. Magri (2008) examined the varying modes and rates of population increase
and decline in Fagus sylvatica L. and analysed the relationships between genetic and demographic
resources. It may now be of interest also to relate the modern distribution of tree species with their
past demographic histories, and relate these processes to the distribution of genetic variation. The
comparative analyses of modern and past distributions, palaeodemographic processes and genetic
3
analyses might concur to pursuit several scopes, such as: i) defining where and at which density a
tree taxa persisted throughout the last glacial period; ii) tracking the evolution of range disjunction
and areal fragmentation; iii) understanding the growth and decline of tree populations in space and
time; iv) establishing conservation actions in areas where woody taxa are historically at risk or
when their genetic resources require special attentions. These multidisciplinary studies contribute to
the practical application of palaeoecology, genetics and phytogeography to long-term biodiversity
maintenance, ecosystem naturalness, conservation evaluation, and response to changing disturbance
regimes (Willis and Birks 2006).
Range dynamics of temperate tree populations during Quaternary climatic oscillations find their
counterpart of biogeographical interest in the disjunction of Mediterranean woody taxa. The history
of Mediterranean tree species is conceived as more complicated than that of those which were
primarily affected by the last glacial period (see Thompson 2005 and references therein). In facts
the cold stages of the Quaternary affected only marginally the distribution of strict Mediterranean
trees. On the contrary, pre-Quaternary palaeogeographic changes and major eustatic sea level
modifications have been responsible for observed distribution and genetic patterns (Magri et al.
2007, Mansion et al. 2008). Different processes and timings affected the two Mediterranean
domains since the Eocene. The western part (including the Tyrrhenian, Ligurian, Provencal,
Algerian, and Alboran basins) started to form during the Oligocene in an overall convergence
motion between African and Eurasian plates and through continental margin break-up events
(Rosenbaum et al. 2002). The eastern part (including the Ionian, Herodotus, and Levant basins)
represents remnants of the Early Mesozoic Neotethys Ocean (Garfunkel 2004). Thus, the
distribution of woody taxa in Mediterranean basin show a profound phytogeographic and genetic
differentiation that is more significantly related to a longitudinal east-west divide rather than a
latitudinal split in northern-southern territories.
Disjunctions in the distribution range of species have fascinated biologists since they were first
detected. Their interpretation has long been regarded as one of the central and most debated
problem in phytogeography (Raven 1972, Thorne 1972, Axelrod 1975). Dispersal across preexisting barriers and vicariance through fragmentation and isolating events have been often
contrasted as competing processes primarily responsible for these biological disjunctions (e.g. Stace
1982). However, few work has been devoted to assess the phylogeographic pattern of
Mediterranean trees showing intraspecific disjunctions and to infer the underlying historical causes.
Disjunct distribution patterns were investigated in some strict Mediterranean woody taxa (Khadari
et al. 2005, Breton et al. 2006, Rodríguez-Sánchez et al. 2009, Migliore et al. 2011). Results from
these studies reported contrasting molecular differentiation between western and eastern
Mediterranean populations.
The above researches induced me to carry out a multidisciplinary research on the distribution,
history and evolution of the Western Palaearctic Buxus species. This group of woody taxa is one of
the tree genera with almost no genetic information (Rosselló et al. 2007). The postglacial history of
Buxus has been studied by some authors (Wegmüller 1984, Lang 1992, Yll et al. 1997). However,
the wealth of recently published modern distribution data and fossil material encourage a new
review.
The genus Buxus L. is not only important for bio-ecological sciences, but also for cancer
research (Ait-Mohamed et al. 2011), chemistry (ur-Rhaman et al. 1991), and material science (Holz
1996). Moreover, Buxus shows a long-lasting link with human culture, as witnessed by its role in
ancient civilisation as a symbol of eternity, renewal and vigour, being found in burial sites and
coffins (Allison 1947). It is also represented in ancient artwork (Caneva and Bohuny 2003),
mentioned in sacred books (Holy Bible, Isaiah 41:19) and early encyclopaedias (Naturalis Historia,
Pliny the Elder, Liber XVI). The wood of Buxus received very early attention because of its
hardiness and quality. It is the most ancient material ever used for writing-boards ever recovered
(Ulu Burun diptych, 14th century B.C.; Pendleton and Warnock 1990), and for building tools such
as hammerheads, nails, and wooden joints (Fell 1998). The particular sound properties of boxwood
4
make it suitable for crafting high quality musical instruments. Its quality seems to be timeless and it
made the fortune of landowners and entire villages. For example, after imports from turkey ended
(John Huntley, personal communication), the Huntley family at Boxwell Court, Glouchestershire
(UK) became the primary supplier of boxwood in England during the 17th century. A second
example is the village of Aiguines, Var (FR), which in the late 19th century became the Europan
center for the production of boxwood bowling balls (Wallet 1983).
Last but not least, Buxus was well know since the Roman times, through the Middle Ages and
Renaissance periods up to the present day being the model plant for the Ars Topiaria. Nowadays
more than 200 Buxus cultivars are known (Batdorf 2005).
The aims of the present study can be summarized as follows:

Which is the distribution of Buxus in the Western Palaearctic area? Molecular surveys as
well as fossil reviews require careful consideration for the frequency and abundance of
Buxus populations over the area. Data from online vegetation databases, Herbaria, Floras
and field surveys encourage an attempt to depict a detailed scenario.

Is the modern distribution of Buxus the result of a post-glacial migration from limited
glacial refugia? Buxus is often considered a thermophilous (sub-)Mediterranean woody
taxon. From the refugia in the Mediterranean regions and in the Southern and Western
Alps Buxus might have recolonized the northern parts of its modern distribution area
(Wegmüller 1984, Lang 1992). The abundance of recently published fossil data of Buxus
may contribute defining times and modes of its postglacial expansion.

Are there regional differences in the Holocene dynamics of Buxus populations in Europe?
Buxus is presently distributed in both Central and Southern Europe. The combined use of
modern and past distribution data may highlight and explain similarities and
dissimilarities in the behaviour of populations located in different territories of the
Western Palaearctic area

Which are the different roles played by natural population dynamics, climate change and
human impact in shaping the European distribution of Buxus in the last 15,000 years?
Considering that the modern distribution of plant species is the result of a combination of
factors, including location of glacial refugia and ecological responses to post-glacial
climate changes and human activities, it may be of interest to evaluate which factor was
especially influential on Buxus.

Is Buxus a vulnerable genus in Europe? While B. balearica is considered vulnerable in
some parts of its distribution range (Blanca 1999), no conservation actions focus on
B. sempervirens, apart from weak regulations at local or regional levels in Central Europe.
The post-glacial history of Buxus may help distinguish areas where the populations are
vigorous and do not require any specific conservation action from areas where they are
weak and demand special attention due to ongoing climate change and increasing human
impact.

How many Buxus species are found in the Western Palaearctic area? The taxonomic
framework of Buxus has long been the subject of controversies (Davis 1982, Guseinova
1996, Sonboli et al. 2004). The characters used to differentiate species are subject to a
wide range of variation in response to the high ecological plasticity of the genus. Novel
DNA fingerprinting techniques may help disentangling this taxonomical problem.
5

Does the phylogeographic pattern of Buxus shows some commonalities with other woody
taxa of the Western Palaearctic area? Comparative phylogeographic studies on temperate
European trees revealed common patterns in the organization of genetic diversity. These
patterns were related to shifts in distribution ranges that occurred in response to
Quaternary climatic cycles and to post-glacial recolonization process (Petit et al. 2003).
The lack of data fosters a molecular survey to study the effects of past climatic changes
on the genetic diversity of Buxus.

Does the molecular differentiation of Buxus mirrors the extreme disjunction of Western
and Eastern Mediterranean populations? Empirical data drawn from many sources would
predict the existence of a genetic break between the two Mediterranean domains. It might
be of concern to shed light on the origin of the disjunct distribution and to interpret the
results in light of Mediterranean palaeogeography.

Does the comparison of genetic and palaeobotanic data provides new insights on the
historical causes responsible for the distribution of genealogical lineages of Buxus in the
Western Palaearctic area? Recent advances in phylogeographic inferences showed the
importance of past population demography in shaping the modern genetic resources of
plant species. However, demographic models usually rely on genetic data. Thus, it might
be desirable to evaluate if independent palaeodemographic evidence, based on thorough
reviews of fossil data, can be used to explain current patterns of genetic diversity.
6
REFERENCES
Ait-Mohamed, O., Battisti, V., Joliot, V., Fritsch, L., Pontis, J., Medjkane, S., Redeuilh, C.,
Lamouri, A., Fahy, C., Rholam, M., Atmani, D., Ait-Si-Ali, S. (2011) Acetonic extract of Buxus
sempervirens induces cell cycle arrest, apoptosis and autophagy in breast cancer cells. PLoS
One, 6, 1-11.
Allison, J. (1947) Buxus sempervirens in a late Roman burial in Berkshire. New Phytologist, 46,
122.
Avise, J.C., Arnold, J., Ball, R.M., Bermingham, E., Lamb, T., Neigel, J.E., Reeb, C.A., Saunders,
N.C. (1987) Intraspecific phylogeography: the mitochondrial DNA bridge between population
genetics and systematics. Annual Review of Ecology and Systematics, 18, 489-522.
Axelrod, D. I. (1975) Evolution and biogeography of Madrean-Tethyan sclerophyll vegetation.
Annals of the Missouri Botanical Garden, 62, 280-334.
Batdorf, L.R. (2005) International Registration List of Cultivated Buxus L. The Boxwood Bullettin,
45, 98-109.
Bennett, K.D., Tzedakis, P.C., Willis, K.J. (1991) Quaternary refugia of north European trees.
Journal of Biogeography, 18, 103-115.
Bessedik, M. (1983) Le genre Buxus L. (Nagyipollis KEDVES 1962) au Tertiaire en Europe
occidentale: évolution et implications paléogaographiques. Pollen et Spores, 25, 461-486.
Bhagwat, S., Willis, K.J. (2008) Species persistence in northerly glacial refugia of Europe: a matter
of chance or biogeographical traits? Journal of Biogeography, 35, 464-482.
Birks, H.J.B. (1989) Holocene Isochrone Maps and Patterns of Tree-Spreading in the British Isles.
Journal of Biogeography, 16, 503-540.
Blanca, G., (1999) Libro Rojo de la flora silvestre amenazada de Andalucía. Consejería de Medio
Ambiente, Junta de Andalucía, Sevilla, Spain.
Breton, C., Tersac, M. , Bervillé, A. (2006) Genetic diversity and gene flow between the wild olive
(oleaster, Olea europaea L.) and the olive: several Plio-Pleistocene refuge zones in the
Mediterranean basin suggested by simple sequence repeats analysis. Journal of Biogeography,
33, 1916-1928.
Caneva, G., Bohuny, L. (2003) Botanical analysis of Livia’s villa painted flora (Prima Porta, Roma).
Journal of Cultural Heritage, 4, 149-155.
Davis, P.H. (1982) Flora of Turkey and the East Aegean Islands. Edinburgh, Edinburgh University
Press.
Davis, P.H., Hedge, I.C. (1971) Floristic links between N.W. Africa and S.W. Asia. Annalen Des
Naturhistorischen Museums in Wien, 75, 43-57.
Fell, V. (1998) Iron Age ferrous hammerheads from Britain. Oxford Journal of Archaeology, 17,
207-225.
7
Garfunkel, Z. (2004) Origin of the Eastern Mediterranean Basin: A reevaluation. Tectonophysics,
391, 11-34.
Guseinova, S.O. (1996) O semennom razmnozhenii i taksonomicheskom range Buxus hyrcana
(Buxaceae). Botanicheskii Zhurnal (Moscow and Leningrad), 81, 49-59.
Holz, D. (1996) Acoustically important properties of xylophone-bar materials: can tropical woods
be replaced by European species? Acustica/Acta acustica, 82, 878-884.
Huntley, B., Birks, H.J.B. (1983) An atlas of past and present pollen maps for Europe: 0–13 000
years ago. Cambridge University Press, Cambridge.
Khadari, B., Grout, C., Santoni, S., Kjellberg, F. (2005) Contrasted genetic diversity and
differentiation among Mediterranean populations of Ficus carica L.: A study using mtDNA
RFLP. Genetic Resources and Crop Evolution, 52, 97-109.
Köhler, E., Brückner, P. (1989) The genus Buxus (Buxaceae): aspects of its differentiation in space
and time. Plant Sysematics and Evolution, 162, 267-383.
Kvaček, Z., Bůžek, Č., Holý , F. (1982) Review of Buxus fossils and a new large-leaved species
from the Miocene of Central Europe. Review of Palaeobotany and Palynology, 37, 361-394.
Lang, G. (1992) Some aspects of European late- and post-glacial flora history. Acta Botanica
Fennica, 144, 1-17.
Magri, D. (2008) Patterns of post-glacial spread and the extent of glacial refugia of European beech
(Fagus sylvatica). Journal of Biogeography, 35, 450-463.
Magri, D., Fineschi, S., Bellarosa, R., Buonamici, A., Sebastiani, F., Schirone, B., Simeone, M.C.,
Vendramin, G.G. (2007) The distribution of Quercus suber chloroplast haplotypes matches the
palaeogeographical history of the western Mediterranean. Molecular Ecology, 16, 5259-5266.
Mansion, G., Rosenbaum, G., Schoenenberger, N., Bacchetta, G., Rosselló, J.A., Conti, E. (2008)
Phylogenetic analysis informed by geological history supports multiple, sequential invasions of
the Mediterranean Basin by the angiosperm family Araceae. Systematic Botany, 57, 269-285.
Migliore, J., Baumel, A., Juin, M., Médail, F. (2011) From Mediterranean shores to central Saharan
mountains: key phylogeographical insights from the genus Myrtus. Journal of Biogeography, 39,
942-956.
Pendleton, M., Warnock, P. (1990) Scanning electron microscope aided wood identification of a
Bronze Age diptych. IAWA Bulletin, 11, 255-260.
Petit, R.J., Aguinagalde, I., de Beaulieu, J.-L., Bittkau, C., Brewer, S., Cheddadi, R., Ennos, R.,
Fineschi, S., Grivet, D., Lascoux, M., Mohanty, A., Müller-Starck, G., Demesure-Musch, B.,
Palmé, A., Martín, J.P., Rendell, S., Vendramin, G.G. (2003) Glacial Refugia: Hotspots but not
melting pots of genetic diversity. Science, 300, 1563-1565.
Petit, R.J., Brewer, S., Bordács, S., Burg, K., Cheddadi, R., Coart, E., Cottrell, J., Csaikl, U.M., van
Dam, B., Deans, J.D., Espinel, S., Fineschi, S., Finkeldey, R., Glaz, I., Goicoechea, P.G.,
8
Jensenn, J.S., König, A.O., Lowe, A.J., Flemming Madsen, S., Mátyás, G., Munro, R.C.,
Popescu, F., Slade, D., Tabbener, H., de Vries, S.G.M., Ziegenhagen, B., de Beaulieu, J.-L.,
Kremer, A. (2002) Identification of refugia and post-glacial colonisation routes of European
white oaks based on chloroplast DNA and fossil pollen evidence. Forest Ecology and
Management, 156, 49-74.
Raven, P.H. (1972) Plant species disjunctions: a summary. Annals of the Missouri Botanical
Garden, 59, 234-246.
Rodríguez-Sánchez, F., Guzmán, B., Valido, A., Vargas, P., Arroyo, J. (2009) Late Neogene history
of the Laurel Tree (Laurus L., Lauraceae) based on phylogeographical analyses of
Mediterranean and Macaronesian populations. Journal of Biogeography, 36, 1270-1281.
Rosenbaum, G., Lister, G.S., Duboz, C. (2002) Reconstruction of the tectonic evolution of the
western Mediterranean since the Oligocene. Journal of the Virtual Explorer, 8, 107-130.
Rosselló J.A., Lázaro, A., Cosín, R., Molins, A. (2007) A phylogeographic split in Buxus balearica
(Buxaceae) as evidenced from nuclear ribosomal markers: when ITS paralogues are welcome.
Journal of Molecular Evolution, 64, 143-157.
Sales, F., Hedge, I.C. (1996) Biogeographical aspects of selected SW Asiatic woody taxa. Annalen
Des Naturhistorischen Museums in Wien, 98, 149-161.
Stace, C.A. (1989) Dispersal versus vicariance - no contest! Journal of Biogeography, 16, 201-202.
Sonboli, A., Azizian, D., Khosravi, F. (2004) Anatomical study of three Buxus L. species
(Buxaceae). Journal of Science (Al-Zahra University), 17, 43-50.
Taberlet, P., Fumagalli, L., Wust-Saucy, A.-G., Cosson, J.-F. (1998) Comparative phylogeography
and postglacial colonization routes in Europe. Molecular Ecology, 7, 453-464.
Thompson, J.D. (2005) Plant evolution in the Mediterranean. Oxford University Press, Oxford.
Thorne, R.F. (1972) Major disjunctions in the geographical ranges of seed plants. Quarterly Review
of Biology, 47, 365-411.
ur-Rahman, A., Ahmed, D., Asif, E., Ahmad, S., Sener, B., Turkoz, S. (1991) Chemical
Constituents of Buxus sempervirens. Journal of Natural Products, 54, 79-82.
von Balthazar, M., Endress, P.K (2002) Reproductive structures and systematics of Buxaceae.
Botanical Journal of the Linnean Society, 140, 193-228.
von Balthazar, M., Endress, P.K., Qiu, Y.-L. (2000) Phylogenetic relationships in Buxaceae based
on nuclear internal transcribed spacers and plastid ndhF sequences. International Journal of
Plant Sciences, 161, 785-792.
Wallet, M. (1983) Auiguines: Un village de tourneurs sur bois. Edition Jeaune, Lafitte, Marseille.
Wegmüller, S. (1984) Zur Ausbreitungsgeschichte von Buxus sempervirens L. im Spät- und
Postglazial in Süd- und Mitteleuropa. Dissertationes Botanicae, 72, 333-344.
9
Willis, K.J., Birks, H.J.B. (2006) What is natural? The need for a long-term perspective in
biodiversity conservation. Science, 314, 1261-1265.
Willis, K.J., Van Andel, T.H. (2004) Trees or no trees? The environments of central and eastern
Europe during the last glaciation. Quaternary Science Reviews, 23, 2369-2387.
Yll, E.-I., Pérez-Obiol, R., Pantaleón-Cano, J., Roure, J.M. (1997) Palynological evidence for
climatic change and human activity during the Holocene on Minorca (Balearic Islands).
Quaternary Research, 48, 339-347.
10
ANNALI DI BOTANICA
Ann. Bot. (Roma), 2011, 1: 45–58
LATE GLACIAL AND HOLOCENE HISTORY
OF BUXUS SEMPERVIRENS L. IN ITALY
DI DOMENICO F.1*, LUCCHESE F.1, MAGRI D.2
1
Dipartimento di Biologia Ambientale, Laboratorio di Botanica Sistematica, Università di Roma Tre, Roma
Dipartimento di Biologia Ambientale, Laboratorio di Paleobotanica, Sapienza Università di Roma, Roma
2
*Corresponding author: Francesco Di Domenico, e-mail: [email protected]
(RECEIVED 07 FEBRUARY 2011; ACCEPTED 22 FEBRUARY 2011)
ABSTRACT - In the course of the Holocene, plant species experienced changes in their area of distribution and population density in response to climate
change, biotic processes and human activities. The combined use of modern and past distribution data provides a powerful tool for assessing the directions and the rates of the changes that took place. Buxus sempervirens L. (common box) is an evergreen angiosperm present in Italy with a scattered
and fragmented distribution resulting from its persistence in the Peninsula through the last glacial maximum and the Holocene. Buxus experienced a
progressive population growth in the course of the Holocene, with different modes and times from region to region, depending on the different
densities of the starting nuclei of Buxus populations. Populations located at latitudes between 41°N and 43°N were already rather dense during the
late glacial. Buxus increased in the course of the Holocene especially in N Italy, while it underwent a severe reduction in S Italy, to the point of disappearing from Sicily and Apulia. Our results demonstrate that the knowledge of Buxus history is especially important in the context of future plant
distribution changes, providing a starting point for conservation action and sustainable management of biodiversity.
KEY WORDS: BUXUS DISTRIBUTION, POLLEN, SICILY, APULIA, POSTGLACIAL
INTRODUCTION
During the Holocene, plant species experienced changes in
their distribution area and population density in response to
climatic trends, habitat changes, biotic processes and human
impact. Combined data of modern and past distribution may
provide a powerful tool to assess the history of plant species,
as well as the directions and rates of changes. The genus
Buxus has been the subject of many investigations due to
its importance, encompassing scientific (conservational:
Zimmermann et al., 2010; palaeoecological: Wegmüller,
1984; biogeographical: Raven & Axelrod, 1974; phytochemical: ur-Rahman et al., 1991; Lorua et al., 2000), cultural
(Leporatti & Ghedira, 2009), as well as economic interest
(Record, 1921; Batdorf, 2006; Köhler, 2007). However, only
a single study deals with the Holocene history of Buxus in
Europe (Wegmüller, 1984), which needs to be updated.
Buxus sempervirens L. was present in Italy in the N
Apennines during the evaporitic Messinian (ca. 5.9-5.6 My
BP; Bertini & Martinetto, 2011). It has persisted in the Italian
Peninsula until the present, while many other floral elements
underwent extinction (Bertini, 2010). Thus, the present
distribution is the result of a very long history, which was
definitively shaped in the course of the Holocene.
The knowledge of Buxus sempervirens distribution in Italy
is fragmented across several local/regional Floras, Herbaria
and field surveys, and no review depicting its complete
distribution in our country is currently available. The species
was considered threatened in the IUCN 1992 Redlists, but it
is not included in more recent Redlists, even though it is
experiencing a severe reduction all over the country, possibly
because of human impact and increasing aridity. Buxus
sempervirens is the characteristic species of the Natural
Habitat types (Natura 2000) of Community interest coded as
5110: “Stable xerothermophilous formations with Buxus
sempervirens on rock slopes (Berberidion p. p.)”. However,
11
46
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
the species is present also in other habitats such as moist
gorges and ravines with mesic conditions.
The aim of the present study is (i) to depict the area of
distribution of Buxus sempervirens in Italy, (ii) to reconstruct
its history through pollen data, (iii) to discuss possible
geographic patterns of distribution in the light of ecological
considerations, (iv) to verify whether Buxus experienced
changes in frequency and distribution over the country in the
course of the Holocene.
In Italy, the current natural distribution of Buxus
sempervirens is scattered throughout the Peninsula north of
40°N and shows a center of distribution around 42°N, in the
middle of the country (Fig. 1, list of sources in Appendix 1).
PRESENT DAY BIOLOGY, ECOLOGY
AND DISTRIBUTION
The phylogenetic position of the genus Buxus L. has been
thouroughly investigated (Köhler & Brückner, 1989, von
Balthazaar et al., 2000). The genus encompasses about 100
species, distributed in two major centers of diversity
(Caribbean-Latin America and E Asia) and a minor
one (Africa). The Asian section (Eubuxus) includes the
“Mediterranean” taxa as western representatives (Buxus
sempervirens L. and Buxus balearica Lam.).
Buxus sempervirens is an evergreen angiosperm, tree or
shrub, growing up to 1-3 (2-8) m height in our country. The
species is slow growing and long-living, reaching maturity
only after several (3-8) years (personal observation). Buxus
sempervirens is monoecious and with proterogynous,
functionally unisexual flowers (male flower may show
pistil rudiments), ambophilous (wind, insects: Diptera,
Hymenoptera) and self pollinating, as well as capable of
vegetative reproduction from broken or buried branches. It
usually flowers from March to April, but can show a great
variation of anthesis in relation to altitudinal and latitudinal
gradients (personal observation). Fruits are dry, loculicidal
and dehiscent capsules with leathery exocarp and persistent
stylodia. Seeds, with a caruncle reduced to two small white
lobes, show a local (1-5 m, balistic) to medium (3-10 m,
myrmecochory) dispersal distance (Fiori & Paoletti, 1908;
Debussche & Lepart, 1992; Köhler, 2007).
Buxus sempervirens occurs in most of Europe (Portugal,
Spain, France, United Kingdom, Ireland, Germany, Belgium,
Italy, Luxembourg, Switzerland, Austria, Italy, Slovenia,
Croatia, Montenegro, F.Y.R.O.M., Albania, Serbia, Kosovo,
Greece, Turkey) and in some parts of northern Africa
(Morocco, Algeria) and W Asia (Georgia, Iran, Azerbaijan,
Russia). In the latter countries the species is usually referred
to as Buxus colchica Pojark. In most cases, Buxus
sempervirens distribution is bound to particular substrates
(chalks, ophiolites and tuffs), in both open (garigues) and
forest areas (thermophilous and mesophilous broad-leaved
deciduous and thermophilous broad-leaved evergreen
forests), frequently on exposed rock slopes along river beds
and moist valleys or basins.
Fig.1. Location of Italian sites where pollen of Buxus was found (red dots),
compared to its modern distribution (green squares, references in Appendix 1).
1. Dura-Moor (Seiwald, 1980; EPD); 2. Castelraimondo di Forgaria (Accorsi
et al., 1992); 3. Bosco del Cansiglio (Kral, 1969); 4. Schwarzsee (Seiwald, 1980);
5. Lago della Costa (Kaltenrieder et al., 2009; Kaltenrieder et al., 2010); 6. Lago
di Ledro (Beug, 1964); 7. Laghetto di Castellaro (Bertoldi, 1968); 8.Lago di
Gaiano (Gehrig, 1997); 9. Lago di Ganna (Schneider & Tobolski, 1985); 10. Selle
di Carnino (de Beaulieu, 1977; EPD); 11. Centa K2 (Arobba et al., 2004);
12. Sestri Levante (Bellini et al., 2009); 13. Lago di Bargone (Cruise et al., 2009);
14. Lago Padule (Watson, 1996; EPD); 15. Lago di Vrazzano (Bertoldi &
Buccella, 1983); 16. Lago dell’Accesa (Drescher-Schneider et al., 2007);
17. Lagaccione (Magri, 1999); 18. Lago di Ripa Sottile (Ricci Lucchi et al., 2000);
19. Valle di Castiglione (Follieri et al., 1988); 20. Roma – Valle del Colosseo
(Celant & Magri, 1999); 21. Maccarese – Lingua d’Oca-Interporto (Di Rita
et al., 2010; Di Rita, pers. comm.); 22. Pesceluna (Di Rita, pers. comm.);
23. Coppa Nevigata – Lago Salso (Di Rita et al., 2011); 24. Lago Grande di
Monticchio (Allen et al., 2000); 25. Lago Alimini Piccolo (Di Rita & Magri,
2009); 26. Lago di Pergusa (Sadori & Narcisi, 2001; Sadori, pers. comm.);
27. Biviere di Gela (Noti et al. 2009); 28. Gorgo Basso (Tinner et al., 2009).
At these localities Buxus sempervirens lives in different
habitats (thermophilous broad-leaved evergreen and
mesophilous broad-leaved deciduous forests; supramediterranean garigues), often on exposed calcareous rock slopes in
river valleys and moist gorges or all around intermountain
basins that hosted large lakes during the Pleistocene. Buxus
is also found in volcanic gorges and ravines of Latium and
Tuscany, where a particular microclimate occurs in relation
12
47
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
to topographic effects (thermal inversion, high moisture) and,
sometimes, to the presence of thermal springs (Giacchi
1974). The co-occurrence of Buxus sempervirens with several
ferns (14 species), subtropical Tertiary relicts (zonal: Laurus
nobilis L., Ruscus aculeatus L.; extrazonal: Ilex aquifolium
L., Daphne laureola L.) and microthermal species (Hepatica
nobilis Miller, Veronica montana L., Euphorbia dulcis L.,
Galanthus nivalis L.) in volcanic gorges at Parco di Veio in
Latium is significant and related to long-term favorable
climatic conditions (Di Domenico & Lucchese, 2007).
Most of the present-day Italian distribution data (list of
sources in Appendix 1) come directly from field surveys, as
part of an ongoing phylogeographic research. Other locations
were derived from bibliographical sources (Natura 2000
database, national and regional floras, publications) and
herbarium specimens. In this respect, it is notable the poor
representation of Buxus sempervirens in many studies, which
consider the species almost or completely absent in our
country.
From a climatic point of view, the present-day Italian
populations are found between a mean annual minimum
temperature of 8.0 ± 5.8 °C, mean annual maximum
temperature of 16.7 ± 7.5 °C and mean annual precipitation
of 860 ± 122 mm. Such a wide variance suggests that
macroclimate parameters are not adequate to represent the
species climatic optimum and that most of the populations
are located in sites where a particular microclimate occurs.
Climatic parameters were extracted through interpolation
of climatic grids at a resoluton of 2.5’ x 2.5’ arc-minutes
(Worldclim database, www.worldclim.org) using Diva-Gis
(Hijmans et al., 2004; 2005).
Fig. 2. Location of Italian sites where pollen of Buxus was found (red dots) in different time windows, compared to its modern distribution (green squares). The late glacial
and Holocene sites correspond to Fig. 1. The sites between 130 and 80 ka BP are: 1. Azzano Decimo (Pini et al., 2009); 2. Lago di Fimon (Pini et al., 2010); 3. Lagaccione
(Magri, 1999); 4. Lago Lungo (Calderoni et al., 1994); 5. Lago di Vico (Magri & Sadori, 1999); 6. Valle di Castiglione (Follieri et al., 1988); 7. Lago Grande di
Monticchio (Allen et al., 2000); 8. Cànolo nuovo (Grüger, 1977).
13
48
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
FOSSIL RECORD OF BUXUS
Characteristics of Buxus pollen
Buxus has a pantoporate pollen grain of 29-38 µm diameter,
exine 2-2.5 µm thick, pori of 1.5-2 µm diamater (Beug,
2004). By comparison, Brückner (1993) reported a diameter
of 18-34 µm, exine thickness of 1.2-2.5 µm and pori diameter
of 0.5-3 µm. These characteristics have a great discriminating
power (Wegmüller, 1984; Beug, 2004), and so we can
exclude misidentification of Buxus pollen in the reviewed
palynological studies.
Holocene pollen records of Buxus from Italy
Buxus pollen was found in 28 (25%) out of the 112 Holocene
continental sites reviewed in the present study (Fig. 1, list of
sites and references in appendix 2), although its frequency in
sediments is generally very low (<1%). It is possible that
Buxus was not reported in some of these sites because only
selected taxa were shown in the published diagrams. The
pollen sites are distributed as follows: 86 sites in N Italy, 17
sites in C Italy, 9 sites in S Italy; Buxus pollen was found in
the 17%, 41% and 67% of the sites, respectively.
The Holocene history of Buxus in Italy is presented by
grouping the pollen sites in five age classes (ages are always
reported in calendar years Before Present) (Fig. 2): 14–10 ka,
10–7 ka, 7–4 ka, 4–1 ka and 1-0 ka. In Fig. 2 the location of
the pollen records with Buxus during the forest phases
preceding the last pleniglacial (130-80 ka) is also shown, for
comparison with the Holocene data.
Late glacial records are located both in C Italy (Lago
dell’Accesa 13.7 ka: Drescher-Schneider et al., 2007; Valle di
Castiglione 13 ka: Alessio et al., 1986, Di Rita, pers. comm.;
Lagaccione 11.5 ka: Magri, 1999), and in S Italy (Lago
Grande di Monticchio ca. 12 ka: Allen et al., 2000). Following
these early occurrences, Buxus appears in C Italy along the
coasts of Latium (Pesceluna 9.7 ka: Di Rita, pers. comm.;
Maccarese – Lingua d’Oca-Interporto 8 ka: Di Rita et al.,
2010), and in N Apennines deposits (Lago di Bargone 9.9 ka:
Cruise et al., 2009; Lago Padule 9.8 ka: Watson, 1996; EPD)
and Liguria (Sestri Levante 8 ka: Bellini et al., 2009).
Afterwards, Buxus becomes visible in Sicily (Lago di
Pergusa 8.3 ka: Sadori & Narcisi, 2001; Sadori, pers. comm.;
Gorgo Basso 8.1 ka: Tinner et al., 2009; Biviere di Gela 7.3
ka: Noti et al., 2009). Buxus then appears in Apulia (Coppa
Nevigata - Lago Salso 6.3 ka: Di Rita et al., 2011; Di Rita,
pers. comm.; Lago Alimini Piccolo 5.5 ka: Di Rita & Magri,
2009). Subsequently it is found at some sites N of the Po
river, in Lombardy (Lago di Ledro 4.8 ka: Beug 1964; Lago
di Gaiano 4.3 ka: Gehrig, 1997), and in Veneto (Lago della
Costa 3 ka: Kaltenrieder et al., 2009; Kaltenrieder et al.,
2010). Between 4 and 1 ka, Buxus is present in almost all
regions, with new records in Latium (Lago di Ripa Sottile
2.5 ka: Ricci Lucchi et al., 2000; Valle del Colosseo in Rome
2 ka: Magri, unpublished data), in Liguria (Centa K2 ca. 1.9
ka: Arobba et al., 2004), in Emilia-Romagna (Lago di Vrazzano
1.1 ka: Bertoldi & Buccella, 1983), and in Veneto (fossil
wood at Castelraimondo di Forgaria 1.9 ka: Accorsi et al.,
1992). In this period a decline of Buxus in S Italy becomes
manifest, as its presence is not recorded anymore in Sicily
(after 2 ka at Biviere di Gela and 1.4 ka at Gorgo Basso), and
in Apulia (after 4.1 ka at Coppa Nevigata - Lago Salso and
1.5 ka at Lago Alimini Piccolo). By contrast, Buxus increases
in the last 1000 years N of the Po river, mostly in the
pre-Alpine belt: in Lombardy (Lago di Ganna 1 ka: Schneider
& Tobolski, 1985; Laghetto di Castellaro approx. 1 ka:
Bertoldi, 1968), in Veneto (Dura-Moor 0.7 ka: Seiwald, 1980;
EPD; Bosco del Cansiglio 0.7 ka: Kral, 1969), in Trentino
Alto-Adige (Schwarzsee ca. 0.6 ka: Seiwald, 1980), and in
Piedmont (Selle di Carnino 0.5 ka: de Beaulieu, 1977; EPD).
Three pollen sites show especially long records of Buxus:
Lago dell’Accesa (13.7 and 2.8 ka), Lago Bargone (9.9-1 ka)
and Gorgo Basso (8.1-1.4 ka).
The plot of the age of Buxus occurrences against the latitude
of pollen sites (Fig. 3) shows that Buxus followed a latitudinal
gradient in its Holocene appearance, being present since the
late glacial at latitudes between 43°N and 41°N (14-10 ka, 4
sites) and then showing up in pollen sites at higher and lower
latitudes. The second group of appearances (10-5 ka) took
place in the N Apennines (Liguria and Tuscany) and Sicily.
These were followed (5-1 ka), by several sites N of the Po
river (Lombardy, Trentino Alto-Adige). Interestingly, Buxus
appears rather early (ca. 8.5 ka) in the sites south of 38°N,
which are located in areas where it does not presently occur.
Considering the whole Holocene record, the number of sites
where Buxus is found increases through time until about
2 ka. Buxus then disappeared from a number of sites where
actually it is not present anymore. In particular, it disappeared
from Sicily at 2-1.5 ka, from Apulia at 1.5 ka, and from the
Veneto plain at 2 ka.
Fig.3. Age-latitude plot of Buxus occurrences in Italian pollen records, grouped
by regional/geographical location.
14
49
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
DISCUSSION
The latitudinal pattern found in the time series (Fig. 3) cannot
be interpreted as a spread from C Italy towards higher and
lower latitudes, which is an unrealistic scenario considering
the dispersal abilities of Buxus (Fiori & Paoletti, 1908;
Debussche & Lepart, 1992; Monestiez & Chadoeuf, 2002;
Köhler, 2007). Plausibly, at the onset of the Holocene the
populations in Tuscany, Latium and Basilicata were already
rather dense. For this reason, the postglacial expansion of
Buxus was detectable in these regions earlier than elsewhere.
The positive and significant relationship between the
abundance of Buxus pollen and the population density in
present-day vegetation provides some support to this
hypothesis. (Cañellas-Boltà et al., 2009).
Despite the delayed postglacial increase of Buxus populations
in N and S Italy, it is most likely that Buxus persisted in large
areas of the Italian Peninsula during the last glacial period
(Fig. 2). In fact, during the forest phases preceding the last
pleniglacial (130-80 ka), Buxus was present in all the long
pollen records studied in Italy, including Cànolo Nuovo
(Grüger, 1977), Lago Grande di Monticchio (Allen &
Huntley, 2009), Valle di Castiglione (Follieri et al., 1988),
Lago di Vico (Magri & Sadori, 1999), Lago Lungo
(Calderoni et al., 1994), Lagaccione (Magri, 1999), Azzano
Decimo (Pini et al., 2009), and Lago di Fimon (Pini et al.,
2010). Besides, between 30 and 18 ka Buxus is recorded at
Lago della Costa (Kaltenrieder et al., 2009) and Lago Grande
di Monticchio (Allen & Huntley, 2000). In most of these sites
Buxus is present also during the Holocene, indicating that it
persisted locally through the last glacial period. However, in
a number of sites Buxus underwent a severe reduction, so that
it is not found during the Holocene at Cànolo Nuovo (Grüger,
1977), Azzano Decimo (Pini et al., 2009), and Lago di Fimon
(Pini et al., 2010).
No long pollen sequences extending beyond the last glacial
maximum are available from Sicily and Apulia. However, the
presence of Buxus in these southern regions before the
last glacial is most likely, considering its early Holocene
appearances. The marked regional reduction of Buxus to
the point of its disappearance was never realized by
palaeobotanists, who did not interpret the fossil record in the
context of its modern absence. Conversely, the absence of
Buxus in Apulia and Sicily was never questioned by
botanists, who did not consider the possibility that Buxus
could extend so much to the south just a few millennia ago.
A point worthy of discussion is the dynamics of Buxus in the
last 2 ka. In this period its representation increases in N Italy
and vanishes in S Italy. It is very difficult to ascribe this trend
to either climate change or to the possible impact of human
activities. Concerning anthropogenic causes, a reduction of
Buxus may have been indirectly caused by grazing pressure,
while an increase may be linked to the ornamental interest in
Buxus sempervirens. Regarding the grazing pressure, it is
unlikely that Buxus was directly affected by herbivores, as
the leaves are at best unpalatable (if not toxic) thanks to the
presence of several alkaloids (Russel et al., 1997). For the
same reason, it is possible that Buxus was cut down to favor
livestock grazing, as evidenced for the pasturing regimes in
present-day calcareous grasslands (Barbaro et al., 2004).
In the respect of climate, the regions of S Italy, where Buxus
is presently missing, were subject to a progressive
desertification in the last few millennia (Di Rita & Magri,
2009), while the coeval records of N Italy, where Buxus
increases, do not show any trend towards arid conditions
(Finsinger & Tinner, 2006). Other determinant causes for the
decline of Buxus, such as plant pathogens and intraspecific
competition, cannot be ruled out. Interestingly, also Buxus
balearica Lam. experienced a severe reduction in the course
of the last few thousands of years, whose causes are still
subject of debate (Yll et al., 1997).
The present study offers new hints for the sustainable
management of biodiversity, showing that Buxus
sempervirens is undergoing a severe reduction in the southern
part of its range. In this respect, particular attention should be
paid to the conservation of the taxon in such areas, which do
not appear prone to favor a long-term persistence of Buxus
sempervirens.
CONCLUSIONS
The use of modern distribution data in conjunction with past
distributions provided new insights about the Holocene
history of Buxus sempervirens in Italy. In particular, our
approach led to the following conclusions:
– The present distribution of Buxus sempervirens in
Italy, reconstructed using field survey, Floras and
Herbarium accessions, is rather fragmented and
scattered, with a center of frequency and abundance
around 42°N, in the middle of the Peninsula.
– Pollen data shows that during the Holocene Buxus
experienced a progressive population growth with
different modes and times from region to region,
depending on the initial densities of the nuclei of
Buxus populations at the end of the last glacial period.
– The populations of Buxus located at latitudes between
41°N and 43°N are detected by pollen analyses
already during the late glacial, when they were probbly
already rather dense.
– In the course of the Holocene, Buxus populations
increased especially in N Italy, where the species
15
50
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
could take advantage of a favourable climate and
possibly also of human activities because of its
ornamental value.
– In S Italy Buxus underwent a severe reduction after
2 ka, to the point of being currently absent in Sicily,
Calabria, and Apulia.
– Disentangling the causes for the reduction of Buxus
in S Italy may provide new hints for a sustainable
management of the taxon, which requires adequate
protection measures in the light of its fragmented
distribution and overall reduced population size.
Arobba D., Caramiello R., Firpo M., 2004. Contributi
paleobotanici alla storia dell’evoluzione di una pianura
costiera: il caso di Albenga. In R.C. de Marinis & G. Spadea
(eds), I Liguri, 76–78, Skira, Ginevra-Milano.
Barbaro L., Dutoit T., Anthelme F., Corcket E., 2004.
Respective influence of habitat conditions and management
regimes on prealpine calcareous grasslands. Journal of
Environmental Management 72, 261–275.
Batdorf L.R., 2006. International Registration List of
Cultivated Buxus L. The Boxwood Bullettin, American
Boxwood Society, 19 pp.
Bellini C., Mariotti Lippi M., Montanari C., 2009. The
Holocene landscape history of the NW Italian coasts. The
Holocene 19, 1161-1172.
ACKNOWLEDGEMENTS
The European Pollen Database (EPD) and Natura 2000
contributors for the Italian sites are acknowledged. We wish
to thank Federico Di Rita, Laura Sadori and Alessandra
Celant for additional information on fossil data. FDD is
grateful to Prof. Josep Rosselló for his encouragement and
to Eleni Dimler-Bruni for her helpful suggestions. This work
was supported by grants attributed to the Authors by the
University Roma Tre and Sapienza University of Rome, and
the Italian Ministry of University and Research (MIUR),
ncluding PRIN funds.
REFERENCES
Accorsi C.A., Bandini Mazzanti M., Forlani L.,
Marchesini M., 1992. Castelraimondo scavi 1988-1990.
Cataloghi e Monografie Archeologiche Dei Civici Musei di
Udine 2, 255-264.
Alessio M., Allegri L., Bella F., Calderoni G., Cortesi C., Dai
Pra G., De Rita D., Esu D., Follieri M., Importa S., Magri D.,
Narcisi B., Petrone V., Sadori L., 1986. 14C dating,
geochemical features, faunistic and pollen analyses of the
uppermost 10 m core from Valle di Castiglione (Rome,
Italy). Geologica Romana 25, 287-308.
A l l e n J . R . M . , Wa t t s W. A . , H u n t l e y B . , 2 0 0 0 .
Weichselian palynostratigraphy, palaeovegetation and
palaeoenvironment: the record from Lago Grande di
Monticchio, Southern Italy. Quaternary International 73/74,
91-110.
Allen J.R.M. & Huntley B., 2009. Last interglacial
palaeovegetation, palaeoenvironments and chronology:
A new record from Lago Grande di Monticchio, southern
Italy. Quaternary Science Reviews 28(15-16), 1521-1538.
Bertini A., 2010. Pliocene to Pleistocene palynoflora and
vegetation in Italy: State of the art. Quaternary International
225, 5-24.
Bertini A. & Martinetto E., 2011. Reconstruction of
vegetation transects for the Messinian–Piacenzian of Italy
by means of comparative analysis of pollen, leaf and
carpological records. Palaeogeography Palaeoclimatology
Palaeoecology, doi:10.1016/j.palaeo.2010.09.005.
Bertoldi R. & Buccella L., 1983. Indicatori colturali in una
sequenza palinologica tardo-postglaciale del preappennino
emiliano. Giornale Botanico Italiano 117(1), 52.
Bertoldi R., 1968. Ricerche pollinologiche sullo sviluppo
della vegetazione tardiglaciale e postglaciale nella regione
del lago di Garda. Studi Trentini di Scienze Naturali, Sez B,
45, 87-162.
Beug H.J., 2004. Leitfaden der Pollenbestimmung. Ed.
Verlag Dr. Friedrich Pfeil, München.
Beug H.J., 1964. Untersuchungen zur spät- und
postglazialen Vegetationgeschichte im Gardaseegebiet unter
besonderer Berücksichtigung der mediterranen Arten. Flora
154, 401-444.
Brückner P., 1993. Pollen morphology and taxonomy of
Eurasiatic species of the genus Buxus (Buxaceae). Grana 32,
65-78.
Calderoni G., Carrara C., Ferreli L., Follieri M., Gliozzi E.,
Magri D., Narcisi B., Parlotto M., Sadori L., Serva L., 1994.
Palaeoenvironmental, palaeoclimatic and chronological
interpretations of a late-Quaternary sediment core from
Piana di Rieti (central Apennines, Italy). Giornale di
Geologia 56, 43-72.
Cañellas-Boltà N., Rull V., Vigo J., Mercadé A., 2009.
Modern pollen–vegetation relationships along an
altitudinal transect in the central Pyrenees (southwestern
16
51
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
Europe). The Holocene 19(8), 1185-1200.
30, 329-356.
Celant A. & Magri D., 1999. Archaeobotanical study of a
sediment core from the Colosseum valley. Congresso della
Società Botanica Italiana. (p. 40). Ferrara.
Gehrig R., 1997. Pollenanalytische Untersuchungen zur
Vegetations- und Klimageschichte des Val Camonica
(Norditalien). Dissertationes Botanicae 276, 1-152.
Cruise G.M., Macphail R.I., Linderholm J., Maggi R.,
Marshall P.D., 2009. Lago di Bargone, Liguria, N Italy:
a reconstruction of Holocene environmental and land-use
history. The Holocene 19, 987-1003.
Giacchi M., 1974. Una stazione di Buxus sempervirens L. in
Val di Merse (Siena). Informatore Botanico Italiano 6(2),
153-157.
de Beaulieu J.-L., 1977. Contribution pollenanalytique à
l’histoire tardiglaciaire et Holocène de la végétation des
Alpes méridionales françaices. Ph.D. Dissertation. Université
d’Aix-Marseille, Marseille, France.
Grüger E., 1977. Pollenanalytische Untersuchung zur
würmzeitlichen Vegetationsgeschichte von Kalabrien
(Süditalien). Flora 166, 475-489.
Debussche M. & Lepart J., 1992. Establishment of woody
plants in Mediterranean old fields: opportunity in space and
time. Landscape Ecology 6, 133-145.
Hijmans R.J., Guarino L., Bussink C., Mathur P., Cruz M.,
Barrentes I., Rojas E., 2004. DIVA-GIS. Vsn. 5.0.
A geographic information system for the analysis
of species distribution data. Manual available at
http://www.diva-gis.org.
Di Domenico F. & Lucchese F., 2007. Le emergenze
floristiche nel Parco di Veio. Flora rara e conservazione degli
habitat pregevoli. Atti del Convegno, La Biodiversità del
Parco di Veio a dieci anni dalla sua istituzione, pp. 58-61.
Hijmans R. J., Cameron S. E., Parra J.L., Jones P.G.,
Jarvis A., 2005. Very high resolution interpolated climate surfaces for global land areas. International Journal of
Climatology. 25, 1965-1978.
Di Rita F. & Magri D., 2009. Holocene drought,
deforestation and evergreen vegetation development in the
central Mediterranean: a 5500 year record from Lago
Alimini Piccolo, Apulia, southeast Italy. The Holocene 19(2),
295-306.
Kaltenrieder P., Belis C.A., Hofstetter S., Ammann B.,
Ravazzi C., Tinner W., 2009. Environmental and climatic
conditions at a potential Glacial refugial site of tree species
near the Southern Alpine glaciers. New insights from
multiproxy sedimentary studies at Lago della Costa
(Euganean Hills, Northeastern Italy). Quaternary Science
Reviews 28, 2647-2662.
Di Rita F., Celant A., Magri D., 2010. Holocene
environmental instability in the wetland north of the Tiber
delta (Rome, Italy): sea-lake-man interactions. Journal of
Paleolimnology 44, 51-67.
Di Rita F., Simone O., Caldara M., Gehrels W.R., Magri D.,
2011. Holocene environmental changes in the coastal
Tavoliere Plain (Apulia, southern Italy): a multiproxy
a p p r o a c h . P a l a e o g e o g r a p h y, P a l a e o c l i m a t o l o g y,
Palaeoecology, submitted.
Drescher-Schneider R., de Beaulieu J.-L., Magny M.,
Walter-Simonnet A.-V., Bossuet G., Millet L., Brugiapaglia
E., Drescher A., 2007. Vegetation history, climate and
human impact over the last 15,000 years at Lago
dell’Accesa (Tuscany, Central Italy). Vegetation History and
Archaeobotany 16, 279-299.
Finsinger W. & Tinner W., 2006. Holocene vegetation and
land-use changes in response to climatic changes in the
forelands of the southwestern Alps, Italy. Journal of
Quaternary Science 21, 243-258.
Kaltenrieder P., Procacci G., Vannière B., Tinner W., 2010.
Vegetation and fire history of the Euganean Hills (Colli
Euganei) as recorded by Lateglacial and Holocene
sedimentary series from Lago della Costa (northeastern
Italy). The Holocene 20(5), 679-695.
Köhler E., 2007. Buxaceae. In: The families and genera of
flowering plants. Ed. Klaus Kobinski. Springer Berlin
Heidelberg New York.
Köhler E. & Brückner P., 1989. The genus Buxus
(Buxaceae): aspects of its differentiation in space and time.
Plants Systematics and Evolution 162, 267-383.
Kral F., 1969. Zur Frage der natürlichen Fichtenverbreitung
im Fagetum des Bosco del Cansiglio. Pollenanalitische
Untersuchungen in den Venetianer Alpen. Mitt. OstalpinDinarischen Pflanzensoziol. Arbeitsgem. 9, 261-273.
Fiori A. & Paoletti G., 1908. Flora analitica d’Italia. Padova,
Tipografia del Seminario.
Leporatti M.L. & Ghedira K., 2009. Comparative analysis of
medicinal plants used in traditional medicine in Italy and
Tunisia. Journal of Ethnobiology and Ethnomedicine
5, 31-38.
Follieri M., Magri D., Sadori L., 1988. 250,000-year pollen
record from Valle di Castiglione (Rome). Pollen et Spores
Lorua F., Duvala D., Aumelasb A., Akeba F., Guédonc D.,
Guedj R., 2000. Four steroidal alkaloids from the leaves of
17
52
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
Buxus sempervirens. Phytochemistry 54(8), 951-957.
Magri D., 1999. Late Quaternary vegetation history at
Lagaccione near Lago di Bolsena (central Italy). Review of
Palaeobotany and Palynology 106, 171-208.
Magri D. & Sadori L., 1999. Late Pleistocene and Holocene
pollen stratigraphy at Lago di Vico (central Italy). Vegetation
History and Archaeobotany 8, 247-260.
Monestiez P. & Chadoeuf J., 2002. Spatial statistics for
dynamics characterization: Buxus sempervirens invasion in
extensive farming systems. Pp. 403- 406 in 53ème Session de
l’Institut International de Statistique , edited by ISI Institut
International de Statistique, Voorburg.
Noti R., van Leeuwen J.F.N., Colombaroli D., Vescovi E.,
Pasta S., La Mantia T., Tinner W., 2009. Mid- and
ate-Holocene vegetation and fire history at Biviere di Gela,
a coastal lake in southern Sicily, Italy. Vegetation History and
Archaeobotany 18(5), 371-387.
Pini R., Ravazzi C., Donegana M., 2009. Pollen stratigraphy,
vegetation and climate history of the last 215 ka in the
Azzano Decimo core (plain of Friuli, north-eastern Italy).
Quaternary Science Reviews 28(13-14), 1268-1290.
Pini R., Ravazzi C., Reimer P.J., 2010. The vegetation and
climate history of the last glacial cycle in a new pollen record
from Lake Fimon (southern Alpine foreland, N-Italy).
Quaternary Science Reviews 29, 3115-3137.
Raven P.H. & Axelrod D.I., 1974. Angiosperm Biogeography
and Past Continental Movements. Annals of the Missouri
Botanical Garden 61( 3), 539-673.
Record S.J., 1921. Boxwoods of commerce. Bullettin of the
Torrey Botanical Club 48(11), 297-306.
Ricci Lucchi M., Calderoni G., Carrara C., Cipriani N.,
Esu D., Ferreli L., Girotti O., Gliozzi E., Lombardo M.,
Longinelli A., Magri D., Nebbiai M., Ricci Lucchi F.,
Vigliotti L., 2000. Late Quaternary record of the Rieti
basin, central Italy: paleoenvironmental and paleoclimatic
evolution. Giornale di Geologia 62, 105-136.
Russel A.B., Hardin J.W., Grand L., Fraser A., 1997.
Poisonous Plants of North Carolina. North Carolina State
University, Poison Control Center.
Sadori L. & Narcisi B., 2001. The postglacial record of
environmental history from Lago di Pergusa (Sicily). The
Holocene 11, 655-671.
Schneider R. & Tobolski K., 1985. Lago di Ganna Late-Glacial and Holocene environments of a lake in the
Southern Alps. Dissertationes Botanicae 8, 229-271.
Seiwald A., 1980. Beiträge zur Vegetationsgeschichte Tirols
IV: Natzer Plateau - Villanderer Alm. Ber. nat.-med. Verein
Innsbruck 67, 31-72.
Tinner W., van Leeuwen J.F.N., Colombaroli D., Vescovi E.,
van der Knaap W.O., Henne P.D., Pasta S., D’angelo S.,
La Mantia T., 2009. Holocene environmental and climatic
changes at Gorgo Basso, a coastal lake in southern Sicily,
Italy. Quaternary Science Reviews 28, 1498-1510.
ur-Rahman A., Ahmed D., Asif E., Ahmad S., Sener B.,
Turkoz S., 1991. Chemical Constituents of Buxus
sempervirens. Journal of Natural Products 54 (1), 79-82.
von Balthazar M., Endress P.K., Qiu Y.-L., 2000.
Phylogenetic relationships in Buxaceae based on nuclear
internal transcribed spacers and plastid ndhF sequences.
International Journal of Plant Sciences 161, 785-792.
Watson C., 1996. The vegetational history of the northern
Apennines, Italy: information from three new sequences and
a review of Regional vegetational change. Journal of
Biogeography 23, 805-841.
Wegmüller S., 1984. Zur Ausbreitungsgeschichte von Buxus
sempervirens L. im Spät- und Postglazial in Süd- und
Mitteleuropa. Dissertationes Botanicae (Festchrift Welten)
72, 333-344.
Yll E.-I., Pérez-Obiol R., Pantaleón-Cano J., Roure J.M.,
1997. Palynological Evidence for Climatic Change and
Human Activity during the Holocene on Minorca (Balearic
Islands). Quaternary Research 48, 339-347.
Zimmermann M., Vischer-Leopold M., Ellwanger G.,
Ssymank A., Schröder E., 2010. The EU Habitats
Directive and the German Natura 2000 Network of Protected
Areas as Tool for Implementing the Conservation of Relict
Species. In: Habel J.C. & Assmann T. (eds), Relict Species:
Phylogeography and Conservation biology. Springer-Verlag
Berlin Heidelberg, Part V pp. 323-340.
APPENDIX 1
The vast majority of modern data for Buxus sempervirens L.
(110 occurrences) comes from field surveys by FDD and FL.
Other occurrences were retrieved from Herbarium Specimens
(12 occurrences, Herbaria of Sapienza and Roma Tre
Universities). These data were integrated by the following
Floras (18 occurrences):
Arcangeli G., 1882. Compendio della Flora Italiana. Ed.
Ermanno Loescher, Torino.
Balbis J.H., 1806. Flora Taurinensis. Ex Typographia
Johannis Grossi, Torino.
18
53
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
Beguinot A., 1909. Flora Padovana. Premiata Società
Cooperativa Tipografica, Padova.
APPENDIX 2
Bertoloni A., 1832. Mantissa Plantarum Florae Alpium
Apuanarum. Ex Typographaeo Emygdii ab Ulmo et Iosephi
Tiocchi, Bononiae.
Bertoloni A., 1833. Flora Italica. Ex Typograpileo Richardi
Masii.
Bracciforti A., 1877. Flora Piacentina. Tipografia F. Salari,
Piacenza.
Caruel T., 1860. Prodromo della Flora Toscana. Ed. Forni,
Firenze.
Comolli G., 1857. Flora Comense. Tipografia degli Eredi
Bizzoni, Pavia.
Fiori A. & Paoletti G., 1908. Flora Analitica d’Italia.
Tipografia del Seminario, Padova.
Fiori A., 1923. Nuova Flora Analitica d’Italia. Tipografia di
M. Ricci, Firenze.
Gavioli O., 1974. Synopsis Florae lucanae. Nuovo
Giornale Botanico Italiano 54, 10-278.
Gelmi E., 1893. Prospetto della Flora Trentina. Ed. Theodor
Oswald Weigel, Leipzig.
Gibelli G. & Pirotta R., 1882. Flora del Modenese e del
Reggiano. Tipografia di G.T. Vincenzi e Nipoti, Modena.
Gortani L. & Gortani M., 1969. Flora Friulana. Ed. Forni,
Bologna.
Paolucci L., 1890. Flora Marchigiana. Premiato
Stabilimento Tipo-Litografico Federici, Pesaro.
Parlatore F., 1848. Flora Italiana. Tipografia Le Monnier,
Firenze.
Passerini G., 1852. Flora dei Contorni di Parma. Tipografia
Carmignani, Parma.
Pignatti S., 1982. Flora d’Italia. Edagricole, Bologna.
Senni L., 1943. La vegetazione dei monti Albani. La Rivista
Forestale Italiana 5, 141-151.
Tenore M., 1838. Ad florae neapolitanae syllogem.
Tipografia del Fibreno, Napoli.
Terracciano N., 1873. Enumeratio plantarum vascularium
sponte nascentium. Nuovo Giornale Botanico Italiano
5(4), 225-260.
List of Italian sites reviewed in the present study: 1. Schwarzsee (Seiwald, 1980);
2. Dura-Moor (Seiwald, 1980;) 3. Rinderplatz (Seiwald, 1980); 4. Sommersuss
(Seiwald, 1980); 5. Comelico (Kral, 1986); 6. Pescosta (Borgatti et al., 2007);
7. Borghetto alto (Moe et al., 2007); 8. Pian Venezia (Speranza et al., 2007);
9. Val Vidrola sotto (Moe et al., 2007); 10. Lago Basso (Fedele & Wick, 1996);
11. Torbiera Ghighel (Braggio Morucchio et al., 1993); 12. Borghetto sotto (Moe
et al., 2007); 13. Agordo (Dai Pra & Giardini, 2001); 14. Passo del Tonale
(Gehrig, 1997); 15. Castelraimondo di Forgaria (Accorsi et al., 1992); 16. Lago
Nero di Cornisello (Filippi et al., 2005a); 17. Col di Val Bighera (Gehrig, 1997);
18. Lago Ragogna (Monegato et al., 2007); 19. Pian di Gembro (Pini, 2002);
20. Palù bei Edolo (Gehrig, 1997); 21. Palughetto di Cansiglio (Ravazzi, 2002);
22. Bosco del Cansiglio (Kral, 1969) 23. Palù di Livenza (Pini, 2004); 24. Lago
di Lavarone (Filippi et al., 2005b); 25. Lago di Ganna (Schneider & Tobolski,
1985); 26. Lago di Ledro (Beug, 1964); 27. Lago del Segrino (Wick, 1996);
28. Torbiera di Santa Anna (Brugiapaglia, 2007); 29. Azzano Decimo (Pini et al.,
2009); 30. Forcellona (Kral, 1980); 31. Laghetti del Crestoso (Scaife, 1997);
32. Laghetto di Biandronno (Schneider, 1978); 33. Lago di Champlong
(Brugiapaglia, 2007); 34. Torbiera di Champlong (Brugiapaglia, 2007);
35. Torbiera di Pilaz (Brugiapaglia, 2007); 36.Lago di Annone (Wick & Mohl,
2006); 37. Lago di Loditor (Brugiapaglia, 1996); 38. Lago di Gaiano (Gehrig,
1997); 39. Lago di Villa (Brugiapaglia, 1996); 40. Torbiera del Lago d’Iseo
(Bertoldi & Consolini, 1989); 41. Fiorentina (Miola et al., 2006); 42. Ca’ Tron
(Miola et al., 2006); 43. Lago della Costa (Kaltenrieder et al., 2009; Kaltenrieder
et al. 2010); 44. Palazzetto (Miola et al. 2006); 45. Lago Lucone (Valsecchi et al.
2006); 46. Lago di Fimon (Valsecchi et al., 2008); 47. Venezia ARS1 (Serandei
et al., 2005); 48. Torbiera di Alice (Schneider, 1978); 49. Lago di Viverone
(Schnerider, 1978); 50. Laghetto di Castellaro (Bertoldi, 1968); 51. Lago Falin
(Caramiello et al., 1995); 52. Lago Piccolo di Avigliana (Finsinger & Tinner,
2006); 53. Parma terramara (Cremaschi et al., 2006); 54. Casanova (Cruise,
1990); 55. Berceto (Bertoldi et al., 2007); 56. Terramara di Montale (Mercuri
et al., 2006); 57. Agoraie (Cruise, 1990); 58. Torbiera del Lajone (Braggio
Morucchio et al., 1978; Guido et al., 2004a); 59. Bubano Quarry Est (Ravazzi
et al., 2006); 60. Lagdei (Bertoldi, 1980; Bertoldi et al., 2007); 61. Lago Baccio
(Mori Secci, 1996); 62. Val Bisagno (Montanari et al., 1997); 63. Prato Spilla C
(Lowe, 1992); 64. Prato Spilla A (Lowe, 1992); 65. Rapallo (Bellini et al., 2009);
66. Pavullo nel Frignano (Vescovi et al., 2007); 67. Chiavari (Guido et al., 2004b);
68. Lago di Bargone (Cruise et al., 2009); 69. Lago Padule (Watson, 1996);
19
54
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
70. Vrazzano (Bertoldi & Buccella, 1983); 71. Sestri Levante (Bellini et al.,
2009); 72. Lago Capello (Bertoldi et al., 1986); 73. Laghi dell’Orgials (Ortu et al.,
2006); 74. Pian Marchisio (Ortu et al., 2008); 75. Rifugio Mondovì (Ortu et al.,
2008); 76. Torbiera del Biecai (Ortu et al., 2008); 77. Lago del Vei del Bouc
(Finsinger, 2001); 78. Lago Pratignano (Watson, 1996); 79. Ospitale (Watson,
1996); 80. Selle di Carnino (de Beaulieu, 1977); 81. Lago del Greppo (Ravazzi
et al., 2006); 82. Lago Nero (Mori Secci, 1996); 83. Centa K2 (Arobba et al.,
2004); 84. Lago di Massaciuccoli (Colombaroli et al., 2007; Mariotti Lippi et al.,
2007); 85. Pisa (Bellini et al., 2009); 86. Arno M1 (Ricci Lucchi et al., 2006);
87. Colfiorito (Brugiapaglia & de Beaulieu, 1995); 88. Lago dell’Accesa
(Drescher-Schneider et al., 2007); 89. Ombrone (Biserni & van Geel, 2005);
90. Lago di Mezzano (Sadori et al., 2004); 91. Lagaccione (Magri, 1999);
92. Lago Lungo (Calderoni et al., 1994); 93. Lago di Ripa Sottile (Ricci Lucchi
et al., 2000); 94. Lago di Vico (Magri & Sadori, 1999); 95. Stracciacappa
(Giardini, 2006); 96. Lago di Martignano (Kelly & Huntley, 1996); 97. Caldara
di Manziana (Biondi et al., 1998); 98. Lago Battaglia (Caroli & Caldara, 2006);
99. Roma – Valle del Colosseo (Celant & Magri, 1999); 100. Valle di Castiglione
(Alessio et al., 1986; Follieri et al., 1988); 101. Maccarese – Lingua
d’Oca-Interporto (Di Rita et al., 2010); 102. Lago Albano (Lowe et al., 1996;
Mercuri et al., 2002); 103. Pesceluna (Di Rita, pers. comm.); 104. Portus (Sadori
et al., 2010); 105. Coppa Nevigata – Lago Salso (Caldara et al., 1999, Di Rita
et al., 2011); 106. Lago Grande di Monticchio (Allen et al. 2000); 107. Lago
d’Averno (Grüger & Thulin 1998); 108. Lago Alimini Piccolo (Di Rita & Magri,
2009); 109. Cànolo Nuovo (Schneider, 1985); 110. Gorgo Basso (Tinner et al.,
2009); 111. Lago di Pergusa (Sadori & Narcisi, 2001); 112. Biviere di Gela
(Noti et al., 2009).
Bertoldi R., 1968 . Ricerche pollinologiche sullo sviluppo
della vegetazione tardiglaciale e postglaciale nella regione
del lago di Garda. Studi Trentini di Scienze Naturali, Sez B,
45, 87-162.
Bertoldi R., 1980. Le vicende vegatazionali e climatiche nella
sequenza paleobotanica wurmiana e post-wurmiana di
Lagdei (Appennino settentrionale). Ateneo Parmense Acta
Naturalia 16, 147-175.
Bertoldi R. & Buccella L., 1983. Indicatori colturali in una
sequenza palinologica tardo-postglaciale del preappennino
emiliano. Giornale Botanico Italiano 117(1), 52.
Bertoldi R. & Consolini M., 1989. Deglaciazione e sviluppo
della vegetazione tardiglaciale e postglaciale nella regione
del lago di Iseo attraverso analisi polliniche e datazione 14C.
Memorie della Società Geologica Italiana 42, 139-145.
Bertoldi R., Chelli A., Roma R., Tellini C. 2007. New data
from northern Apennines (Italy) pollen sequences spanning
the last 30,000 yrs. Il Quaternario 20(1), 3-20.
REFERENCES
Bertoldi R., Timossi A. M., Borghi M. L., Magri D., 1986.
Studio di una piccola torbiera montana dell’Appennino
Emiliano attraverso i microfossili vegetali. L’Italia Forestale
e Montana 41(4), 216-230.
Accorsi C.A., Bandini Mazzanti M., Forlani L., Marchesini
M., 1992. Castelraimondo scavi 1988-1990. Cataloghi e
Monografie Archeologiche Dei Civici Musei di Udine 2,
255-264.
Beug H.J., 1964. Untersuchungen zur spät- und postglazialen
Vegetationgeschichte im Gardaseegebiet unter besonderer
Berücksichtigung der mediterranen Arten. Flora 154,
401-444.
Alessio M., Allegri L., Bella F., Calderoni G., Cortesi C., Dai
Pra G., De Rita D., Esu D., Follieri M., Importa S., Magri D.,
Narcisi B., Petrone V., Sadori L., 1986. 14C dating,
geochemical features, faunistic and pollen analyses of the
uppermost 10 m core from Valle di Castiglione (Rome, Italy).
Geologica Romana 25, 287-308.
Biondi E., Brugiapaglia E., Tedeschini Lalli L., 1998.
Indagine geobotanica sulla “Caldara di Manziana” (Italia
centrale). Fitosociologia 35, 65-76.
Allen J.R.M., Watts W.A., Huntley B., 2000. Weichselian
palynostratigraphy, palaeovegetation and palaeoenvironment:
the record from Lago Grande di Monticchio, Southern Italy.
Quaternary International 73/74, 91-110.
Arobba D., Caramiello R., Firpo M., 2004. Contributi
paleobotanici alla storia dell’evoluzione di una pianura
costiera: il caso di Albenga. In R.C. de Marinis & G. Spadea
(eds), I Liguri, 76–78, Skira, Ginevra-Milano.
Biserni G. & van Geel B., 2005. Reconstruction of Holocene
palaeoenvironment and sedimentation history of the
Ombrone alluvial plain (South Tuscany, Italy). Review of
Palaeobotany and Palynology 136, 16-28.
Borgatti L., Ravazzi C., Donegana M., Corsini A., Marchetti
M., Soldati M., 2007. A lacustrine record of early Holocene
watershed events and vegetation history, Corvara in Badia,
Dolomites (Italy). Journal Quaternary Science 22(2),
173-189.
Bellini C., Mariotti Lippi M., Montanari C., 2009. The
Holocene landscape history of the NW Italian coasts. The
Holocene 19, 1161–1172.
Braggio Morucchio G., Guido M. A., Montanari C., 1978.
Studio palinologico e vegetazione della torbiera del Lajone
presso Piampaludo (Gruppo M. Beigua, Appennino ligure
occidentale). Archivio Botanico e Biogeografico Italiano 54,
115-136.
Benvenuti M., Mariotti-Lippi M., Pallecchi P., Sagri M.,
2006. Late-Holocene catastrophic floods in the terminal Arno
River (Pisa, Central Italy) from the story of a Roman riverine
harbour. The Holocene 16, 863-876.
Braggio Morucchio G., Cavaliere S., Cornara L., 1993.
Analisi polliniche nella torbiera del Ghighel (Val Formazza
- Alpi Occidentali) e ricostruzione della paleovegetazione.
Candollea 48(2), 475-484.
20
55
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
Brugiapaglia E., 1996. Dynamique de la vegetation
Tardiglaciaire et Holocene dans les Alpes Italiennes
Nord-Occidentales. PhD Thesis, Université de Saint-Jerome,
Marseille, France.
Brugiapaglia E., 2007. Evoluzione altitudinale spaziotemporale degli alberi durante gli ultimi 12000 anni in Valle
d’Aosta. Il Quaternario 20(2), 195-212.
Brugiapaglia E. & de Beaulieu J.-L., 1995. Etude de la
dynamique végétale tardiglaciaire et Holocène en Italie
centrale: le marais de Colfiorito (Ombrie). Les Comptes
Rendus de l’Académie des Sciences 321, série IIa, 617-622.
Caldara M., Cazzella A., Fiorentino G., Lopez R.,
Magri D., Simone O., 1999. Primi risultati di una ricerca
paleoambientale nell’area di Coppa Nevigata (Foggia).
In: A. Gravina (ed.) Atti 19° Convegno sulla PreistoriaProtostoria e Storia della Daunia. San Severo 1998, 199-236.
San Severo, Foggia.
Calderoni G., Carrara C., Ferreli L., Follieri M., Gliozzi E.,
Magri D., Narcisi B., Parlotto M., Sadori L., Serva L., 1994.
Palaeoenvironmental, palaeoclimatic and chronological
interpretations of a late-Quaternary sediment core from Piana
di Rieti (central Apennines, Italy). Giornale di Geologia 56,
43-72.
history. The Holocene 19, 987-1003.
Dai Pra G. & Giardini M., 2001. Cenni geomorfologici,
litologia e analisi pollinica dei sedimenti olocenici dell’antico
lago di Agordo (Dolomiti Orientali). Il Quaternario 14(2),
187-198.
de Beaulieu J.-L., 1977. Contribution pollenanalytique à
l’histoire tardiglaciaire et Holocène de la végétation des
Alpes méridionales françaices. Ph.D. Dissertation. Université
d’Aix-Marseille, Marseille, France.
Di Rita F. & Magri D., 2009. Holocene drought,
deforestation and evergreen vegetation development in the
central Mediterranean: a 5500 year record from Lago Alimini
Piccolo, Apulia, southeast Italy. The Holocene 19(2),
295-306.
Di Rita F., Celant A., Magri D., 2010. Holocene environmental
instability in the wetland north of the Tiber delta (Rome,
Italy): sea-lake-man interactions. Journal of Paleolimnology
44, 51-67.
Di Rita F., Simone O., Caldara M., Gehrels W.R., Magri D.,
2011. Holocene environmental changes in the coastal
Tavoliere Plain (Apulia, southern Italy): a multiproxy
approach. Palaeogeography, Palaeoclimatology, Palaeoecology,
submitted.
Caramiello R. A., Siniscalco C., Savant Ros I., 1995. Analisi
palinologica della torbiera del Lac Falin (alta valle di Viù,
Alpi occidentali). Giornale Botanico Italiano 129(2), 233.
Caroli I. & Caldara M., 2006. Vegetation history of Lago
Battaglia (eastern Gargano coast, Apulia, Italy) during the
middle-late Holocene. Vegetation History and Archaeobotany
16(4), 317-327.
Drescher-Schneider R., de Beaulieu J.-L., Magny M.,
Walter-Simonnet A.-V., Bossuet G., Millet L., Brugiapaglia E.,
Drescher A., 2007. Vegetation history, climate and human
impact over the last 15,000 years at Lago dell’Accesa
(Tuscany, Central Italy). Vegetation History and
Archaeobotany 16, 279-299.
Celant A. & Magri D., 1999. Archaeobotanical study of a
sediment core from the Colosseum valley. Congresso della
Società Botanica Italiana (p. 40). Ferrara.
Fedele F. G. & Wick L., 1996. Glacial/Postglacial transition
South of Splügen Pass: environment and human activity. Il
Quaternario 9(1), 541
Colombaroli D., Marchetto A., Tinner W., 2007. Long-term
interactions between Mediterranean climate, vegetation and
fire regime at Lago di Massaciuccoli (Tuscany, Italy). Journal
of Ecology 95, 755-770.
Filippi M. L., Heiri O., Arpenti E., Angeli N., Bortolotti M.,
Lotter A.F., van der Borg K., 2005. Studio paleolimnologico
del Lago Nero di Cornisello (Parco Naturale AdamelloBrenta, Trentino). Studi Trentini di Scienze Naturali, Acta
Geologica 82, 261-278.
Cremaschi M., Pizzi C., Valsecchi V., 2006. Water
management and land use in the terramare and a possibile
climatic co-factor in their abandonment: The case study of
the Terramara of Poviglio Santa Rosa (northern Italy).
Quaternary International 151, 87-98.
Cruise G.M., 1990. Holocene peat initiation in the Ligurian
Apennines, northern Italy. Review of Palaeobotany and
Palynology 63, 173-182.
Cruise G.M., Macphail R.I., Linderholm J., Maggi R.,
Marshall P.D., 2009. Lago di Bargone, Liguria, N Italy:
a reconstruction of Holocene environmental and land-use
Filippi M. L., Heiri O., Arpenti E., Angeli N., Bortolotti M.,
Lotter A. F., van der Borg K., 2005b. Evoluzione
paleoambientale dal Tardoglaciale a oggi ricostruita
attraverso lo studio dei sedimenti del Lago di Lavarone
(Altopiano di Folgaria e Lavarone, Trentino). Studi Trentini
di Scienze Naturali, Acta Geologica 82: 279-298.
Finsinger W., 2001. Vegetation history and human impact at
the Lago del Vei del Bouc (Argentera Massif, Maritime
Alps). Quaternaire 12, 223-233.
Finsinger W. & Tinner W., 2006 . Holocene vegetation and
21
56
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
land-use changes in response to climatic changes in the
forelands of the southwestern Alps, Italy. Journal of
Quaternary Science 21, 243-258.
im Fagetum des Bosco del Cansiglio. Pollenanalitische
Untersuchungen in den Venetianer Alpen. Mitt. OstalpinDinarischen Pflanzensoziol. Arbeitsgem. 9, 261-273.
Follieri M., Magri D., Sadori L., 1988. 250,000-year pollen
record from Valle di Castiglione (Rome). Pollen et Spores
30, 329-356.
Kral F., 1980. Zur postglazialen Vegetationsgeschichte am
Südrand der Ostalpen. Pollenanalytische Untersuchungen im
Val-Sugana-Umkreis. Botanische Jahrbücher für Systematik
101, 575 - 593.
Gehrig R., 1997. Pollenanalytische Untersuchungen zur
Vegetations- und Klimageschichte des Val Camonica
(Norditalien). Dissertationes Botanicae 276, 1-152.
Giardini M., 2007. Late Quaternary vegetation history at
Stracciacappa (Rome, central Italy). Vegetation History and
Archaeobotany 16, 301-316.
Gobet E., Tinner W., Hubschmid P., Jansen I., Wehrli M.,
Ammann B., Wick L., 2000. Influence of human impact and
bedrock differences on the vegetational history of the
Insubrian Southern Alps. Vegetation History and
Archaeobotany 9, 175–187.
Grüger E. & Thulin B., 1998. First results of biostratigraphical
investigations of Lago d’Averno near Naples relating to the
period 800 BC-800 AD. Quaternary International 47/48,
35-40.
Guido M.A., Mariotti Lippi M., Menozzi B.I., Placerani S.,
Montanari C., 2004. Il paesaggio vegetale montano della
Liguria centro-occidentale nell’età del Ferro: area del monte
Beigua (Savona). In R.C. de Marinis & G. Spadea (eds),
I Liguri, 91-95, Skira, Ginevra-Milano.
Guido M. A., Mariotti Lippi M., Menozzi B. I., Placerani S.,
Montanari C., 2004b. Ambienti costieri nella riviera ligure di
Levante tra le età del Bronzo e del Ferro: aree di Rapallo e di
Chiavari. In R.C. de Marinis & G. Spadea (eds), I Liguri,
78-81, Skira, Ginevra-Milano.
Kaltenrieder P., Belis C.A., Hofstetter S., Ammann B.,
Ravazzi C., Tinner W., 2009. Environmental and climatic
conditions at a potential Glacial refugial site of tree species
near the Southern Alpine glaciers. New insights from
multiproxy sedimentary studies at Lago della Costa
(Euganean Hills, Northeastern Italy). Quaternary Science
Reviews 28, 2647-2662.
Kaltenrieder P., Procacci G., Vannière B., Tinner W., 2010.
Vegetation and fire history of the Euganean Hills (Colli
Euganei) as recorded by Lateglacial and Holocene
sedimentary series from Lago della Costa (northeastern
Italy). The Holocene 20(5), 679-695.
Kelly M. G. & Huntley B., 1991. An 11000-year record of
vegetation and environment from Lago di Martignano,
Latium, Italy. Journal of Quaternary Science 6, 209-224.
Kral F., 1969. Zur Frage der natürlichen Fichtenverbreitung
Kral F., 1986. Zur postglazialen Vegetationsgeschichte
in den südlichen Ostalpen. Botanische Jahrbücher für
Systematik 106, 409-417.
Lowe J.J., 1992. Lateglacial and early Holocene lake
sediments from the northern Appennines, Italy—pollen
stratigraphy and radiocarbon dating. Boreas 21, 193-208.
Lowe J., Accorsi C.A., Bandini Mazzanti M., Bishop A., van
der Kaars S., Forlani L., Mercuri A.M., Rivalenti C., Torri P.,
Watson C., 1996. Pollen stratigraphy of sediment sequences
from lakes Albano and Nemi, (near Rome) and from the
central Adriatic, spanning the interval fromoxygen isotope
stage 2 to the present day. In: Guilizzoni, P. and Oldfield,
F. (Eds.). Palaeoenvironmental Analysis of Italian Crater lake
and Adriatic Sediments (PALICLAS). Memorie dell’Istituto
Italiano di Idrobiologia 55, 71-98.
Magri D., 1999. Late Quaternary vegetation history at
Lagaccione near Lago di Bolsena (central Italy). Review of
Palaeobotany and Palynology 106, 171-208.
Magri D. & Sadori L., 1999. Late Pleistocene and Holocene
pollen stratigraphy at Lago di Vico (central Italy). Vegetation
History and Archaeobotany 8, 247-260.
Mariotti Lippi M., Guido M., Menozzi B.I., Bellini C.,
Montanari C., 2007. The Massaciuccoli Holocene pollen
sequence and the vegetation history of the coastal plains by
the Mar Ligure (Tuscany and Liguria, Italy). Vegetation
History and Archaeobotany 16, 267-277.
Mercuri A.M., Accorsi C.A., Bandini Mazzanti M., 2002.
The long history of Cannabis and its cultivation by the
Romans in central Italy, shown by pollen records from
Lago Albano and Lago di Nemi. Vegetation History and
Archaeobotany 11 (4), 263-276.
Mercuri A.M., Accorsi C.A., Bandini Mazzanti M., Bosi G.,
Trevisan Grandi G., Cardarelli A., Labate D., Olmi L. , Torri
P., 2006 . Cereal fields from the Middle-Recent Bronze Age,
as found in the Terramara di Montale, in the Po Plain (Emilia
Romagna, Northern Italy), based on pollen, seeds/fruits and
microcharchoals. In: J.P. Morel, Tresserras J. and Matalama
J.C. (ed.). The Archaeology of Crop Fields and Gardens.
Proceedings of the first conference on crop fields and garden
archaeology (Barcelona-ES, 1-3- June 2006), Centro Studi
Europei per I Beni Culturali, Ravello, Edipuglia, pp.
22
57
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
251-270.
Miola A., Bondesan A., Corani L., Favaretto S., Mozzi P.,
Piovan S., Sostizzo I., 2006. Wetlands in the Venetian Po
Plain (northeastern Italy) during the Last Glacial Maximum:
Interplay between vegetation, hydrology and sedimentary
environment. Review of Palaeobotany and Palynology
141(1-2), 53-81.
Moe D., Fedele F.G., Engan Maude A., Kvamme M., 2007.
Vegetational changes and human presence in the low-alpine
and subalpine zone in val Febbraro, upper Valle di Spluga
(Italian central Alps), from the Neolithic to the Roman
period. Vegetation History and Archaeobotany 16(6),
431-451.
vegetation and climate history of the last 215 ka in the
Azzano Decimo core (plain of Friuli, north-eastern Italy).
Quaternary Science Reviews 28(13-14), 1268-1290.
Ravazzi C., 2002. Late Quaternary history of spruce in
southern Europe. Review of Palaeobotany and Palynology
120, 131-77.
Ravazzi C., Donegana M., Vescovi E., Arpenti E., Caccianiga
M., Kaltenrieder P., Londeix L., Marabini S., Mariani S.,
Pini R., Vai G. B., Wick L., 2006. A new Late-glacial site
with Picea abies in the northern Apennine foothills: an
exception to the model of glacial refugia of trees. Vegetation
History and Archaeobotany 15(4), 357-371.
Monegato G., Ravazzi C., Donegana M., Pini R., Calderoni
G., Wick L., 2007. Evidence of a two-fold glacial advance
during the last glacial maximum in the Tagliamento end
moraine system (eastern Alps). Quaternary Research 68,
284-302.
Ricci Lucchi M., Calderoni G., Carrara C., Cipriani N.,
Esu D., Ferreli L., Girotti O., Gliozzi E., Lombardo M.,
Longinelli A., Magri D., Nebbiai M., Ricci Lucchi F.,
Vigliotti L., 2000. Late Quaternary record of the Rieti basin,
central Italy: paleoenvironmental and paleoclimatic
evolution. Giornale di Geologia 62, 105-136.
Montanari C., Guido M.A., Cornara L., Placereani S., 1997.
Tracce polliniche di boschi neolitici di abete bianco in Val
Bisagno (area urbana di Genova). Biogeographia 19,
133-143.
Ricci Lucchi M., Aguzzi M., Amorosi A., Rossi V., Vaiani S.
C., 2006. Holocene climatic changes on the western coast of
Italy (Arno coastal plain, Tuscany). Geophysical Research
Abstracts 8, 06304.
Mori Secci M., 1996. Vicende oloceniche dell’Appennino
Tosco-Emiliano ricostruite attraverso le analisi palinologiche.
Webbia 51, 83-120.
Sadori L. & Narcisi B., 2001. The postglacial record of
environmental history from Lago di Pergusa (Sicily). The
Holocene 11, 655-671.
Noti R., van Leeuwen J.F.N., Colombaroli D., Vescovi E.,
Pasta S., La Mantia T., Tinner W., 2009. Mid- and
late-Holocene vegetation and fire history at Biviere di Gela,
a coastal lake in southern Sicily, Italy. Vegetation History and
Archaeobotany 18(5), 371-387.
Sadori L., Giraudi C., Petitti P., Ramrath A., 2004. Human
impact at Lago di Mezzano (central Italy) during the Bronze
Age: a multidisciplinary approach. Quaternary International
113, 5-17.
Ortu E., Brewer S. & Peyron O., 2006 - Pollen-inferred
palaeoclimate reconstructions in mountain areas: problems
and perspectives. Journal of Quaternary Science 21: 615-627.
Ortu E., David F. & Caramiello R., 2003. Role of man in the
history of vegetation of the Ellero valley (maritime Alps,
Italy). C R Biologies 326(7), 631-637.
Ortu E., Peyron O., Bordon A., de Beaulieu J.-L., Siniscalco
C., Caramiello R., 2008. Lateglacial and Holocene climate
oscillations in the South-western Alps: An attempt at
quantitative reconstruction. Quaternary International 190,
71-88.
Pini R., 2002. A high-resolution Late-Glacial - Holocene
pollen diagram from Pian di Gembro (Central Alps, Northern
Italy). Vegetation History and Archaeobotany 11, 251-262.
Pini R., 2004. Late Neolithic vegetation history at the
pile-dwelling site of Palù di Livenza (northeastern Italy).
Journal of Quaternary Science 19, 769-781.
Pini R., Ravazzi C., Donegana M., 2009. Pollen stratigraphy,
Sadori L., Giardini M., Giraudi C., Mazzini I., 2010. The
plant landscape of the imperial harbour of Rome. Journal of
Archaeological Science 37(12), 3294-3305.
Scaife R., 1997. Pollen analysis of the Laghetti del Crestoso
corrie basin. In C. Baroni & P. Biagi (eds) Excavations at the
high altitude mesolithic site of Laghetti del Crestoso
(Bovegno, Brescia – northern Italy), 64-77. Ateneo,
Accademia di scienze lettere ed arti. Brescia, Italy.
Schneider H., Höfer D., Irmler R., Daut G., Mäusbacher R.,
2010. Correlation between climate, man and debris flow
events - A palynological approach. Geomorphology 120,
48-55.
Schneider R.E., 1978. Pollenanalytische Untersuchungen zur
Kenntnis der spät- und postglazialen Vegetationsgeschichte
am Südrand der Alpen zwischen Turin und Varese (Italien).
Botanische Jahrbücher für Systematik 100, 26-109.
Schneider R., 1985. Analyse palynologique dans
l’Aspromonte en Calabre (Italie meridionale). Cahiers ligures
23
58
DI DOMENICO F. / Ann. Bot. (Roma), 2011, 1: 45–58
de préhistoire et de protohistoire n.s. 2, 279-288.
Schneider R. & Tobolski K., 1985. Lago di Ganna - LateGlacial and Holocene environments of a lake in the Southern
Alps. Dissertationes Botanicae 8, 229-271.
Seiwald A., 1980. Beiträge zur Vegetationsgeschichte Tirols
IV: Natzer Plateau - Villanderer Alm. Ber. nat.-med. Verein
Innsbruck 67, 31-72.
Serandrei-Barbero R., Bertoldi R., Canali G., Donnicia S.,
Lezziero A., 2005. Paleoclimatic record of the past 22,000
years in Venice (Northern Italy): Biostratigraphic evidence
and chronology. Quaternary International 140-141, 37-52.
Speranza A., Ravazzi C., Barini C., Carton A., van Geel B.,
Mommersteeg H., 1996. Holocene vegetation development
and human impact in the central Alps: the “Pian Venezia”
palaeobotanical record (Trento, Italy). Il Quaternario 9,
737-744.
Tinner W., van Leeuwen J. F. N., Colombaroli D., Vescovi
E., van der Knaap W. O., Henne P. D., Pasta S., D’Angelo
S., La Mantia T., 2009. Holocene environmental and climatic
changes at Gorgo Basso, a coastal lake in southern Sicily,
Italy. Quaternary Science Reviews 28, 1498-1510.
Valsecchi V., Finsinger W., Tinner W., Ammann B., 2008.
Testing the influence of climate, human impact and fire on
the Holocene population expansion of Fagus sylvatica in the
southern Prealps (Italy). The Holocene 18, 603-614.
Valsecchi V., Tinner W., Finsinger W., Ammann B., 2006.
Human impact during the Bronze Age on the vegetation at
Lago Lucone (northern Italy). Vegetation History and
Archaeobotany 15, 99-113.
Vescovi E., Ravazzi C., Arpenti E., Finsinger W., Pini R.,
Valsecchi V., Wick L., Ammann B., Tinner W., 2007.
Interactions between climate and vegetation during the
Lateglacial period as recorded by lake and mire sediment
archives in Northern Italy and Southern Switzerland.
Quaternary Science Reviews 26, 1650-1669.
Watson C. S., 1996. The vegetational history of the northern
Apennines, Italy: information from three new sequences and
a review of Regional vegetational change. Journal of
Biogeography 23, 805-841.
Wick L. & Mohl A., 2006. The mid-Holocene extinction of
silver fir (Abies alba) in the Southern Alps a consequence of
forest fires? Palaeobotanical records and forest simulations.
Vegetation History and Archaeobotany 15(4), 435-444.
Wick L., 1996. Late-Glacial and early Holocene palaeoenvironments in Brianza, northern Italy. Il Quaternario 9,
653-660.
24
Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
Contents lists available at SciVerse ScienceDirect
Perspectives in Plant Ecology, Evolution and Systematics
journal homepage: www.elsevier.de/ppees
Research article
Buxus in Europe: Late Quaternary dynamics and modern vulnerability
Francesco Di Domenico a,∗ , Fernando Lucchese a , Donatella Magri b
a
b
Dipartimento di Biologia Ambientale, Università degli Studi Roma Tre, Viale G. Marconi 446, Rome, Italy
Dipartimento di Biologia Ambientale, Sapienza Università di Roma, Piazzale Aldo Moro, 5, 00185 Rome, Italy
a r t i c l e
i n f o
Article history:
Received 27 February 2012
Received in revised form 26 June 2012
Accepted 3 July 2012
Keywords:
Population increase
Plant traits
Conservation
Distribution
Last glacial period
Refugia
a b s t r a c t
The suggested location of broadleaved evergreen trees in Europe during the last full-glacial has traditionally favoured a southerly refugial model, which proposes survival in the Mediterranean peninsulas
and recolonization of central and northern Europe during the Holocene. This hypothesis is not always
substantiated by thorough reviews of original past and modern occurrence data, or considered in the
light of plant traits and autoecology. Our approach focuses on the genus Buxus with the aim of exploring (i) the relationship between the location of refugia and post-glacial population dynamics, (ii) past
processes determining density, fragmentation and local extinctions of modern populations, and (iii) the
vulnerability of Buxus in the context of the undergoing environmental changes. We compiled a database
of over 3600 modern occurrences and 676 fossil sites to reconstruct the distribution of Buxus in Europe
since 30 ka cal BP. The location of fossil finds and the plant traits of Buxus indicate that it persisted widely
across its modern distribution through the last glacial period with modes varying from region to region.
The E Pyrenees, W Alps, and Jura Mts hosted dense populations, which expanded exponentially during the
whole Holocene, and resulted in a modern continuous distribution area. In contrast, the Mediterranean
Peninsulas hosted sparse populations, which increased exponentially only during the first half of the
Holocene, clearly decreased in the last 4.5 ka BP and resulted in a highly fragmented modern distribution
area, most likely in relation to the climate trends towards dry conditions of the last few millennia. These
results challenge the common view that the Mediterranean regions are the exclusive and most important
refuge areas for evergreen broadleaved trees and stress the importance of considering long-term population dynamics based on fossil data to evaluate the vulnerability of modern fragmented plant populations
in view of conservation actions.
© 2012 Elsevier GmbH. All rights reserved.
Introduction
The suggested location of broadleaved evergreen trees in Europe
during the last full-glacial has traditionally favoured a southerly
refugial model, which proposes survival in the three Mediterranean
peninsulas and recolonization of central and northern Europe during the Holocene. Modern studies involving fossil and genetic data
(Magri et al., 2006; Willis and Van Andel, 2004) indicate that some
broadleaved deciduous (e.g. Fagus sylvatica L., Corylus avellana L.)
and needleleaved evergreen (e.g. Taxus baccata L., Pinus sylvestris L.,
Juniperus communis L.) trees may have survived in central Europe.
No reports for northerly persistence of evergreen broadleaved
woody taxa (e.g. Ilex aquifolium L., Hedera helix L., Buxus sempervirens L.) are known, and in the common view such species were
∗ Corresponding author.
E-mail address: [email protected] (F. Di Domenico).
confined to the Mediterranean Peninsulas during the last glacial
period (Bennett et al., 1991; Bhagwat and Willis, 2008).
Recently, some authors showed that biogeographical plant traits
of woody species play an important role in determining the chances
of local persistence through the last glacial period in central
and northern Europe (Bhagwat and Willis, 2008). In particular, it
was shown that species with a full-glacial distribution including
northerly locations were wind-dispersed, habitat-generalist trees
with the ability to reproduce vegetatively.
On these premises, we decided to study the history of Buxus
L., a genus with two shrub/tree (up to 15 m) species in Europe,
B. sempervirens L. and B. balearica Lam. The genus has a long history in Europe, as witnessed by a continuous fossil record since
the Miocene, through the Pliocene and Early Pleistocene (Kvaček
et al., 1982; Leroy and Roiron, 1996). A few studies deal with its
Holocene history (Lang, 1992; Wegmüller, 1984; Yll et al., 1997; Di
Domenico et al., 2011), but they are either limited in time and/or
space or require updating with the large amount of data published
in the last 20 years.
1433-8319/$ – see front matter © 2012 Elsevier GmbH. All rights reserved.
http://dx.doi.org/10.1016/j.ppees.2012.07.001
25
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
In this paper, a reconstruction of the late-glacial and Holocene
history of Buxus in Europe is presented using modern and past distribution data and interpreted in the light of the ecological features,
with the aim of answering the following questions:
• is the modern distribution of Buxus the result of a post-glacial
migration from limited glacial refugia? Buxus is often considered a thermophilous (sub-)mediterranean woody taxon. From
the refugia in the Mediterranean regions and in the southern and
western Alps Buxus would have recolonized the northern parts of
its modern distribution area (Lang, 1992; Wegmüller, 1984). The
wealth of recently published modern and fossil data of Buxus may
contribute defining times and modes of its postglacial expansion.
• are there regional differences in the Holocene dynamics of Buxus
populations in Europe? Buxus is presently distributed in both central and southern Europe. The combined use of modern and past
distribution data may highlight and explain similarities and dissimilarities in the behaviour of populations located in different
bioclimatic areas.
• which are the different roles played by natural population
dynamics, climate change and human impact in shaping the European distribution of Buxus in the last 15,000 years? Considering
that the modern distribution of plant species is the result of a
combination of factors, including location of glacial refugia and
ecological responses to post-glacial climate changes and human
activities, it may be of interest to evaluate which factor was especially influential on Buxus.
• is Buxus a vulnerable genus in Europe? While B. balearica is
considered vulnerable in some parts of its distribution range
(Blanca, 1999), no conservation actions focus on B. sempervirens,
apart from weak regulations at local or regional levels in central
Europe. The post-glacial history of Buxus may help distinguish
areas where the populations are vigorous and do not require any
specific conservation action from areas where they are weak and
demand special attention in the context of the ongoing climate
change and increasing human impact.
The combined analyses of modern distribution and fossil
records may concur to define where Buxus is “natural” in Europe
and serve as a scientific basis on which to base conservation actions
in areas where Buxus populations are at risk. In this sense, they contribute to the practical application of palaeoecological studies to
long-term biodiversity maintenance, ecosystem naturalness, conservation evaluation, habitat alteration, and changing disturbance
regimes (Willis and Birks, 2006).
Materials and methods
A thorough survey of the modern distribution data of Buxus has
been carried out, including more than 3600 occurrences. Online
Vegetation Databases were queried for the distribution of Buxus
in Portugal, Spain, France, Germany, Switzerland, and Austria
(Appendix S1). A number of sites (0.7%) that appeared incorrectly
georeferenced were checked with the original source. The distribution of Buxus in Italy follows Di Domenico et al. (2011). For all
the other European countries, the distribution was reconstructed
using (i) original field data, (ii) national and regional floras, and (iii)
papers reporting the natural occurrence of Buxus (Appendix S1).
The distributions of B. sempervirens and B. balearica are represented
in Fig. 1 by green and orange squares, respectively.
To reconstruct the Holocene history of Buxus, we compiled a
pollen database consisting of 650 sites throughout Europe (Fig. 1
and Appendix S2). The characteristics of Buxus pollen are unmistakable (Beug, 2004; Wegmüller, 1984), so misidentifications in
355
the reviewed palynological studies are unlikely. Considering the
low frequency of Buxus pollen and its short dispersal distances
(Cañellas-Boltà et al., 2009), we assumed that the occurrence of
Buxus pollen indicates its presence in the surrounding vegetation.
Records of Buxus pollen were extracted from published
diagrams, and ages were determined with the associated
depth-age model, when available, or by linear interpolation
between radiocarbon ages. All dates were calibrated to calendar years before present (cal BP) using the online program
Calib 6.0 (http://calib.qub.ac.uk/calib/calib.html). Since many published diagrams show only selected pollen taxa, the full list
of species was checked in the European Pollen Database
(www.europeanpollendatabase.net), when available. The complete list and references of the reviewed pollen sites are reported
in Appendix S2. The presence or absence of Buxus pollen is represented in Fig. 2 by red or yellow dots, respectively.
In addition to pollen data, the past distribution of Buxus was
enhanced by a macrofossil dataset (wood and leaves) derived from
the published literature (26 sites). The complete list and references of the macrofossil sites are presented in Appendix S3. The
chronological setting of the macrofossils is based on radiocarbon
measurements at all the sites. The macrofossil sites of Buxus are
represented in Fig. 2 by red triangles.
Modern and past distribution data were processed in GIS environment using the Mollweide Equal-Area Projection (Figs. 1 and 2).
The Bioclimatic Map of Europe (Rivas-Martínez et al., 2004;
http://www.globalbioclimatics.org/) was used to assign each site
to a Macro-Bioclimatic zone (Fig. 1).
Distribution, ecology and plant traits of Buxus in Europe
Distribution and ecology
B. sempervirens L. has a gregarious and locally abundant distribution but it is also absent in vast areas where suitable habitats
are present (Tutin et al., 1968). It shows a centre of continuity,
abundance and frequency in the Pyrenees, Southern France, French
Prealps and all around the Jura Mts (Fig. 1). Fragmented populations
are found in the southern European Peninsulas (Iberian, Italian,
and Balkan) and in central France. Exiguous stands are located
in Britain, Belgium, Luxembourg, Germany, Sardinia, Montenegro,
and Kosovo. B. sempervirens also occurs, outside the studied area,
in Morocco, Algeria, Turkey, Georgia, Iran and Kazakhstan. B. sempervirens is a species with a wide ecological niche, being generally
found between sea-level and 2000 m (e.g. Pyrenees, Mount Olympus, Pindos). It mainly lives on limestone, but it may also be found
on ophiolite and volcanic tuff. B. sempervirens is often found along
slopes, river valleys, canyons, gorges, ravines, and thermal springs,
which provide suitable sub-humid conditions. It is present in a
wide range of vegetation types, such as deciduous and evergreen
broadleaved forests, evergreen needled woodlands, garigues, and
calcareous grasslands. Calcareous grasslands with B. sempervirens
on rock slopes (Berberidion p.p.) are a natural habitat type of community interest which requires the designation of special areas of
conservation (Natura 2000 code 5110).
B. balearica Lam. shows a distribution restricted to the Balearic
Islands and to southern Spain, with a single population located in
Sardinia (Tutin et al., 1968). Besides, it occurs in Morocco, Algeria and Turkey. It is found on limestone (Balearic Islands, Sardinia)
and on the metamorphic crystalline rocks with ultramafic inclusions of the Alboran Block (southern Spain). Despite its restricted
distribution, B. balearica occurs in both deciduous and evergreen
broadleaved forests (Balearic Islands, Sardinia) as well as in evergreen needled woodlands with Pinus halepensis Mill. (southern
26
356
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
Fig. 1. Modern distribution of Buxus in Europe: B. sempervirens is represented in dark green, and B. balearica in orange. The yellow dots represent the reviewed fossil sites (cf.
Supplementary Information S2 and S3). Red triangles represent fossil records between 30 and 15 ka cal BP. The Mediterranean (pink) and Temperate (pale green) bioclimatic
zones follow Rivas-Martínez et al. (2004).
Spain). Also B. balearica is most often found in canyons, river valleys
and rocky cliffs, in sub-humid conditions. The International Union
for Conservation of Nature (IUCN) lists the severely reduced populations of B. balearica as vulnerable (VU), and at risk of extinction
in Andalusia (Blanca, 1999).
The common denominator of the wide range of habitats where
both B. sempervirens and B. balearica occur is a high level of air
moisture. In most cases, Buxus is found not far from rivers or water
basins, but the presence of water alone does not fulfil the optimal requirements of Buxus. Mechanisms of moisture entrapment
are always present in environments hosting large Buxus populations. For example, in gorges and ravines the presence of water
and the effects of thermal inversion act jointly in maintaining subhumid and relatively stable conditions. In wide fluvial valleys, an
adequate level of air humidity is granted by large amounts of flowing waters, which may be vaporized thanks to emerging rocks,
drops and waterfalls. In supra-mediterranean garigues, most often
located in intermountain basins, Buxus is able to live thanks to the
stagnation of mist. Around thermal springs, significant amounts of
moisture are granted by the high degree of water evaporation in
relation to high temperatures.
Plant traits
B. sempervirens shows a unique and well-known set of plant
traits, which are important to enhance or reduce the long-term
possibility of survival and determine its diffusion during past and
future changes of climate and habitat (e.g. Barboni et al., 2004;
Bhagwat and Willis, 2008). Plant traits may explain the persistence
of Buxus over Europe in the last millions of years (Kvaček et al.,
1982), and elucidate its modern geographical distribution pattern.
The leaves of Buxus present a very well-adapted photoprotective
mechanism in which the same retro-carotenoids can be activated
in response to both winter and summer light stress (Hormaetxe
et al., 2007). The plasticity of such a mechanism, induced in both
supra- and suboptimal conditions, plays a critical role in plant acclimation to extreme temperature. This photoprotective mechanism
was found to be plastic enough to compensate for seasonal changes
in irradiance related with overstorey canopy phenology, thereby
reducing the risk of winter photoinhibition and optimizing the
annual carbon gain even under light limitation (García-Plazaola
et al., 2008). This means that Buxus is a shade tolerant species.
The leaves of Buxus are also resilient to frost, thanks to their
ability to form ice lenses in the mesophyll lacunas during frost
periods (Hacker and Neuner, 2007; Hatakeyama and Kato, 1965;
Montemartini, 1907). Thus, the leaves of Buxus are well adapted
to cold stress and avoid frost damages by lowering their freezing
point below air temperature (Hatakeyama and Kato, 1965). Another
important feature of the leaves of Buxus is their ability to increase
the osmotic value of cell sap up to extreme values, a factor influencing plant resistance to cold stress. While many species, some
of which share ecological features with Buxus (e.g. I. aquifolium),
are able to increase the density of their cell sap up to 20 atm but
die at 25 atm, Buxus is able to increase its concentration from 33 to
73 atm without any injury (Walter, 1929). In comparison to other
plant species, Buxus has a low uptake of toxic metals (Karataglis
et al., 1982), being able to establish large populations on ultramafic
soils, as evidenced by its presence on ophiolites in Italy, Greece and
27
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
Fig. 2. Geographical distribution of the fossil records of Buxus in Europe during the last 15 ka cal BP: , Buxus pollen absent;
Green and orange squares correspond to the modern distribution of B. sempervirens and B. balearica, respectively.
Turkey. Ophiolites are usually poor in species composition, due to
the high concentration of heavy toxic metals (Cr, Ni, Co, Fe, Mg),
providing a relatively competition-free substrate.
Other important plant traits are related to the reproductive biology of Buxus. Both B. sempervirens and B. balearica are long-lived and
extremely slow growing. It is not uncommon to observe individuals
500 years old. Buxus is able to reproduce vegetatively from broken
, Buxus pollen present;
357
, Buxus macrofossil.
or buried branches. This is a very well known feature which makes
the taxon a target of great ornamental interests.
The dispersal of Buxus consists mainly of an ejection mechanism that may suffice for localized seed dispersal. Abiotic agents
such as rain and flowing water may account for dispersal over
longer distances (Köhler, 2007). Myrmechocory was observed in B.
sempervirens (Köhler, 2007) and B. balearica (Lázaro and Traveset,
28
358
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
2005), but few ant species are effective dispersers while most are
consumers. Detailed studies of the seed dispersal distances of Buxus
reports a mean range of 1.5 m yr−1 and a maximum of 3 m yr−1
(Debussche and Lepart, 1992).
Late Quaternary history of Buxus in Europe
Fossil pollen and macroremains of Buxus were found in 204 out
of the 676 sites reviewed in the present study. In Figs. 1 and 2
different time windows of the Late Quaternary history of Buxus are
presented.
During the last glacial period (30–15 ka), sites with Buxus fossils (henceforth ‘Buxus sites’) are found in the Iberian Peninsula,
Balearic Islands (Minorca), E Pyrenees, Jura Mts, Italian Peninsula,
E Adriatic Coast, and central Greece (Fig. 1). In the late-glacial period
(15–12 ka) they increased in the Jura Mts, Italy and Greece (Fig. 2).
The beginning of the Holocene (12–10 ka) is characterized by an
overall increase of Buxus sites in the Pyrenees and Italy, and new
appearances in central France and Corsica. From 10 to 8 ka Buxus
was more frequent in the Pyrenees, Jura Mts and Corsica. Besides,
it appeared in Sicily, S and NW France, W Alps and S Germany. This
increasing trend continued at 8–6 ka, and unprecedented discoveries are found in N France and S England. The widespread increase
of Buxus sites culminated at 6–4 ka, with remarkable increments
especially in C France, SE Spain, and Greece. Between 4 and 2 ka,
there was a reduction in the number of sites in S Spain and S Greece,
whereas Buxus still increased elsewhere in Europe (Fig. 2). The
reduction in S Europe continued between 2 ka and the present time.
In the last thousand years this decrease was especially marked, so
that no Buxus site located in the modern range of B. sempervirens is
found south of 40◦ N. A clear decline is evident also in the modern
range of B. balearica, where Buxus pollen is found only in two pollen
records, in Mallorca and Sierra Nevada, respectively.
Regionally, the history of Buxus appears rather diversified:
Portugal. The past occurrence of Buxus in Portugal is documented only in two sites by fossil woods during the pleniglacial
(28 and 21.5 ka BP), and around 3 ka BP (Figueiral and Terral, 2002; Figueiral and Bettencourt, 2004). The Portuguese
populations have been much reduced during the whole Holocene,
as at present.
E Spain and Balearic Islands. The occurrence of Buxus during the last
glacial period is documented at Navarrés around 30 ka BP (Carrión
and Van Geel, 1999) and in Minorca around 20 ka cal BP (Yll et al.,
1997). Buxus was continuously present in E Spain and the Balearic
Islands, but after ca. 5 ka cal BP a clear decrease in fossil finds
became manifest, culminating with a high degree of fragmentation of the continental populations and the modern absence of
Buxus from Ibiza and Minorca.
Pyrenees, S France. In the E Pyrenees Buxus is recorded at approximately 30 ka BP at Banyoles (Pérez-Obiol and Julià, 1994), while
it showed up relatively late in S France. The number of Buxus sites
continuously increased during the Holocene, so that the modern
distribution of B. sempervirens is almost continuous and with rich
populations.
Massif Central. Buxus pollen is found at several sites throughout
the Holocene, starting around 9 ka cal BP and ending around 1 ka
cal BP, although Denèfle et al. (1980) relate its appearance to the
Gallo-Roman utilization.
NW France. Various authors consider Buxus native in the area,
including the Massif Armoricain, on the basis of its early (between
10 and 9 ka cal BP at Jaunay), continuous and widely distributed
finds throughout the Holocene (Ouguerram and Visset, 2001;
Cyprien et al., 2004; Visset and Bernard, 2006; Joly and Visset,
2009).
England. Single pollen grains from three sites suggest a sparse presence since ca. 7 ka cal BP (Oldfield and Statham, 1963; Bartley and
Morgan, 1990; Waller and Hamilton, 2000). Macroremains from
the Roman period are related to coffin lining in burial practices at
Roden Downs (Allison, 1947). Since there appear to be no records
of lining coffins with Buxus in Roman burials on the ‘Continent’, the
contention that the tree was introduced by the Romans to serve
such a task has little support (Godwin, 1975). These finds, scattered in space and time, are unlikely to be the result of repeated
contamination episodes and support the native status of Buxus in
England. At present, Buxus is regarded as naturally occurring at
Box Hill (Surrey), Ellensborough (Buckinghamshire) and Boxwell
(Cotswolds), where large populations occur on chalk or oolite
escarpments (Wigginton, 1999).
Vosges. Buxus pollen is found in only two sites in the last 2 ka BP.
Buxus was present during the last interglacial at La Grande Pile,
but it disappeared before the last glacial maximum (de Beaulieu
and Reille, 1992).
W Alps, Jura Mts and Swiss Alpine forelands. The fossil record of
Buxus starts around 18 ka cal BP at Bibersee (van der Knaap and
van Leeuwen, 2001). The number of Buxus sites in the W Alps and
the Jura Mts continuously increased throughout the post-glacial
period and together with the Pyrenees and S France constituted
the centre of abundance and frequency of B. sempervirens distribution.
Austria. Buxus is reported as a non-native species (Online-Flora
von Österreich: http://flora.vinca.at/), but at some sites in the Enns
and Steyr valleys this statement is questioned (Essl, 2002). Buxus
pollen occurred at three Tyrolese sites between 2.5 and 2 ka BP,
after which time it disappeared.
E Adriatic coast. Sparse Buxus sites are found from the late-glacial
period to a few centuries ago. At present, no natural populations
are reported. The fossil evidence suggests that the Croatian populations were already fragmented during the Holocene, but the
final disjunction between the Balkan and the N Italian populations
is very recent, dating back to the last centuries.
Central and N Italy. The populations located between 41◦ N and
43◦ N were already rather dense during the late-glacial period.
Buxus increased in the course of the Holocene especially in N Italy
(Di Domenico et al., 2011).
S Greece and S Italy. Fossil finds of Buxus are recorded since the
last glacial period. A clear increase in the number of Buxus sites
is observed starting from the late-glacial period. During the last
4500 years Buxus underwent a dramatic reduction, to the point of
becoming locally extinct south of 40◦ N, with the exception of a
few sites in Greece (Pindos, Kerkyra and Euboea Islands).
Discussion
A comprehensive consideration of the plant traits of Buxus and of
its past and modern distributions provides the basis for a discussion
on the location of glacial refugia and Holocene population increase,
and on the mechanisms underlying the marked reduction of the
Mediterranean populations in the last few millennia.
Glacial refugia
According to previous studies, Buxus spread from the Mediterranean peninsulas and the southern and western Alps, reaching
the northern part of its modern distribution by approximately
6 ka BP (Lang, 1992; Wegmüller, 1984). In the light of the wealth
29
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
of fossil data collected in recent years and of the advances of
ecological studies, we propose a revision of this hypothesis.
In the time span 30–15 ka BP, in addition to the sparse fossil
finds located in the Mediterranean Peninsulas, there were already
Buxus sites in the Swiss Alpine forelands, where a progressive
increase during the late-glacial period points to a local presence
rather than to sporadic finds of reworked pollen grains. These data,
together with the late-glacial records, indicate that the glacial refugia of Buxus in central Europe reached a latitude of at least 47◦ N
(Figs. 1 and 2).
By ca. 8.5 ka cal BP, Buxus appeared in NW France at approximately the same latitude, but at over 600 km distance. Around
7 ka cal BP it is found in England. Given the maximum distance
of 3 m yr−1 for seed dispersal in modern populations (Debussche
and Lepart, 1992), dissemination from so far away appears very
unlikely. Buxus might have expanded its range through fluvial
transport, but this mechanism has not been documented in the field
(Köhler, 2007). For example, the Jura Mountain range, where several early pollen occurrences are found, may have been the source
area for downstream migration into E France and Germany through
the Moselle valley. However, fluvial transport cannot easily explain
the appearance of Buxus pollen in NW France and England, as no
rivers connect these areas with the southern populations (Godwin,
1975; Vanden Berghen, 1955). Thus, alternative hypotheses should
be considered.
A relict status for the populations of the northern part of France
and Germany is suggested by the geometrical morphometric analysis of the box midge, Monarthropalpus buxi Laboulbène, showing
that the northern and southern French populations are completely
distinct from each other as a result of long-term isolation (Baylac
and Daufresne, 1996). The presence of Buxus in England (Phillips,
1974), and in the marine core MD04-2845 off W France (M.F.
Sanchez Goñi, personal communication) during the last interglacial
period, being a condicio sine qua non for its survival during the last
glacial period, is another argument supporting a possible northwestern persistence. This hypothesis is also supported by the plant
traits of Buxus, which demonstrate a great adaptability to changing climate, light, soil and community structure, especially with
respect to resilience to frost conditions. Considering the early
appearance and sparse location of the fossil records in NW France
and England, as well as the very limited dispersal abilities of Buxus
and its ecological plasticity, it cannot be excluded that the taxon
survived at the northern fringe of its modern distribution during
the last glacial period at densities too low to be detected by pollen
analysis. In any case, even disregarding NW France and England, the
location of the late-glacial and early Holocene fossil finds suggests
that it persisted widely across its modern distribution in central
Europe (Fig. 2).
Holocene population dynamics
The comparison between the modern distribution and the location of Buxus sites in E Spain during the last 1 ka demonstrates
that exiguous populations are often undetected by pollen analysis (Fig. 2). By contrast, where populations are dense (E Pyrenees),
fossil finds are generally frequent. Thus, the appearance of Buxus
in the fossil record may reasonably reflect a process of local population growth to a detectable degree. If so, the fossil occurrences
between 30 and 12 ka BP indicate that the populations located in
the Betic Mts, Eastern Pyrenees, Jura Mts, Italy, Croatia and Greece
must have been dense enough to be detected. In particular, the
populations located in the E Pyrenees, W Alps, and Jura Mts show a
continuous record of Buxus throughout the Holocene and still host
large populations.
359
Fig. 3. Scatter plot of the number of sites with fossil records of Buxus in Europe
against time. Temperate (triangles) and Mediterranean (circles) populations were
distinguished according to their location in the Bioclimatic Map by Rivas-Martínez
et al. (2004).
Currently, the largest populations of Buxus are located in the
Temperate bioclimatic zone, as defined by Rivas-Martínez et al.,
2004 (Fig. 1). By contrast, the Mediterranean bioclimate zone hosts
mostly fragmented populations (including both B. sempervirens and
B. balearica). With the aim of checking if the different behaviours
of the modern populations in the Temperate and Mediterranean
zones are related to the Holocene population dynamics of Buxus,
the number of Buxus sites in each bioclimatic zone has been plotted
against time (Fig. 3).
The Temperate populations steadily increase from the lateglacial period to the present time, with a continuous and
exponential trend (linearized regression coefficients R2 = 0.95,
F = 22.4, df = 14, p = 0.02). This increase may be considered to reflect
a population growth, as already discussed for other taxa (Magri,
2008). On this basis, the modern populations of Buxus located in
the Temperate zone can be estimated to be over 70 times more
numerous than at 15 ka BP, as a result of a natural biological process. Interestingly, the strong impact of human activities over the
last thousands of years appears to have not influenced at all this
multiplicative process, which is still continuing with undiminished
vigour, even in densely populated areas.
In the Mediterranean bioclimate zone, the postglacial population increase was exponential until approximately 4.5 ka BP, with
a pattern similar to the Temperate zone (Fig. 3). The number of
Mediterranean Buxus sites markedly decreased after 4.5 ka BP,
especially in the last thousand years. The maximum number of
Buxus sites during the postglacial, around 4.5 ka BP, was only 7
times greater than 15 ka BP. Interestingly, the number of Mediterranean Buxus sites in the last thousand years only doubles its
late-glacial abundance.
On the whole, it appears that the Temperate and Mediterranean
bioclimatic zones played a quite different role for the conservation
of Buxus. During the last glacial period, the E Pyrenees, W Alps, and
Jura Mts hosted relatively dense populations, recorded very early
by pollen analysis, which vigorously expanded in the course of the
Holocene, and resulted in a modern continuous distribution area.
By contrast, the refugial Mediterranean areas hosted far less robust
populations, which increased only moderately until 4.5 ka BP, then
30
360
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
declined and resulted in the modern highly fragmented distribution area, which is still undergoing a process of severe reduction.
These results challenge the hypothesis that during the last
glacial period the evergreen broadleaved trees were confined only
to the “southern refugia” and demonstrate that a thorough review
and careful interpretation of the original fossil datasets may lead
to a substantial revision of past plant distributions.
Reduction and vulnerability of the Mediterranean populations
A number of Mediterranean Buxus populations underwent a
dramatic reduction up to local extinction during the last few millennia. They are scattered in different regions, including Minorca,
Ibiza, Sardinia, Sicily, Apulia, Basilicata, Croatia, Thessaly, W Greece,
and Peloponnese. This reduction is even more striking considering
that in these regions Buxus persisted throughout the last glacial
period (Figs. 1 and 2).
The disappearance of Buxus pollen from a site may be reasonably considered to reflect a population decrease to a degree at
which the taxon is not detectable anymore or even extirpated. It
is very difficult to define the causes for this reduction, but the different behaviour of the Temperate and Mediterranean populations
may be the key to suggest an explanation. Climate change, human
impact, and biological processes are the most plausible factors.
In respect of climate, the reduced modern Buxus populations
located in the Mediterranean bioclimatic zone are always found in
moist fluvial valleys, gorges, ravines and around mountain basins,
where climate conditions are particularly stable in the respect of
water availability. This behaviour evidences the strong ecological
link with available moisture (Lenoble and Broyer, 1945; Pigott and
Pigott, 1993). A natural decline of Buxus may have been reasonably enhanced by the progressive aridification process affecting the
Mediterranean regions during the late Holocene, as documented by
various proxy climate records (e.g. Di Rita et al., 2011; Jalut et al.,
2009; Magny et al., 2011; Zanchetta et al., 2007). Conversely, in the
Temperate bioclimatic zone no similar trend towards dry conditions is documented that could hamper the continuous and natural
population growth of Buxus throughout the Holocene.
Concerning human activities, the impact of agriculture may
have only indirectly affected the decline of Buxus on calcareous land, in gorges, and on exposed rocks, which were hardly
exploitable. Pastoral activities may have affected the abundance
of Buxus. However, grazing should not have directly provoked a
decline of Buxus, since its leaves are toxic and unpalatable to herbivores (Borchard et al., 2011). Cutting down of Buxus by humans to
favour livestock grazing, a common practice in present-day calcareous grasslands (Barbaro et al., 2004), may have caused its
reduction. However, Buxus is often used to delimit grazing plots,
especially in the Pyrenees, and once pastures are abandoned, it
rapidly expands and recolonizes the surrounding areas, to the point
of being considered invasive (Debussche and Lepart, 1992). All in
all, the abundance of Buxus in highly populated areas of the Temperate bioclimatic zone, especially in S France, compared to the
sparse distribution of Buxus in similarly populated areas in the
Mediterranean zone suggests that human activities may have not
been a major driving factor for increases and reductions of Buxus
populations.
Other possible causes for the Mediterranean decline of Buxus,
such as plant parasites and pathogens, stochastic events (i.e. fires),
and competition cannot be ruled out. However, parasites and
pathogens (e.g. Diptera: M. buxi Laboulbène, Fungi: Cylindrocladium
buxicola Henricot, Puccinia buxi Sowerby) occur widely in both the
Temperate and Mediterranean zones, so they do not explain the
differences in the modern distribution of Buxus. Concerning fires,
Buxus is sometimes reported as a passive pyrophyte (Le Houérou,
1980), so it is probable that fires could only have had minor effects
in reducing the Mediterranean populations. About competition,
modern studies show that Buxus populations have positive feedbacks in the establishment of woodland species (e.g. Quercus, Fagus,
and Abies), whose growth has negative feedbacks on Buxus abundance as they become mature trees (Debussche and Lepart, 1992;
García-Plazaola et al., 2008). Once more, this observation does not
support a different behaviour of the Temperate and Mediterranean
Buxus populations and therefore does not explain their decline in
the Mediterranean zone.
The fossil data indicate that in the southernmost regions
of Europe, Buxus can be considered a vulnerable tree. Sporadic occurrences of modern limited stands of Buxus in southern
Mediterranean regions are a natural residue of a wider distribution
that is going towards an eventual disappearance. Considering the
very low dispersal ability and growth rate of Buxus, there is very
little chance that small residual populations in the southernmost
regions of Europe will naturally increase to the point of becoming source areas for Buxus in a future glacial period, if they are not
included in red data books and adequately protected.
Conclusions
The combined use of modern and past distribution data proves
to be a powerful tool to reconstruct the location of the glacial refugia of Buxus in Europe, explain the configuration of its modern
populations, their density and degree of fragmentation, regional
behaviour, responses to climate change and anthropogenic impact,
as well as to determine the vulnerability of Buxus in the context of
the undergoing natural and anthropogenic environmental changes.
We have answered our research questions as follows:
• Is the modern distribution of Buxus the result of a post-glacial
migration from limited glacial refugia? In addition to the sites
located in the Mediterranean Peninsulas, fossil records from the
Jura Mts during the pleni- and late-glacial periods document the
location of Buxus refugia up to a latitude of at least 47◦ N. The
sparse early-Holocene occurrences in central Europe confirm that
Buxus may have persisted across most of its modern distribution,
although at very low densities, especially in the northern part
of its range. This hypothesis is strongly supported by the plant
traits of Buxus, which is a poor pollen and seed disperser, and is
able to withstand extremely low temperatures. In any case, the
modern distribution of Buxus in central Europe is not the effect of
a postglacial migration from southern refugia, where Buxus had
only a sparse distribution during the last glacial period.
• Are there regional differences in the Holocene dynamics of Buxus
populations in Europe? The fossil data indicate that during the
pleni- and late-glacial periods the populations located in the E
Pyrenees, W Alps, and Jura Mts (Temperate bioclimatic zone)
were relatively dense. Their seventy-fold expansion in the course
of the Holocene resulted in a continuous modern distribution
area. They increased exponentially from the late-glacial period
to the present time following a natural biological process. In contrast, the southern Peninsulas (Mediterranean bioclimatic zone)
hosted far less robust populations, which increased only moderately during the first half of the Holocene and resulted in a
modern highly fragmented distribution area. The populations
located in Minorca, Ibiza, Sardinia, Sicily, Apulia, Basilicata, Croatia, Thessaly, W Greece, and Peloponnese have been completely
extirpated, so that the number of Mediterranean Buxus sites in
the last thousand years only doubles the late-glacial period.
• Which are the different roles played by natural population
dynamics, climate change and human impact in shaping the
31
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
European distribution of Buxus in the last 15,000 years? In the
Temperate regions, the exponential population increase of the
postglacial period points to a typical biological process, not significantly affected by either climate or human activity. In the
Mediterranean regions, the natural population growth is limited
to the first half of the Holocene. Afterwards, an aridification process may be considered the most likely cause for the population
decline that fragmented the Buxus distribution, mainly reducing it to moist habitats. There is little support for alternative
causes (human activities, fires, phytopathogens, and competition), which would have equally affected the Buxus populations
of both the Temperate and the Mediterranean areas.
• Is Buxus a vulnerable genus in Europe? Most of the Buxus populations located in the Temperate zone are vigorous, so that they
do not require specific conservation actions. However, the populations in the northern part of the Temperate range (England,
NW France, Belgium, Luxembourg, and Germany) are very fragmented, which indicates a general weakness. These populations
may not be vigorous enough to survive future glacial reductions.
In the Mediterranean bioclimatic area, Buxus populations are
severely fragmented and the undergoing reduction is expected
to produce an even more dramatic fragmentation. It is noteworthy that the Buxus populations that were extirpated from the
southern regions in the course of the present interglacial had
persisted through the last glacial period. Considering the very
low dispersal ability and growth rate of Buxus, there is very little
chance that the southern territories of Europe will be naturally
recolonized and become refuge areas for Buxus in a future glacial
period. Thus, conservation actions of natural populations of Buxus
are of primary importance in the southernmost territories of the
Mediterranean Peninsulas.
Acknowledgements
The European Pollen Database and Natura 2000 contributors
are acknowledged. We wish to thank Carole Bégeot, Alessandra
Celant, Federico Di Rita, Laura Sadori, Maria Fernanda Sanchez
Goñi, Luisa Santos, and Pim van der Knaap for additional information on fossil data. Rolf Sievers, Klaus Ammann, Guillaume Decocq,
Stephen Meyer, Livio Poldini, Chrys Veryen, and Wolfgang Willner are acknowledged for comments on the modern distribution
of Buxus. We are grateful to Charles Turner and Josep Rosselló for
help in field work and useful discussions, and to Julia Mortera for
statistical support. This work was supported by grants attributed to
the authors by the University Roma Tre and Sapienza University of
Rome, and the Italian Ministry of University and Research (MIUR),
including PRIN funds.
Appendix A. Supplementary data
Supplementary data associated with this article can
be found, in the online version, at http://dx.doi.org/
10.1016/j.ppees.2012.07.001.
References
Allison, J., 1947. Buxus sempervirens in a late Roman burial in Berkshire. New Phytol.
46, 122.
Barbaro, L., Dutoit, T., Anthelme, F., Corcket, E., 2004. Respective influence of habitat conditions and management regimes on prealpine calcareous grasslands. J.
Environ. Manage. 72, 261–275.
Barboni, D., Harrison, S.P., Bartlein, P.J., Jalut, G., New, M., Prentice, I.C., Sanchez-Goñi,
M.F., Spessa, A., Davis, B.A.S., Stevenson, A.C., 2004. Relationships between plant
traits and climate in the Mediterranean region: an analysis based on pollen data.
J. Veg. Sci. 15, 635–646.
Bartley, D.D., Morgan, A.V., 1990. The palynological record of the King’s Pool,
Stafford, England. New Phytol. 116, 177–194.
361
Baylac, M., Daufresne, T., 1996. Wing venation variability in Monarthropalpus
buxi (Diptera: Cecidomyiidae) and the quaternary coevolution of box (Buxus
sempervirens L.) and its midge. A geometrical morphometric analysis. In: Marcus,
L.F., Corti, M., Loy, A., Naylor, G., Slice, D.E. (Eds.), Advances in Morphometrics.
Plenum Press, New York, pp. 285–301.
Bennett, K.D., Tzedakis, P.C., Willis, K.J., 1991. Quaternary refugia of north European
trees. J. Biogeogr. 18, 103–115.
Beug, H.J., 2004. Leitfaden der Pollenbestimmung. Verlag Dr. Friedrich Pfeil,
München.
Bhagwat, S., Willis, K.J., 2008. Species persistence in northerly glacial refugia of
Europe: a matter of chance or biogeographical traits? J. Biogeogr. 35, 464–482.
Blanca, G., 1999. In: Blanca, G. (Ed.), Libro Rojo de la flora silvestre amenazada de
Andalucía. Consejería de Medio Ambiente, Junta de Andalucía, Sevilla, Spain.
Borchard, F., Berger, H.-J., Bunzel-Drüke, M., Fartmann, T., 2011. Diversity of
plant–animal interactions: possibilities for a new plant defense indicator value?
Ecol. Indic. 11, 1311–1318.
Cañellas-Boltà, N., Rull, V., Vigo, J., Mercadé, A., 2009. Modern pollen–vegetation
relationships along an altitudinal transect in the central Pyrenees (southwestern
Europe). The Holocene 19, 1185–1200.
Carrión, J.S., Van Geel, B., 1999. Fine-resolution Upper Weichselian and Holocene
palynological record from Navarrés (Valencia Spain) and a discussion about factors of Mediterranean forest succession. Rev. Palaeobot. Palynol. 106, 209–236.
Cyprien, A.L., Visset, L., Carcaud, N., 2004. Evolution of vegetation landscapes during the Holocene in the central and downstream Loire basin (Western France).
Veget. Hist. Archaeobot. 13, 181–196.
de Beaulieu, J.-L., Reille, M., 1992. The last climatic cycle at La Grande Pile (Vosges,
France): a new pollen profile. Quaternary Sci. Rev. 11, 431–438.
Debussche, M., Lepart, J., 1992. Establishment of woody plants in Mediterranean old
fields: opportunity in space and time. Landscape Ecol. 6, 133–145.
Denèfle, M., Valadas, B., Vilks, A., Lintz, G., 1980. L’évolution Holocène de la végétation en Limousin. Bull. AFEQ 17, 189–199.
Di Domenico, F., Lucchese, F., Magri, D., 2011. Late glacial and Holocene history of
Buxus sempervirens L. in Italy. Ann. Bot. (Roma) 1, 45–58.
Di Rita, F., Simone, O., Caldara, M., Gehrels, W.R., Magri, D., 2011. Holocene environmental changes in the coastal Tavoliere Plain (Apulia, southern Italy): a
multiproxy approach. Palaeogeogr. Palaeoclimatol. Palaeoecol. 310, 139–151.
Essl, F., 2002. Verbreitung und Gesellschaftsanschluss des Buchsbaumes (Buxus sempervirens L.) im oberösterreichischen Enns- und Steyrtal. Verh. Zool.-Bot. Ges.
Österreich 139, 75–95.
Figueiral, I., Bettencourt, A., 2004. Middle/Late Bronze Age plant communities and
their exploitation in the Cavado Basin (NW Portugal) as shown by charcoal
analysis: the significance and co-occurrence of Quercus (deciduous) – Fabaceae.
Veget. Hist. Archaeobot. 13, 219–232.
Figueiral, I., Terral, J.-F., 2002. Late Quaternary refugia of Mediterranean taxa in the
Portuguese Estremadura: charcoal based palaeovegetation and climatic reconstruction. Quaternary Sci. Rev. 21, 549–558.
García-Plazaola, J.I., Esteban, R., Hormaetxe, K., Becerril, J.M., 2008. Seasonal
reversibility of acclimation to irradiance in leaves of common box (Buxus sempervirens L.) in a deciduous forest. Flora 203, 254–260.
Godwin, H., 1975. The History of the British Flora. A Factual Basis for Phytogeography. Cambridge University Press, Cambridge.
Hacker, J., Neuner, G., 2007. Ice propagation in plants visualized at the tissue level
by infrared differential thermal analysis (IDTA). Tree Physiol. 27, 1661–1670.
Hatakeyama, I., Kato, J., 1965. Studies on the water relation of Buxus leaves. Planta
65, 259–268.
Hormaetxe, K., Becerril, J.M., Hernández, A., Esteban, R., García-Plazaola, J.I., 2007.
Plasticity of photoprotective mechanisms of Buxus sempervirens L. leaves in
response to extreme temperatures. Plant Biol. 9, 59–68.
Jalut, G., Dedoubat, J.J., Fontugne, M., Otto, T., 2009. Holocene circum-Mediterranean
vegetation changes: climate forcing and human impact. Quaternary Int. 200,
4–18.
Joly, C., Visset, L., 2009. Evolution of vegetation landscapes since the Late Mesolithic
on the French West Atlantic coast. Rev. Palaeobot. Palynol. 154, 124–179.
Karataglis, S., Babalonas, D., Kabasakalis, B., 1982. The ecology of plant populations
growing on serpentine soils. II. Ca/Mg ratio and the Cr, Fe, Ni, Co concentrations
as development factors of Buxus sempervirens L. Phyton 22, 317–328.
Köhler, E., 2007. In: Kobinski, K. (Ed.), Buxaceae. The Families and Genera of Flowering Plants. Springer, Berlin, Heidelberg, New York.
Kvaček, Z., Bůžek, Č., Holý, F., 1982. Review of Buxus fossils and a new largeleaved species from the Miocene of Central Europe. Rev. Palaeobot. Palynol. 37,
361–394.
Lang, G., 1992. Some aspects of European late- and post-glacial flora history. Acta
Bot. Fenn. 144, 1–17.
Lázaro, A., Traveset, A., 2005. Spatio-temporal variation in the pollination
mode of Buxus balearica (Buxaceae), an ambophilous and selfing species:
mainland–island comparison. Ecography 28, 640–652.
Le Houérou, N., 1980. L’impact de l’homme et de ses animaux sur la forêt méditerranéenne. Forêt Méditerranéenne 2, 31–44.
Lenoble, F., Broyer, C., 1945. Sur la distribution de Buxus sempervirens L. en France.
Bull. Soc. Bot. France 92, 118–131.
Leroy, S.A.G., Roiron, P., 1996. Latest Pliocene pollen and leaf floras from Bernasso
palaeolake (Escandorgue Massif, Hérault, France). Rev. Palaeobot. Palynol. 94,
295–328.
Magny, M., Vannière, B., Calo, C., Millet, L., Leroux, A., Peyron, O., Zanchetta, G., La
Mantia, T., Tinner, W., 2011. Holocene hydrological changes in south-western
32
362
F. Di Domenico et al. / Perspectives in Plant Ecology, Evolution and Systematics 14 (2012) 354–362
Mediterranean as recorded by lake-level fluctuations at Lago Preola, a coastal
lake in southern Sicily, Italy. Quaternary Sci. Rev. 30, 2459–2475.
Magri, D., 2008. Patterns of post-glacial spread and the extent of glacial refugia of
European beech (Fagus sylvatica). J. Biogeogr. 35, 450–463.
Magri, D., Vendramin, G.G., Comps, B., Dupanloup, I., Geburek, T., Gömöry, D.,
Latałowa, M., Litt, T., Paule, L., Roure, J.M., Tantau, I., van der Knaap, W.O., Petit,
R., de Beaulieu, J.-L., 2006. A new scenario for the Quaternary history of European beech populations: palaeobotanical evidence and genetic consequences.
New Phytol. 171, 199–221.
Montemartini, L., 1907. Contributo alla biologia fogliare del Buxus sempervirens L.
Atti dell’Istituto Botanico dell’Università di Pavia 10, 239–244.
Oldfield, F., Statham, D.C., 1963. Pollen analytical data from Urswick Tarn and Ellerside Moss, N. Lancashire. New Phytol. 62, 53–66.
Ouguerram, A., Visset, L., 2001. Histoire de la végétation et première mise en évidence d’un milieu marin pendant l’HoIocène dans la vallée de l’Erdre et le Val
de Gesvres (Bassin versant de la Loire, Massif Armoricain, France). Quaternaire
12, 189–199.
Pérez-Obiol, R., Julià, R., 1994. Climatic change on the Iberian Peninsula recorded in
a 30,000-year pollen record from Lake Banyoles. Quaternary Res. 41, 91–98.
Phillips, L., 1974. Vegetational history of the Ipswichian/Eemian interglacial in
Britain and continental Europe. New Phytol. 73, 589–604.
Pigott, C.D., Pigott, S., 1993. Water as a determinant of the distribution of trees at
the boundary of the Mediterranean zone. J. Ecol. 81, 557–566.
Rivas-Martínez, S., Penas, A., Díaz, T.E., 2004. Bioclimatic Map of Europe. Cartographic Service, University of León, Spain.
Tutin, T.G., Heywood, V.H., Burges, N.A., Moore, D.M., Valentine, D.H., Walters, S.M.,
Webb, D.A. (Eds.), 1968. Flora Europaea, vol. 2. Rosaceae to Umbelliferae. Cambridge University Press, Cambridge.
van der Knaap, W.O., van Leeuwen, J.F.N., 2001. Vegetationsgeschichte und menschlicher Einfluss in der Umgebung des Bibersees zwischen 2600 und 50 v. Chr.
Cham-Oberwil Hof (Kanton Zug). In: Gnepf Horisberger, U., Hämmerle, S. (Eds.),
Befunde und Funde aus der Glockenbecherkultur und der Bronzezeit. Antiqua
33. Schweizerische Gesellschaft für Ur- und Frühgeschichte, Basel, pp. 181 and
194–199.
Vanden Berghen, C., 1955. Étude sur les irradiations de plantes méridionales dans
la vallée de la Meuse wallone. Bull. Soc. Roy. Bot. Belg. 87, 29–55.
Visset, L., Bernard, J., 2006. Évolution du littoral et du paysage, de la presqu’île de
Rhuys à la rivière d’Étel (Massif Armoricain – France), du Néolithique au Moyen
Âge. Archéosciences 30, 143–156.
Waller, M.P., Hamilton, S., 2000. Vegetation history of the English chalklands: a midHolocene pollen sequence from the Caburn, East Sussex. J. Quaternary Sci. 15,
253–272.
Walter, H., 1929. Die osmotischen Werte und die Kältenschäden unserer wintergrünen Pflanzen während der Winterperiode. Ber. Dtsch. Bot. Ges. 47, 338–348.
Wegmüller, S., 1984. Zur Ausbreitungsgeschichte von Buxus sempervirens L. im
Spät- und Postglazial in Süd- und Mitteleuropa. Dissertationes Botanicae 72,
333–344.
Wigginton, M.J., 1999. British Red Data Books: 1. Vascular plants. Nature Conservancy Council, Peterborough, UK.
Willis, K.J., Birks, H.J.B., 2006. What is natural? The need for a long-term perspective
in biodiversity conservation. Science 314, 1261–1265.
Willis, K.J., Van Andel, T.H., 2004. Trees or no trees? The environments of central and
eastern Europe during the last glaciation. Quaternary Sci. Rev. 23, 2369–2387.
Yll, E.-I., Pérez-Obiol, R., Pantaleón-Cano, J., Roure, J.M., 1997. Palynological evidence for climatic change and human activity during the Holocene on Minorca
(Balearic Islands). Quaternary Res. 48, 339–347.
Zanchetta, G., Borghini, G., Fallick, A.E., Bonadonna, F.P., Leone, G., 2007.
Late Quaternary palaeohydrology of Lake Pergusa (Sicily, southern Italy)
as inferred by stable isotopes of lacustrine carbonates. J. Paleolimnol. 38,
227–239.
33
Genetic structure and taxonomical boundaries in the western Palaearctic Buxus species
Di Domenico1*, F., Rosato2, M., Lucchese1, F., Rosselló2, J.A.
1
Università degli Studi Roma Tre, Viale G. Marconi 446, 00146, Rome, Italy
2
Jardí Botánic, Universidad de Valencia, c/ Quart 80, E-46008, Valencia, Spain
*
Corresponding Author: [email protected]
Abstract
The genus Buxus L. is present in the Western Palaearctic area with five closely related taxa showing
little interspecific and wide infraspecific morpho-anatomical differentiation.. The lack of clearly
distinctive features and the strong, manifold variations in the few characters used to differentiate
these species is reflected by an uncertain taxonomical framework. ISSR fingerprinting was used to
study species boundaries, to relate the degree of genetic differentiation with the range of
morphological variability and to test the role of the geographic isolation of Buxus populations in
shaping the genetic diversity of this genus. A total of 109 bright and reproducible bands (83%
polymorphic) were obtained after screening 264 individuals from five different species with seven
primers. Ordination and numerical regionalisation analyses revealed the existence of two main
groups, one including B. sempervirens, B. colchica and B. hyrcana, and one including B. balearica
and B. longifolia. In the first group it was not possible to distinguish samples according to
taxonomy or geographic distribution, whereas ISSR revealed the existence genetic boundaries
between species and populations in the second. The samples of B. balearica could be clearly
distinguished according to their occurrence NW Africa, S Spain and W Mediterranean islands.
Matrix correlation analyses evidenced a significant and differential role of geographic isolation on
the genetic diversity of the two groups of species. Available data suggest that clear taxonomical
barriers only exist between B. sempervirens and B. balearica. Neveretheless, B. balearica and B.
longifolia proved to be genetically different. The magnitude of genetic variation and the effects of
geographic isolation on population differentiation indicate that B. sempervirens, B. colchica and B.
hyrcana are not distinct taxonomic entities. With some cautions, also the genetic differentiation
between B. balearica and B. longifolia should not be ascribed to taxonomical boundaries, but rather
to a high degree of population differentiation.
Keywords: Buxus, taxonomy, ISSR, ordination, numerical regionalisation, isolation by distance
34
Introduction
The genus Buxus L. is one of the five genera of the Buxaceae. This family comprises around 100
species distributed in the Northern Hemisphere of the Old and New World, extending to Andean
South America and the Caribbean coast, to South Africa and Madagascar, and to peninsular
Malaysia (Köhler, 2007). Buxus is the most diverse and widespread genus of the family with more
than 90 species located in three centres of diversity. A group of 7-8 relict species occur in Africa
and forms the sect. Probuxus (Köhler & Brückner, 1989). A first centre of diversification is found
in the Caribbean-Latin America with more than 40 species from sect. Tricera (Mathou 1940). A
second diversity hotspot is located in Eurasia and NW Africa, where approximately 40 species form
the sections Buxus and Eugeniobuxus. While sect. Eugeniobuxus is confined to SE tropical Asia
with 6 species (Hatushima, 1942), the sect. Buxus counts more than 30 species distributed across the
Eurasian continent from Japan to Spain and NW Africa. The systematic of the genus is based on
pollen morphology, leaf size and venation, anatomy of the petiole and shoot (see Köhler &
Brückner, 1989 and references therein), morphology of reproductive structures (von Balthazar &
Endress, 2002), and molecular markers (von Balthazar & al., 2000; Van Laere & al., 2011). These
studies concurred to a thorough knowledge of the structure of extant taxa and their systematic
relationships. Also, they allowed tracing the phylogenetic evolution of this ancient genus (Köhler &
Brückner, 1989) with the support of a large availability in space and time of fossil material (see
Kvaček & al., 1982 and Bessedik 1983 and references therein).
Systematic and phylogenetic studies at a continental or global scale provide a clear framework for
the delimitation of Buxus sections. However, the identification of different species within sections is
not always straightforward. On the one hand, leaf venation pattern, anatomy of the petiole and shoot
and pollen morphology are of little utility in discriminating between species because they may show
no variation amongst different members of a section (Köhler & Brückner, 1989). On the other hand,
the distinctive traits such as leaf shape and size and morphology of reproductive structures are
highly variable, at multiple levels ranging from the individual specimen to different populations
(Hatushima, 1942). In groups of closely related Buxus species this plasticity lead to an overlapping
range of variation in diagnostic characters (Davis 1982).
One case exemplifying this problem are Western Palaearctic Buxus, a group of five closely related
species distributed in Central Europe and the Mediterranean area (3 species), and in the Colchic and
Hyrcanian regions (one species each). These species are differentiated on the basis of leaf shape and
size, hairiness of shoots, wood anatomy, morphology of reproductive structures, and molecular
markers (Boissier, 1853; Pojarkova, 1949; Webb, 1968; Davis, 1982; Guseinova 1996; Sonboli &
al., 2004; Van Laere & al., 2011). The variable nature of these morphological characters is probably
the fundament of the diverse taxonomic treatment of Buxus in the literature (Tab. 1). Two vastly
recognised taxonomic entities are the sister (von Balthazar & al., 2000) species B. sempervirens L.
and B. balearica Lam., the latter showing bigger leaves, hairless young shoots, higher style to
capsule ratio and revolute horns (styles) in the mature fruit. The botanical explorations of Turkey
during the middle 19th century lead to the discovery of B. longifolia Boiss. (Boissier, 1853), a
species close to B. balearica but differing for its narrower leaves, pubescent shoots and linearsagittate anthers. On the basis of the progressive increase in leaf and fruit size and variations in the
ontogeny of rudimentary ovary (pistillode), Pojarkova (1947) classified the geographically isolated
populations of B. sempervirens from the Colchis and Hyrcanian regions as B. colchica Pojark. and
B. hyrcana Pojark., respectively. During the last fifty years a number of papers supported one or the
other view, but to date an agreement still has to be found (Guseinova, 1996; Sonboli & al., 2004;
Van Laere & al., 2011).
35
Tab. 1 Alternative taxonomic treatments used in the literature for the Buxus species considered in the present study 1)
Sales & Hedge, 1996; 2) Davis & Hedge, 1971; 3) Davis 1982, 4) Köhler & Brückner, 1989; 5) Boissier, 1853; 6)
Webb, 1968; 7) Rechinger 1966; 8) Pojarkova, 1947; 9) Grossgeim, 1962; 10) Van Laere & al., 2011. * cases in which
the author recognise the taxonomic entity, but suggest synonymy either with B. sempervirens or B. balearica.
DNA fingerprinting techniques like AFLP, RAPD and ISSR, established as valuable means to
assess taxonomic diversity for their high polymorphism and cost efficiency. These markers allow
the detection of species boundaries at several taxonomic levels. In facts, they allow detection of
sibling species within difficult plant complexes (Jiménez & al., 2005; Hardion & al. 2012),
providing an identification tool largely complementary to morphology. AFLPs markers already
proved their efficiency in discriminating between species and cultivars of Buxus grown in nurseries
or gardens (Salvesen & al., 2009; Van Laere & al. 2011). It might be of interest to check their
congruency with other kinds of molecular markers and to extend the investigations to plants
sampled from natural populations. Anchored ISSRs markers, because of the longer primers and
higher annealing temperatures, show a good reproducibility and errors can be minimised with good
laboratory practices. These markers offer a valuable opportunity to further investigate the
taxonomic relationships among Western Palaearctic Buxus.
Material and Methods
Distribution and Ecology of the studied species
B. sempervirens L. (Fig.1) is widely present in Europe, NW Africa and along the SW coasts of the
Black Sea (Quézel & Santa, 1962; Webb, 1968; Davis, 1982). B. colchica Pojark. is distributed
around the SE and E coasts of the Black Sea and deep into the Colchis area (Pojarkova, 1947, 1949;
Grossgeim, 1962; Davis, 1982). B. hyrcana Pojark. is found in the interior areas surrounding the
southern coasts of the Caspian Sea (Pojarkova, 1947, 1949; Grossgeim, 1962; Rechinger, 1966). B.
balearica Lam. show a distribution restricted to S Spain, W Mediterranean Islands (Baleares,
Sardinia), and NW Africa (Gennari, 1864; Quézel & Santa, 1962; Benedí, 1997). B. longifolia
Boiss. occur in S Anatolia and is known only from three locations (Boissier, 1853; Davis, 1982).
These group of more of less geographically isolated species live in a wide range of habitats (broadleaved deciduous and evergreen forests, needle-leaved evergreen forests, supra-mediterranean
garigues) and on different substrates (limestone, ultramafic rocks, tuff) of the Mediterranean and
Temperate areas. As the whole genus, they reproduce sexually through insect and wind pollination,
but are also capable of vegetative reproduction through stolons or buried branches. Seed dispersal
consists of an ejection mechanism that account for a localized dissemination (Debusche & Lepart,
1992), and abiotic agents such as rain and flowing water may account for dispersal over longer
distances (Köhler, 2007). Myrmechocory is observed in B. sempervirens (Köhler, 2007) and B.
balearica (Lázaro & Traveset, 2005), but few ant species are effective dispersers while most are
consumers. The plants are long lived and slow growing, and sought for their wood (Record, 1925)
and ornamental purposes, as the vast number of existing cultivars testify (Batdorf, 2005).
36
Fig. 1 Distribution map of the five Buxus species considered in the present study.
Plant material
A total of 264 individuals from 44 populations of different Buxus species were used for ISSR
analysis (Fig. 2, Appendix 1). Our study was primarily focused on geographic patterns and
taxonomical boundaries of Buxus. For this reason, we analyzed a small number of plants (six) per
site. At each site leaves were collected from plants separated by at least 50-100m to avoid the
sampling of clones. The material was placed in plastic bags with silica gel and stored at room
temperature until required.
DNA extraction, amplification and electrophoresis
Total genomic DNA was isolated and purified using the DNeasyTM Plant Minikit (Qiagen, Hilden,
Germany) following the manufacturer’s instructions. Seven ISSR primers were used (Tab. 3),
manufactured by Eurofins MWG Synthesis GmbH. Each primer was used to amplify 0.5 µL of each
DNA extract using a Taq DNA polymerase (BioTools, B&M Labs, S.A.). All reactions were carried
out in a final volume of 25 µL, containing 2.5 µL of 10× PCR buffer MgCl free (N.E.E.D.), 1.5 µL
of MgCl 50 mM (N.E.E.D.), 4 µL of dNTPs 1.25 mM (Fermentas), 0.5 µL of primer 35 µM, and 1
µL of Taq polymerase 1U/ µL. The conditions of the PCRs included an initial denaturising step of
1.5 min at 94 °C, 35 cycles of 40 s at 94 °C, 60 s of annealing (temperatures for each primer are
given in Tab. 2), followed by a single extension step of 5 min at 72 °C. All PCR analyses were
performed in a Primus thermal cycler (MWG-BIOTECH) and a negative control reaction without
DNA. The entire product (25 µL) of each PCR was examined by electrophoresis in 2% agarose gels
stained with ethidium bromide running at 110 mV and 400 mA for 90 minutes. Digital images of
37
the gels were processed with a graphic software to score for the presence or absence of bands. Each
ISSR band was coded as present or absent (1 or 0, respectively) and a binary data matrix was
compiled. Following suggestions by Grosberg & al. (1996), no attempts were made to code for
band intensity. DNA bands showing quantitative variation in brightness were scored as present,
regardless of their intensities, and absent if they were undetectable.
Data analysis
Phenetic similarity was analyzed using ordination techniques (Principal coordinates analysis, PCoA
and Multidimensional Scaling, MDS) implemented in PAST 2.15 software (Hammer & al., 2001).
The coefficient of Jaccard (1908) was adopted to score band matching, as this method does not take
into account shared band absences. Shared band absences should not be used for statistical analyses
of dominant markers data because they not always indicate a homozygous recessive genotype and
may be caused by a number of different factors. For example, loss of a primer annealing region,
indels in the fragment between the two primer sites, and experimental errors may result in erroneous
interpretations of genetic frequencies. The Jaccard’s similarity matrix was then used as input in the
Principal Coordinates Analysis (PCoA) and Metric Multidimensional Scaling (MDS). Isolines
connecting population-averaged scores for PCoA axis were created with Surfer 11 (Golden
Software Inc.) to geographically represent the observed geographical variation. Numerical
regionalisation analyses using Monmonier’s algorithm (Monmonier 1973) were used to detect
genetic barriers between species using the software Barrier version 2.2 (Manni & al., 2004)
To evaluate the effects of isolation by distance (IBD), pair-wise geographical and genetic distance
(using the 1-Jaccard coefficient) matrices were computed and tested for correlation using the
Mantel version 2.0 package (http://www.sci.qut.edu.au/nrs/mantel.htm).
Results
A total of 109 bright and reproducible bands ranging in size from 200 to 1600bp were obtained after
screening 264 individuals with seven primers (Tab. 2). The vast majority (84%) of ISSR bands were
polymorphic. Nearly all individuals analysed (87%) could be distinguished by a unique multilocus
ISSR profile. Thirty-four specimens shared their profile with one or more individuals from the same
population.
Tab. 2 Primer used, primer sequences, annealing temperature, fragment sizes, number of loci, and
percentage of polymorphic fragments generated by the seven primers used in the study.
38
Population and species-specific ISSR markers were found in some cases (Tab. 3). Sixteen, ten and
four bands were species-specific for B. sempervirens, B. balearica and B. longifolia, respectively.
Of these, nine, ten and one were fixed markers (present in all samples), respectively. No private
markers were detected in B. colchica and B. hyrcana, whereas population specific markers were
detected in the continental (3) and insular (4) populations of B. balearica.
Tab. 3 Species and population ISSR markers found in the W Palaearctic Buxus species considered in
the present study
Principal Coordinates Analysis and Multidimensional Scaling produced overlapping results, thus
only results from PCoA will be showed. Thirteen PCoA axes had eigenvalues greater than 1 and
explained 79.81% of the variance, suggesting that ISSR data were not strongly correlated. The first
two axes explained the 51.31% and 4.87% of the total variation in the dataset, respectively. PCoA
detected two distinct groups separated along the first axis (Fig. 2, top-left). One group include
samples of B. sempervirens, B. colchica and B. hyrcana, and the other B. balearica and B.
longifolia. Within the latter group, two distinct clusters could be distinguished along the second
axis. These clusters corresponded to samples of B. balearica and B. longifolia. Analyses on partial
datasets, one including accessions of B. sempervirens, B. colchica and B. hyrcana, and the other
samples of B. balearica and B. longifolia, revealed additional patterns of genetic variation but also
the absence of strong taxonomical or geographical differentiation. In facts, the samples of B.
sempervirens, B. colchica and B. hyrcana could not be separated according to taxonomic
membership or regional distributions (Fig. 2, central-left). On the contrary B. balearica and B.
longifolia were further differentiated along the first axis accounting for the 40.07% of the total
variation in the dataset (Fig. 2, bottom-left). Moreover, PCoA allowed distinguishing samples of B.
balearica according to their occurrence NW Africa, S Spain and W Mediterranean islands (Fig. 2,
bottom-left). The existence of genetic discontinuities was detected also by numerical regionalisation
analysis. This procedure identified a barrier between the group including B. sempervirens, B.
colchica and B. hyrcana, and the group including B. balearica and B. longifolia (Fig. 2, top-right).
The analysis of partial datasets revealed the absence of a genetic barrier between B. sempervirens,
B. colchica and B. hyrcana (Fig. 2, central-right), whereas B. balearica and B. longifolia were
separated by a genetic barrier (Fig. 2, bottom-right). Correlation analysis of the matrices of
geographic and phenetic distances (1-Jaccard index) using the Mantel test revealed a moderate and
significant correlation between geographic and genetic distances for B. sempervirens, B. colchica,
and B. hyrcana (r2 = 0.22, P < 0.005). Moreover, the Mantel test showed high and significant
correlation for B. balearica and B. longifolia (r2 = 0.81, P < 0.005).
39
Fig. 2 Left: differentiation of B. sempervirens (), B. colchica (
), B. hyrcana (), B. balearica () and B. longifolia
() along the first two PCoA axes for the entire dataset (top), only for samples of B. sempervirens, B. colchica and B.
hyrcana (centre), and only for samples of B. balearica and B. longifolia (bottom). Groups 1, 2 and 3 in the bottom
picture represent samples of B. balearica from NW Africa, W Mediterranean islands and S Spain, respectively. Right:
map showing the distribution of studied samples and isolines connecting population-averaged scores for the first PCoA
axis. The grey shaded area represent the genetic barriers detected with numerical regionalisation analyses using
Monmonier’s maximum difference algorithm.
Discussion
To the present day, few studies focused on the taxonomic framework of the Western Palaearctic
Buxus as a whole. Van Leare & al. (2011) studied the genetic relationships in European and Asiatic
Buxus species based on AFLP markers, genome sizes and chromosome numbers. However, this
study was based on Buxus material from nurseries and private collections. Also, a significant
number of plants were used only from B. sempervirens and B. microphylla, whereas other species
were underrepresented. The current taxonomy Western Palaearctic Buxus has been primarily based
40
on an intuitive perception of the observed patterns of morphological variation. Aside a few studies
on flower ontogeny (Guseinova, 1996), wood anatomy (Köhler & Brückner, 1989; Sonboli & al.,
2004) and pollen morphology (Köhler & Brückner, 1989), the treatment of phenotypic variation in
Buxus was analysed by non-explicit approaches (Pojarkova, 1947; Davis 1982). In the present
study, ISSRs were able to detect molecular variation among closely related Buxus species. Also, it
was possible to differentiate geographic populations in some cases. This offers an opportunity to
discuss the taxonomy and species differentiation of the Western Palaearctic Buxus.
Previous studies indicate that B. sempervirens and B. colchica are two distinct taxa forming an
independent cluster among Eurasiatic Buxus species (Van Laere & al. 2011). All the remaining
species were clustered together, with B. balearica as outgroup. B. hyrcana resulted genetically very
close to the SW Asiatic B. harlandii Hance. By contrast, our ISSRs dataset show that
B. sempervirens, B. colchica and B. hyrcana are closely related species with low levels of genetic
differentiation. In facts, they could not be distinguished according to a separate taxonomic
treatment. Also, ISSR revealed the absence of abrupt genetic changes throughout the distribution
areas of these taxa. Contrasting results from AFLPs and ISSRs might be due to several factors.
Jones & al. (1997) evidenced that among DNA fingerprinting markers, AFLPs were the least
reliable and reproducible. Conflicting genetic relationships may also arise from the uneven selection
of material for analyses, as well as from the absence of plants from natural populations. Our results
agree with morpho-anatomical studies that assigned variations between species to ecological
plasticity and local adaptation (Guseinova, 1996), and support the view that only B. sempervirens
should be considered a valid taxonomic entity (Sales & Hedge 1996).
To date, no explicit morphological study addressed the problem of Mediterranean Buxus taxonomic
distinctiveness. B. longifolia, occurring in Southern Anatolia, was ascribed to B. balearica by
various authors because the observed range of morphological variation does not justify a taxonomic
separation (Davis, 1982; Webb, 1968). The genetic relationships between B. balearica and B.
longifolia were investigated by Rosselló & al. (2007). Their study of ITS sequence variation
revealed that some distinct, but not divergent, ribotypes occur in each species. The present research
revealed that B. balearica and B. longifolia are two closely related species but with different genetic
identity. The distinctiveness detected by ordination and numerical regionalisation analyses requires
careful consideration before drawing any taxonomical conclusion. The effects of geographic
isolation on genetic differentiation may help interpreting the results from ISSR analysis.
Considering that ISSR were able to differentiate the neighbouring populations of B. balearica into
geographic groups, as well as the magnitude of isolation-by-distance effect, the detected genetic
variation might not be related to taxonomic boundaries. On the contrary, such diversity could have
originated by independent evolution from a common gene pool after repeated fragmentation events
leading to the breakdown of a continuous distribution area, as suggested by others (Rosselló & al.
2007). This hypothesis is partly substantiated by the behaviour of Mediterranean Buxus populations
during the present interglacial period. Di Domenico & al. (2012) detected several episodes of local
extinctions around the Mediterranean basin responsible for the disjunct and fragmented distribution
of Buxus in this area. Nevertheless, the absence of Tertiary and Quaternary Buxus fossil finds along
Northern Africa poses some questions on the location of those populations that eventually formed a
continuous distribution area in the past (see Jacobs & al. 2010 and references therein).
41
Literature Cited
Batdorf, L.R. 2005. International Registration List of Cultivated Buxus L. Boxwood Bullettin 45:
98-109.
Benedí, C. 1997. Buxaceae. Pp. 187-189 in: Castroviejo, S. & al. (ed.) Flora Ibérica. Madrid: CSIC,
Real Jardín Botánico.
Bessedik, M. 1983. Le genre Buxus L. (Nagyipollis KEDVES 1962) au Tertiaire en Europe
occidentale: évolution et implications paléogaographiques. Pollen Spores 25: 461-486.
Boissier, E. 1853. Buxaceae. Ser. 1, Vol. 12, p. 107 of: Diagnoses Plantarum Orientalium
novarum.
Browicz, K. 1987. Chorology of the Euxinian and Hyrcanian elements in the woody flora of Asia.
Plant Syst. Evol. 162: 305-314.
Davis, P.H. & Hedge, I.C. 1971. Floristic links between N.W. Africa and S.W. Asia. Ann.
Naturhistor. Mus. Wien 75: 43-57.
Davis, P.H. 1982. Buxaceae. Pp. 630-632 in: Davis, P.H. (ed.), Flora of Turkey and the East
Aegean Islands. Edinburgh, Edinburgh University Press.
Debussche, M. & Lepart, J., 1992. Establishment of woody plants in Mediterranean old fields:
opportunity in space and time. Landscape Ecol. 6: 133-145.
Di Domenico, F., Lucchese, F. & Magri, D. 2012. Buxus in Europe: Late Quaternary dynamics and
modern vulnerability. Perspect. Plant. Ecol. 14: 354-362.
Gennari, P., 1864. Flora medica sarda ossia descrizione della piante medicinali che crescono
spontanee in Sardegna. Sardegna Medica, Firenze.
Grosberg, R.K., Levitan, D.R. & Cameron, B. 1996. Characterization of genetic structure and
genealogies using RAPD-PCR markers: a random primer for the novice and the nervous. Pp. 67100 in: Ferraris J., Palumbi, S.R. (ed.) Molecular Zoology. Advances, strategies and protocols.
Wiley-Liss, New York.
Grossheim, A.A., 1962. Flora Kavkaza. Baku, Moskva, Leningrad.
Guseinova, S.O. 1996. O semennom razmnozhenii i taksonomicheskom range Buxus hyrcana
(Buxaceae). Bot. Zhurn. (Moscow & Leningrad) 81: 49-59.
Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological statistics software package
for education and data analysis. Palaeontol. Electron. 4: 1-9.
Hardion, L., Verlaque, R., Baumel, A., Marianick, J. & Vila, B. 2011. Revised systematics of
Mediterranean Arundo (Poaceae) based on AFLP fingerprinting and morphology. Taxon 61: 12171226.
Hatushima, S., 1942. A revision of the Asiatic Buxus. J. Dept. Agric. Kyushyu Imp. Univ. 6: 261342.
42
Jaccard, P., 1908. Nouvelles recherches sur la distribution florale. Bull. Soc. Vaud. Sci. Nat. 44:
223-270.
Jacobs, B.F., Pan, A.D. & Scotese, C.R. 2010. A Review of the Cenozoic Vegetation History of
Africa. in: Werdelin, L. & Sanders W.J. (ed.), Cenozoic Mammals. University of California Press.
Jiménez, J.F., Sánchez-Gómez, P., Güemes, J. & Rosselló J.A. 2005. Isolated populations or
isolated taxa? A case study in narrowly-distributed snapdragons (Antirrhinum sect. Sempervirentia).
Pl. Syst. Evol. 252: 139-152.
Jones, C., Edwards, K., Castaglione, S. & al. 1997. Reproducibility testing of RAPD, AFLP and
SSR markers in plants by a network of European laboratories. Mol. Breed. 3: 381-390.
Köhler, E. & Brückner, P. 1989. The genus Buxus (Buxaceae): aspects of its differentiation in space
and time. Plant Syst. Evol. 162: 267-383.
Köhler, E. 2007. Buxaceae. Pp. 40-47 in: Kubitzki, K. (ed.), The Families and Genera of Flowering
Plants. Springer, Berlin, Heidelberg, New York.
Kvaček, Z., Bůžek, Č. & Holý, F. 1982. Review of Buxus fossils and a new largeleaved species
from the Miocene of Central Europe. Rev. Palaeobot. Palynol. 37: 361-394.
Lázaro, A. & Traveset, A. 2005. Spatio-temporal variation in the pollination mode of Buxus
balearica (Buxaceae), an ambophilous and selfing species: mainland-island comparison. Ecography
28: 640-652.
Manni, F., Guérard, E. & Heyer, E. 2004. Geographic patterns of (genetic, morphologic, linguistic)
variation: how barriers can be detected by “Monmonier’s algorithm”. Hum. Biol. 76: 173-190.
Mathou, T., 1940. Recherches sur la famille des Buxacées. Dissertation, Faculté des sciences de
Toulouse.
Monmonier, M.S., 1973. Maximum-Difference Barriers: An Alternative Numerical Regionalization
Method. Geogr. Anal. 5: 254-261.
Pojarkova, A.I. 1947. Referat. Nauch.-Issl. Rab. Akad. Nauk SSSR, Biol. 1945, 7.
Pojarkova, A.I., 1949. Buxus. Pp. 385-389 in: Komarov, V.L. (ed.), Flora URSS. URSS Acad. Sci.
Press.
Quézel, P. & Santa, S. 1962. Nouvelle Flore de l’Algérie et des régions désertiques méridionales.
Paris, CNRS.
Rechinger, K.H. 1966. Flora Iranica. Graz.
Record, S.J. 1922. Boxwoods of commerce. Bull. Torrey Bot. Club 48: 297-306.
Rosselló, J.A., Lázaro, A., Cosín, R., Molins, A., 2007. A phylogeographic split in Buxus balearica
(Buxaceae) as evidenced from nuclear ribosomal markers: when ITS paralogues are welcome. J.
Mol. Evo. 64: 143-157.
43
Sales, F. & Hedge, I.C., 1996. Biogeographical aspects of selected SW Asiatic woody taxa. Ann.
Naturhist. Mus. Wien 98: 149-161.
Salvesen, P.H., Kanz, B. & Moe, D. 2009. Historical Cultivars of Buxus sempervirens L. Revealed
in a Preserved 17th Century Garden by Biometry and Amplified Fragment Length Polymorphism
(AFLP). Europ. J. Hort. Sci. 74: 130-136.
Sonboli, A., Azizian, D. & Khosravi, F., 2004. Anatomical study of three Buxus L. species
(Buxaceae). J. Sc. (Al-Zahra University) 17: 43-50.
Van Laere, K., Hermans, D., Leus, L. & Van Huylenbroeck, J. 2011. Genetic relationships in
European and Asiatic Buxus species based on AFLP markers, genome sizes and chromosome
numbers. Plant Syst. Evol. 293: 1-11.
von Balthazar, M. & Endress, P.K. 2002. Reproductive structures and systematics of Buxaceae. Bot.
J. Linn. Soc. 140: 193-228.
von Balthazar, M., Endress, P.K. & Qiu, Y.-L. 2000. Phylogenetic relationships in Buxaceae based
on nuclear internal transcribed spacers and plastid ndhF sequences. Int. J. Plant Sci. 161: 785-792.
Webb, D.A. 1968. Buxaceae. Pp. 925-926 in: Tutin & al. (ed), Flora Europaea. Cambridge,
Cambridge University Press.
44
Appendix 1: Sampled populations of the W Palaearctic Buxus species.
Species
B. sempervirens L.
B. hyrcana Pojark.
B. colchica Pojark.
B. balearica Lam.
B. longifolia Boiss.
Country
Albania
France
France
France
France
Germany
Greece
Greece
Greece
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Luxembourg
Makedonia
Makedonia
Montenegro
Spain
Spain
Spain
Turkey
Turkey
UK
UK
UK
Iran
Iran
Georgia
Italy
Morocco
Morocco
Morocco
Spain
Spain
Spain
Spain
Spain
Turkey
Turkey
Turkey
Site
Çibërraka
Coursegoules
Guise
Les Plans
Tagolsheim
Karden
Katara Pass
Mount Olympus
Pixariá
Bargonasco
Fiastra
Fiume Bussento
Fiume Panaro
Parco di Veio
Pianico
Rapolano
Vicalvi
Ahn
Demir Kapja
Matka
Krasovina
Montgò
Iglesuela del Cid
Sierra de Cazorla
Alihocalar
Gölkug
Box Hill
Boxwell
Ellensborough
Ab Dang Sar
Nour
Nikortsminda
Monte Tassua
Gorges du Todra
Jebha
Talembote
Frigiliana
Rágol
Artá
Cap Ventos
Ternells
Adrasan
Antakya
Feke
Lat
41.17
43.78
49.90
43.76
47.67
50.19
39.79
40.09
38.70
44.28
43.22
40.14
44.34
42.10
45.94
43.28
41.68
49.63
41.15
41.98
42.58
38.81
40.50
37.90
40.81
40.67
51.25
51.63
51.75
36.33
36.58
42.48
39.34
31.59
35.05
35.26
36.76
36.98
39.78
39.16
39.90
36.33
36.25
37.99
Lon
20.25
7.07
3.60
3.28
7.27
7.30
21.20
22.40
23.63
9.47
13.42
15.56
10.96
12.40
10.27
11.59
13.71
6.43
22.29
21.33
19.15
0.13
-0.38
-2.94
29.82
31.63
-0.32
-2.27
-0.79
52.88
52.05
43.06
8.70
-5.58
-4.76
-5.27
-3.88
-2.69
3.35
2.96
2.96
30.51
36.06
35.84
Alt
280
875
140
486
305
124
1516
1455
807
72
215
210
480
230
330
280
477
155
173
393
204
576
1417
1015
546
170
126
176
160
280
-13
1075
150
1000
150
160
350
440
340
50
650
300
150
750
Code
LIB
COU
GUI
LSP
TAG
KAR
KAT
PRI
PIX
BAR
FIA
CAS
PAN
VEI
PIA
RAP
VIC
AHN
DMK
MAT
KRA
MTG
IDC
SDC
ALI
GOL
BXH
BXW
ELL
ADS
NOU
NIK
MTA
TOD
JEB
TAL
FRI
RAG
ART
CPV
TER
ADR
ANT
FEK
45
Chloroplast variation in two closely related woody taxa showing contrasting histories and
distributions
Di Domenico1*, F., Molins2, A., , Rosato3, M., Lucchese1, F., Rosselló3, J.A.
1
Università degli Studi Roma Tre, Viale G. Marconi 446, 00146, Rome, Italy
2
Center for Ecological Research and Forestry Applications, Autonomous University
of Barcelona, Barcelona, Spain,
3
Jardí Botánic, Universidad de Valencia, c/ Quart 80, E-46008, Valencia, Spain
*
Corresponding Author: [email protected]
Abstract
This is a first contribution to the knowledge of chloroplast genetic variation in western Palaearctic
Buxus species. The aim of this research is to investigate the effects of a possible northern
persistence during the last glacial period, evaluate the effects of population dynamics assessed by
palaeobotanical studies and to analyze the phylogeographic signal in the disjunct populations
occurring in the western and eastern Mediterranean area. Seventeen haplotypes were detected after
sequencing 340 individuals from 65 populations of B. sempervirens and B. balearica throughout the
distribution range of the genus in the western Palaearctic area. Nearly all populations were fixed for
a single haplotype, and the organization of cpDNA diversity revealed a strong degree of population
differentiation, low variability within populations and lack of phylogeographic structure. Seventeen
lineages (100%) occur below 43 °N, where Buxus populations are fragmented and experienced a
severe reduction in the last few millennia. By contrast, a single haplotype is present throughout
central and western Europe, where Buxus populations are continuous and experienced an
exponential growth throughout the Holocene. The average number of mutational differences
between lineages indicate that the Mediterranean Peninsulas had quite divergent histories,
especially between W-C (Iberian, Italian) and C-E (Balkan, Anatolian). Statistical parsimony
analyses did not support a phylogeographic break between the southern and the northern territories
of Europe, as well as between Western and Eastern Mediterranean; instead, a well defined group
including those populations from the Balearic islands and Sardinia was differentiated from the
remaining populations. Comparisons of genetic and palaeobotanic data show that a conspicuous
portion the genetic diversity of western Palaearctic Buxus may have been lost in the last few
thousand years. This poses some questions on the importance of the Southern Peninsulas as longterm refuge area and stress the importance of considering long-term population dynamics based on
fossil data to evaluate the vulnerability of modern fragmented plant populations in view of
conservation actions. Extensive haplotype sharing between the two studied species at the western
and eastern edges of their Mediterranean distribution suggest that two independent events of
hybridization, one ancient and one more recent, are the most likely explanation for the lack of a
west-east divide in this genus.
Keywords: Buxus, refugia, phylogeography, hybridization, population dynamics
46
INTRODUCTION
Molecular surveys of plant and animal populations revealed common patterns of genetic
differentiation in the Western Palaearctic area. These patterns were mostly explained with the shifts
in distribution ranges that occurred in response to Quaternary climatic oscillations, with particular
consideration for the effects of the last glacial period. Several studies concurred to define glacial
refugia for hundreds of plant and animal taxa, identify pathways of post-glacial migration, evaluate
the degree of genetic reshuffling and hybridization and, ultimately, to understand the temporal
frame of speciation (Taberlet et al. 1998, Petit et al. 2003, Hewitt 2004). The glacial refugia for
temperate trees and shrubs were several times identified in the Peninsulas that surround the northern
Mediterranean Sea. Their central significance for the long-term persistence of plant taxa has been
broadly supported by interpretation of fossil records (Bennett et al. 1991). Nevertheless, modern
studies involving fossil and genetic data (see Bhagwat & Willis 2008 and references therein)
indicate that some broadleaved deciduous (e.g. Fagus sylvatica L., Corylus avellana L.) and
needleleaved evergreen (e.g. Taxus baccata L., Pinus sylvestris L., Juniperus communis L.) trees
may have survived the last glaciation in central Europe as well. These findings resize the centrality
of Mediterranean refugia in the interpretation of phylogeographic patterns. Also, they offer an
invaluable opportunity to shift the focus towards the mechanisms that allowed populations to
persist, the processes that took place during climatic oscillations and the consequences that those
processes had on the genetic structure and identity of species (Nieto-Feliner 2011).
Strict Mediterranean taxa seems to be influenced by a more complex geological, physiographical
and climatic history. The genetic pattern of many plant and animal populations in the Mediterranean
area is conceived as more ancient and complicated than that of those populations primarily affected
by Quaternary climatic oscillations (see Thompson 2005 and references therein). Surprisingly,
many attractive hypotheses on the origin of strict Mediterranean taxa distribution have remained
unexplored by phylogeographic studies. A common biogeographical pattern between hundreds of
taxonomic and phylogenetically unrelated organisms, including several animal phyla (Ribera &
Blasco-Zumeta 1998, Sanmartín 2003), lichenized fungi (Egea & Alonso 1996), bryophytes
(Montserrat et al. 1981), and vascular plants (Davis & Hedge 1971), is the extreme disjunct
distribution of taxa between the Western and Eastern Mediterranean domains. Dispersal across preexisting barriers and vicariance through fragmentation and isolating events have been often
contrasted as competing processes primarily responsible for these biological disjunctions (Stace
1989). Disjunctions in the range of species have fascinated biologists since they were first detected
and their interpretation has long been regarded as one of the central and heated problems of plant
biogeography (Raven 1972, Thorne 1972). However, few work has been devoted to assess the
phylogeographical pattern of Mediterranean plants showing intraspecific disjunctions and to infer
the underlying historical causes (Rosselló et al. 2007, Kadereit & Yaprak 2008).
In purest form, phylogeographic appraisals deal with the principles and processes governing the
geographical distribution of genealogical lineages whose phylogenetic relationships are known or
can be estimated, especially in groups of closely related species (Avise et al. 1987). Examples of
closely related tree species showing contrasting widespread temperate and Mediterranean disjunct
distributions are few. These rare examples offer a unique occasion to study a complex scenario of
genetic patterns and contrasting histories. For these reasons, we have decided to undertake a
phylogeographic survey on Buxus L. This genus has two shrub or small tree sister species (von
Balthazar et al. 2000) distributed in temperate and Mediterranean Europe, north-western Africa,
Middle East and Caucasia. Western Palaearctic Buxus has a very long history in the area, as
witnessed by a continuous fossil record since the Lower Eocene and Miocene, through the Pliocene
and Early Pleistocene (Bessedik 1983, Kvaček et al., 1982; Leroy and Roiron 1996). The common
box (B. sempervirens L.) is widely present across Europe, and occurs in NW Africa, Middle East
and Caucasia as well (Fig. 1). It lives found on different substrates (limestone, ultramafic rocks,
tuff) and in various vegetation types (deciduous and evergreen broadleaved forests, evergreen
47
needled woodlands, garigues, and calcareous grasslands). B. sempervirens is often confined along
slopes, river valleys, canyons, gorges, ravines, and thermal springs, which provide suitable subhumid conditions. By contrast, the Balearic box (B. balearica Lam.) shows an east-west
Mediterranean disjunction, occurring in S Spain, Balearic Islands, Sardinia, NW Africa and S
Anatolia (Fig. 1). As the common box, it occurs on diverse substrates, including limestone (Balearic
Islands, Sardinia), ophiolite (S Anatolia) and metamorphic crystalline rocks with ultramafic
inclusions (Alboran Block, S Spain). The Balearic box is found in both deciduous and evergreen
broadleaved forests (Balearic Islands, Sardinia, Anatolia), and in needled woodlands as well (S
Spain). Consistently with the whole genus, the common and Balearic box reproduce sexually
through insect and wind pollination, but are also capable vegetative reproduction through stolons or
buried branches. Seed dispersal consists of an ejection mechanism that account for a localized
dissemination (Debusche & Lepart 1992), and myrmechocory was observed in both B.
sempervirens and B. balearica. However but few ant species are effective dispersers while most are
consumers (Lázaro et al. 2006). The plants are long lived and slow growing (Salvesen & Kanz
2009), sought for their wood (Record 1925) and for ornamental purposes, as testified by the vast
number of existing cultivars and hybrids (Batdorf 2005).
The aim of the present study is to i) study the organization of cpDNA variation in two widely
distributed sister species showing contrasting histories and distributions, ii) examine the congruence
between genetic and palaeobotanic data, iii) evaluate historical population dynamics effects on the
modern configuration of genetic diversity, iv) investigate the presence of phylogeographic patterns
and v) elucidate the origins of the extreme disjunct distribution between the Western and Eastern
Mediterranean basin.
MATERIALS AND METHODS
Plant sampling and DNA extraction
Forty-seven and eighteen representative populations of B. sempervirens and B. balearica were
sampled in the field, respectively, covering the entire range of the species (Tab. 1 and Fig. 1). As a
rule, between 50 and 100 m were used as a minimum distance between plants to avoid sampling of
closely related individuals, or even clones. Fresh and healthy leaves from a total of 340 individuals
(222 for B. sempervirens and 118 for B. balearica) were dried in silica gel and stored at room
temperature until DNA extraction. Total genomic DNA was isolated and purified using the
DNeasyTM Plant Minikit (Qiagen, Hilden, Germany) following the manufacturer’s instructions.
DNA amplification and sequencing
The trnT–trnL intergenic spacer was amplified using the universal primers A and B described in
Taberlet et al. (1991). PCR reactions were performed in 50 μL, containing 10X reaction buffer,
0.001% BSA, 2 mm MgCl2, 0.5 mm of each dNTP, 0.6 μm of each primer, c. 50–100 ng genomic
DNA and 3 units DNA polymerase (NETZYME™ DNA Polymerase; NEED SL, Valencia, Spain).
Thermal cycling started with a denaturizing step at 94°C lasting 2 min, followed by 30 cycles each
comprising 50 s denaturizing at 94°C, 50 s annealing at 55°C, and 1.5 min elongation at 72°C, and
ended with a final elongation cycle of 3 min at 72°C. Amplifications were carried out on an ABI
GeneAmp PCR System 9700 Thermal Cycler (Applied Biosystems, Foster City, CA, USA). PCR
products were visualized on 2% agarose gels, purified using the High Pure PCR Product
Purification Kit (Roche Diagnostics, Barcelona, Spain) and sequenced with an ABI 3100 Genetic
Analyzer using the ABI BigDye™ Terminator Cycle Sequencing Ready Reaction Kit (Applied
Biosystems). Samples were sequenced in both the forward and reverse directions. The sequences
obtained were aligned manually.
48
B. sempervirens
COUNTRY
Albania
Austria
France
France
France
France
France
France
France (Corsica)
France (Corsica)
Georgia
Georgia
Germany
Germany
Greece
Greece
Greece
Greece
Iran
Iran
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Italy
Luxembourg
Macedonia
Macedonia
Montenegro
Spain
Spain
Spain
Spain
Turkey
Turkey
Turkey
Turkey
United Kingdom
United Kingdom
United Kingdom
United Kingdom
SITE
Çibërraka
Kienberg
Coursegoules
Tagolsheim
Guise
Les Plans*
Le Pré
Quimperlé
Moltifao
Bonifacio
Nikortsminda
Khobistskali
Karden
Oberdorf
Katara Pass
Pixariá
Naousa
Mount Olympus
Nour
Ab Dang Sar
Bargonasco
Vicalvi
Fiume Bussento
Monteluco di Spoleto*
Mofeta di Rapolano
Pianico
Monte Brione*
Fiume Panaro
Parco di Veio
Fiastra
Torrente Sentino*
Ahn
Matka
Demir Kapja
Krasovina
Ribes del Freser
Iglesuela del Cid
Sierra de Cazorla
Barranc de l'Emboixar
Gölkug
Gökbel
Tevekkeli
Alihocalar
Box Hill
Ellensborough Warren
Boxwell Court
Boxley*
CODE
LIB
KIE
COU
TAG
GUI
LSP
LEP
QUI
MOL
BON
NIK
KHO
KAR
OBE
KAT
PIX
NAO
PRI
NOU
ADS
BAR
VIC
CAS
MON
RAP
PIA
BRI
PAN
VEI
FIA
SCH
AHN
MAT
DMK
KRA
RDF
IDC
SDC
MTG
GOL
GOK
TEV
ALI
BXH
ELL
BXW
BXL
LAT
41.17
47.92
43.78
47.67
49.9
43.76
45.19
47.84
42.48
41.44
42.48
42.64
50.18
47.49
39.79
38.7
40.64
40.08
36.58
36.33
44.28
41.68
40.14
42.82
43.28
45.94
45.88
44.34
42.1
43.22
43.41
49.63
41.98
41.37
42.58
42.3
40.5
37.9
38.81
41.43
41.21
37.48
40.81
51.26
51.75
51.63
51.31
LON
20.25
14.35
7.07
7.27
3.6
3.28
4.13
-3.55
9.12
9.15
43.06
42.2
7.29
7.71
21.2
23.63
22.05
22.39
52.05
52.89
9.47
13.61
15.57
12.82
11.59
10.27
10.87
10.96
12.4
12.41
12.68
6.43
21.33
22.22
19.15
2.1
-0.38
-2.94
0.13
31.85
32.42
36.98
29.82
-0.32
-0.79
-2.26
0.55
ALT
280
530
875
305
140
486
850
38
350
178
1075
320
124
380
1516
807
436
1455
-13
280
72
477
210
610
280
330
230
480
230
215
584
155
393
173
204
1250
1417
1015
576
170
183
817
546
126
160
176
146
N
3
6
6
3
6
1
3
6
6
6
6
6
3
6
6
6
6
6
3
6
6
3
7
2
3
6
1
6
7
6
1
3
6
3
6
5
3
6
6
6
6
5
6
6
3
3
1
h
M
H
H
H
H
H
H
H
C
C
O
S
H
H
R
Q
P
M
O
O
H
C
N
C
C
C
C
C
C(4), N(3)
C
C
H
M
M
C
H
H
L
H
S
R
C
S
H
H
H
H
Tab. 1 Sampling sites, samples sizes (N), and haplotypes (h) recorded in each site for B. sempervirens and B.
balearica. The geographical location of each population is represented in Figure 1. * Sample not included in
the calculation of population genetic diversity parameters.
49
Buxus balearica
COUNTRY
Italy
Italy
Morocco
Morocco
Morocco
Morocco
Morocco
Spain (Baleares)
Spain (Baleares)
Spain (Baleares)
Spain (Baleares)
Spain
Spain
Spain
Spain
Turkey
Turkey
Turkey
SITE
Carbonia
Monte Tassua
Jebha
Boulemane
Cirque de Jaffar
Gorges du Todra
Talembote
Artá
Cap Ventos
Ternelles
Andratx
Frigiliana
Cuevas de Nerja
Velez de Benaulla
Ragol
Adrasan
Feke
Antakya
CODE
MTB
MTA
JEB
BOU
CDJ
GDT
TAL
ART
CPV
TER
AND
FRI
CDN
VDB
RAG
ADR
FEK
ANT
LAT
LON
ALT
N
h
39.20
39.18
33.43
32.57
31.60
35.05
35.27
39.59
39.78
39.15
36.76
36.78
36.98
39.90
36.85
36.34
36.21
37.99
8.53
8.54
-4.62
-4.91
-5.59
-4.76
-5.27
2.42
3.35
2.96
-3.85
-3.88
-2.69
2.97
-3.38
30.51
36.18
35.85
150
150
900
1100
1000
150
160
75
340
50
150
350
440
650
1200
300
150
750
7
6
6
7
7
6
7
7
6
7
7
5
7
6
7
6
7
7
E
E
H
H
H
H
H
G
F
F
H
H
I
G
H
A(4), B(2)
C(6), D(1)
C
Tab. 1 Continued
Data Analysis
Measures of genetic diversity and population differentiation were calculated with two different
analyses: one taking into account the frequencies of haplotypes (HAPLODIV software; Pons & Petit,
1995), and one incorporating both the haplotype frequencies and the genetic distances between
haplotypes (HAPLONST software; Pons & Petit, 1996). Occurrence of a significant phylogeographic
structure was inferred by testing if GST (coefficient of genetic variation over all populations) and
NST (equivalent coefficient taking into account the similarities between haplotypes) were
significantly different by use of 10,000 permutations (PERMUT software, Pons & Petit, 1996).
The geographical structure of genetic variation was assessed by analysis of molecular variance
using ARLEQUIN ver. 3.5.1.3 (Excoffier and Lischer 2010). The total genetic variance was
partitioned into covariance components at different hierarchical levels: (1) without systematic or
regional grouping; (2) only with systematic grouping; (3) with both systematic and regional
grouping, thus exploring the partitioning of the total genetic variance under multiple phylogenetic
and biogeographic hypotheses. The significance levels of the variance components were obtained
by non-parametric permutation using 10,000 replicates. In addition, SAMOVA (spatial analysis of
molecular variance) was used to identify groups of populations that were geographically
homogeneous and maximally differentiated from each other using the software SAMOVA ver. 1.0
(Dupanloup et al., 2002). This method is based on a simulated annealing procedure that maximizes
the proportion of total genetic variance due to differences between groups of populations, and also
leads to the identification of genetic barriers between these groups. The most likely number of
groups (K) was identified by repeatedly running the program for 10,000 iterations for K  {2, ..., 5}
using 3000 random initial conditions, and retaining the largest ΦCT values (i.e. the largest proportion
of total genetic variance due to differences between groups) as predictors of the best grouping of
populations.
Patterns of isolation by distance were assessed by testing the correlation between the matrix of pair
wise ΦST values and the matrix of geographical distances between pairs of populations using a
50
Mantel test (10,000 permutations) implemented in the MANTEL ver. 2.0 package (Liedloff, 1999).
Pair wise ΦST values were computed using ARLEQUIN ver. 3.5.1.3 (Excoffier and Lischer 2010), and
significance levels of the estimated values were obtained by permutation using 10,000 replicates.
Finally, unrooted statistical parsimony networks were constructed using TCS ver. 1.21 software
(Clement et al., 2000). Root probabilities were also calculated using the TCS program following the
method of Castelloe & Templeton (1994).
RESULTS
Sequence and haplotype variation
The length of the trnT–trnL spacer ranged between 665 and 685 bp, and resulted in an aligned
matrix of 720 bp. Site mutations were detected at alignment positions 56, 68, 69, 178, 231, 420,
438, 522 and 588. Insertion–deletion polymorphisms included eight mutations varying between 4
and 25 bp and two mononucleotide poly-T and poly-A stretches composed of 8–10 and 8–11
nucleotides, respectively. These mutations defined seventeen haplotypes (Appendix 1). With the
exception of two populations from Anatolia (B. balearica) and one from Italy (B. sempervirens)
showing two haplotypes, all the sampled populations harbored only a single haplotype (Tab. 1, Fig.
1). A total of 10 (C, H, L-S) and 9 (A-I) haplotypes were detected in B. sempervirens and B.
balearica, respectively. The most frequent and widespread haplotype (H, 40%) occur across C, W
and SW Europe and NW Africa. The second most common chlorotype (C, 19%) is found in the
Italian Peninsula, in Corsica, S Turkey, and in a single population of the Balkan area (KRA). These
two common and widespread haplotypes (C and H) are shared between the two sister species at the
eastern and western sides of their distribution in the Mediterranean area. Seven haplotypes (A, B, D,
I, L, P, Q) occurred only in single populations of the Southern Mediterranean area, of which two in
S Spain, two in the Balkan area, and three in S Turkey. Haplotype M characteristic of the Balkan
region, and N of C-S Italy. Haplotypes S and O are distributed along the Black Sea coasts and the
Colchis/Hyrcanian area, respectively. Haplotype R is detected in two disjunct populations of B.
sempervirens, one in Greece (KAT) and one in N Turkey (GOK). Three haplotypes exclusively
related to B. balearica were found in samples from the Balearic Islands (D, F) and Sardinia (F).
Organization and hierarchical partitioning of cpDNA variation
Parameters of genetic diversity and structure for B. sempervirens and B. balearica are given in
Table 2. Overall, low levels of intrapopulation diversity were detected in B. sempervirens and B.
balearica across all populations, with hs = 0.014 ± 0.014 / vs = 0.009 ± 0.009 and
hs = 0.046 ± 0.032 / vs = 0.010 ± 0.007, respectively. Genetic variation in the Common and Balearic
box was highly structured geographically, based both on unordered and ordered alleles
(Gst = 0.983 ± 0.017, Nst = 0.984 ± 0.016 and Gst = 0.943 ± 0.039, Nst = 0.987 ± 0.009, for B.
sempervirens and B. balearica, respectively). However, a significant phylogeographic structure was
not deduced in either species (Nst = Gst; U= 0.07, P>0.05 and Nst = Gst; U= 1.41, P >0.05 for B.
sempervirens and B. balearica, respectively) nor in the whole dataset (Nst = Gst; U= 1.88, P>0.05).
51
Fig. 1 Geographical distribution of the seventeen chloroplast haplotypes found in B. sempervirens and B.
balearica. Dashed lines entangles populations of B. balearica. For population codes, see Tab. 1. The detailed
distribution of the taxa is presented in the inset map (redrawn from Di Domenico et al. 2012). Green and light
green represent the continuous and fragmented distribution area of Buxus sempervirens, respectively. Red and
light red represent the continuous and fragmented distribution area of Buxus balearica, respectively.
B. sempervirens
N. Populations
N. Individuals
N. Haplotypes
N. Private Haplotypes
Avg. N. Haplotypes
N. Polymorphic Sites
hs
ht
Gst
vs
vt
Nst
Nst – Gst
42
204
10
8
5.14
12
0.014 ± 0.014
0.789 ± 0.048
0.983 ± 0.017
0.009 ± 0.009
0.586 ± 0.098
0.984 ± 0.016
0.001
B. balearica
18
118
9
7
6.56
10
0.046 ± 0.033
0.792 ± 0.085
0.943 ± 0.039
0.010 ± 0.007
0.794± 0.101
0.987± 0.009
0.044
Overall
60
322
17
5.5
19
0.023 ± 0.014
0.801 ± 0.042
0.971 ± 0.017
0.010 ± 0.007
0.640 ± 0.088
0.984 ± 0.011
0.013
Tab. 3 Parameters of population genetic diversity and substructure in western
Palaearctic Buxus. Estimates of genetic diversity and differentiation measures for
unordered (hs, ht, Gst) and ordered (vs, vt, Nst) haplotypes are provided, together with
their standard deviations. hs and vs, intrapopulation diversity, ht and vt, total
diversity, Gst and Nst, degree of differentiation among populations.
52
The results obtained from AMOVA showed that, when no regional or systematic grouping was
considered, most of the cpDNA variation found in the two sister species can be attributed to
differences between populations (98.38%, Table 3). Generally, at all the hierarchical levels
considered, little variation occur within populations (<3%). When variation between species was
considered, AMOVA found that only 12.75% of the cpDNA variation was distributed between the
two taxa. Within each taxon, assuming no regional grouping, AMOVA showed that most of the
variation is found between populations (97.82% and 98.86% for B. sempervirens and B. balearica,
respectively). When variation between regions for each species was taken into account, AMOVA
found that 31.30% of variation was distributed between the Southern and Northern territories for B.
sempervirens, and that only 10.10% of the cpDNA variation was distributed between the Eastern
and Western Mediterranean for B. balearica. Instead, most of the variation among groups (80.06%)
was revealed when the hypothesis involving three regional units for B. balearica was considered
(Iberian Peninsula and North Africa, Western Mediterranean islands, Anatolia).
Hierarchy
Source of variation
df
S.S.
V.c.
% of
variation
a
Among populations
Within populations
59
271
536.97
7.33
1.65
0.03
98.38*
1.62
b
Among species
Among populations within species
Within populations
1
58
271
44.36
492.61
7.33
0.22
1.54
0.03
12.75*
85.74*
1.51*
c
Among populations
Within populations
41
171
281.092
5.143
1.35
0.03
97.82*
2.18
d
Among populations
Within populations
17
100
211.51
2.19
1.90
0.02
98.86*
1.14
e
Among groups
Among populations within groups
Within populations
1
32
135
32.945
130.753
5.143
0.39
0.82
0.04
31.3*
65.64*
3.06*
f
Among groups
Among populations within groups
Within populations
1
16
100
19.19
192.31
2.19
0.21
1.83
0.02
10.10
88.84*
1.06*
g
Among groups
Among populations within groups
Within populations
2
15
100
161.11
50.40
2.19
2.14
0.51
0.02
80.06*
19.12*
0.82*
Tab. 4 Table 4. Analysis of molecular variance (AMOVA) based on trnT-trnL sequence data for B.
sempervirens and B. balearica. (a) Assuming no systematic differentiation; (b) According with systematic
differentiation; (c) Assuming no regional differentiation in B. sempervirens; (d) Assuming no regional
differentiation in B. balearica; (e) southern versus northern territories for B. sempervirens; (f) Eastern
Mediterranean versus Western Mediterranean for B. balearica; (g) three regional units for B. balearica
(Iberian Peninsula and North Africa, Western Mediterranean islands, Anatolia). df = degrees of freedom;
S.S. = sum of squares; V.c. = variance components; * = P < 0.001 after 10,000 permutations (otherwise not
significant).
53
Spatial AMOVA (SAMOVA) did not allow us to clearly identify one single group of maximally
differentiated populations, as ΦCT values increased progressively with increasing values of K (ΦCT
values ranging from 0.688 to 0.788). The first level of divergence (K=2) revealed a well-defined
group including those populations of B. balearica from the Balearic islands and Sardinia showing
the divergent E, F and G haplotypes, which were differentiated from the remaining populations of
both B. sempervirens and B. balearica (68.82%). When three groups were considered (K=3) one
populations of B. sempervirens from Greece showing a private haplotype (NAO) and the
populations from the Colchis and Hyrcanian and area were split from the larger group formerly
detected for K=2. When considering four groups, the population from Greece (NAO) was separated
from those of Georgia and Iran, and placed together with another population from Greece showing a
private haplotype (PIX). Finally, for K=5, the two populations from Greece (NAO and PIX) were
split into two separate groups.
A significant correlation between genetic (ΦST) and geographic distances were found as evidenced
by the Mantel test in both B. sempervirens (r = 0.397, P < 0.005) and B. balearica (r = 0.426, P <
0.01). However, this correlation was not significant when considering the dataset as a whole (r =
0.089, P > 0.05).
Phylogenetic relationships among haplotypes
TCS calculated a 95% parsimony connection limit of 11 steps for the seventeen haplotypes (Fig. 2).
Nine haplotypes (B, C, F, G, I-M, R, S) were nested in the network as interior nodes whereas eight
(A, D, E, H, N, O, P, Q) occupied tip clades. Missing or not sampled intermediate haplotypes in the
network (nine) were identified between F-S (two), P-R and O-R (three) lineages. All other
haplotypes were one mutational step apart from each other. The highest root probability was
assigned by TCS to the Italian-Corsican-Anatolian haplotype C (P = 0.156). The phylogenetic
relationships and the root of the parsimony network remained unchanged after including a sequence
from the Asiatic B. mycrophylla Siebold & Zucc. (GenBank accession NC009599).
Fig. 2 Phylogenetic reconstruction of chloroplast haplotypes in Buxus using statistical parsimony. Circle sizes are
proportional to the mean frequency of each haplotype. Small black circles represent extinct or not sampled
haplotypes. In the dashed box is depicted the relationship between the network for western Palaearctic Buxus and a
sequence from B. mycrophylla.
54
DISCUSSION
Organization of cpDNA diversity in Buxus
In Western Palaearctic Buxus, cpDNA diversity is highly subdivided among populations. In facts,
Buxus settles in the top-right corner of an hypothetical graph comparing GST and NST values of
European temperate tree taxa (see Aguinagalde et al. 2005 and references therein). Also, our
cpDNA dataset shows values of haplotypic diversity close to zero and lack of a phylogeographic
structure. A similar situation is not a general feature of temperate woody taxa genetic diversity.
Comparable values has been detected only in a few species, i.e. Carpinus betulus and C. orientalis
(Grivet & Petit 2003) and in Fraxinus angustifolia, F. excelsior and F. ornus (Heuertz et al. 2006).
This pattern was explained by limited dispersal abilities, low chloroplast mutation rate or low Ne
during glacial periods. By all accounts, the values we have detected in Buxus are not the product of
inappropriate sampling strategy or ability of the marker to detect enough polymorphism. Indeed, we
have sampled Buxus throughout its distribution ranges and our results are in line with previous
studies on tree taxa using standardized sampling protocols (mean number of haplotypes = 16.9,
mean number of populations = 21.3; Petit et al. 2003). Therefore, the organization of genetic
diversity in Buxus can be related to the ecology and reproductive biology, as well as to historical
population dynamics assessed by independent palaeoecological investigations (Kvaček et al. 1982,
Di Domenico et al. 2012).
Both B. sempervirens and B. balearica are poor dispersers (Debusche & Lepart 1992, Lázaro et al.
2006), slow growing and long-lived trees. In addition, the common box shows a rampant ability to
reproduce asexually through stolons or broken/buried branches. Reviews of Buxus Tertiary fossil
occurrences revealed that the genus was able to cope with Western Palaearctic climatic and
palaeogeographic changes since the Eocene. For example, modifications of its ecological position in
the succession from tropical-subtropical, to temperate and ultimately seasonal vegetations without
major variations in its distribution are proofs of its adaptability (Kvaček et al. 1982). This historical
plasticity has its counterpart in some plant traits of Buxus. In facts, boxwood is able to withstand a
vast array of adverse ecological conditions, including frost, cold, heat and changing light regimes
(Hormaetxe et al. 2007, García-Plazaola et al. 2008, Hacker and Neuner 2007, Walter 1929).
Despite a wide geographic distribution in various vegetation types, Buxus is often linked with high
levels of moisture (Lenoble and Broyer 1945, Pigott and Pigott 1993). For this reason, the
populations are usually confined to restricted habitats occurring with a patchy distribution across
the landscape. Examples of these ‘sink’ habitats are river valleys, gorges, canyons, and misty
valleys, which may have hampered genetic exchanges between demographic units. Coupled with
the conservative nature of the cpDNA molecule, the above ecological features are the most likely
explanation for the organization of genetic diversity in Buxus. All in all, this high level of genetic
organization suggest that the phylogeographic signal is related to a protracted persistence over the
Western Palaearctic area. Furthermore, despite the two studied species show contrasting histories
and distributions, the organization of cpDNA diversity is strikingly similar. This suggests that the
observed pattern is a general feature of the genus.
Genetic differentiation in Mediterranean and Temperate Buxus populations
The location and survival of trees in the coldest stages of the last full-glacial has long been of
interest to palaeoecologists, biogeographers, and geneticists. In particular, where species survived in
isolated refugia and the influence that this has had upon the long-term ancestry of populations
remain key research questions. However, the exact location of refugia during the coldest stages of
the full-glacial still remains illusive for many tree and shrub taxa. Emerging evidence from various
fossil proxies, palaeoclimatic modelling and genetic research on broadleaved deciduous (Fagus
sylvatica L., Corylus avellana L.) and needleleaved evergreen (e.g. Taxus baccata L., Pinus
55
sylvestris L., Juniperus communis L.) tree taxa is starting to suggest that the traditional paradigm
that such species were restricted to southern Europe and in particular the southern Peninsulas during
the full-glacial is questionable (Willis & van Andel 2004).
Aside the unusual organization of cpDNA diversity, Buxus matches some of the expected
commonalities of many temperate woody taxa geographic distribution of genetic diversity (Petit et
al. 2003). In facts, the number of haplotypes is maximum in the southern territories of the western
Palaearctic area, especially in the southeastern regions. By contrast, only one haplotype, also found
in putative Mediterranean refugia, occurs across central Europe. It is a common view that such
patterns may reflect the long-term persistence of Buxus in Mediterranean Peninsulas. This view is
also substantiated by palaeobotanical data (Wegmüller 1984, Lang 1994, Di Domenico et al. 2012).
The average number of mutational differences between lineages located in each Peninsula suggest
that every territory had a different history and influence on the genetic diversity of Buxus. The
Italian and Iberian Peninsulas (average number of mutational differences = 2.7) show a similar and
more homogeneous history as compared to Anatolian (average number of mutational differences =
6.5) and Balkan (average number of mutational differences = 9.2). In the latter cases, lineage
divergence suggest an ancient persistence and isolation, as well as an individualistic ecological
response of population units.
Our cpDNA dataset do not provide direct support to Buxus northern persistence, as suggested by
ecological and palaeobotanical data (Di Domenico et al. 2012). In facts, it was not possible to detect
any northern private lineage. Some populations sampled in the close proximity of last glacial
maximum Buxus fossil sites show haplotype H. This indicate that the occurrence of haplotype H in
central Europe may predate the onset of last glaciation. If so, a new interpretation for Buxus
populations genetic homogeneity is desirable. Historical population dynamics and modern
distribution of Buxus in Central Europe may help providing an explanation. During the last glacial
period, the E Pyrenees, W Alps, and Jura Mts hosted relatively dense populations, recorded very
early by pollen analysis. Those populations vigorously expanded throughout the Holocene, and
resulted in a modern continuous distribution area. Therefore, it appear that Buxus populations in
Central Europe were abundant throughout the last glacial period, allowing extensive gene flow to
counterbalance the fixation of newly arisen mutation. Considering that the last glacial maximum
was one of the harshest of the Quaternary, it is possible that extensive gene flow between Central
European populations has occurred throughout the Pleistocene (Magri 2010).
The effects of historical Mediterranean population dynamics on the genetic resources of Buxus
deserve a discussion as well. In the southern territories of Europe the populations vigorously
increased only during the first half of the Holocene. Thereafter, Buxus populations markedly
decreased, probably in relation to increasing aridity around the Mediterranean basin (Di Rita et al.
2011, Jalut et al. 2009, Magny et al. 2011, Zanchetta et al. 2007). This decline is most likely
responsible for the highly fragmented distribution area that Buxus shows in the southern territories
of Europe. Lessons from ecology tell that fragmentation is the most serious threat to biological
diversity and the main driver of extinctions (Wilcox & Murphy 1985). Thus, fragmentation may
have played an important role in the local extinction of Buxus populations scattered in different
territories, including Minorca, Ibiza, Sardinia, Sicily, Apulia, Basilicata, Croatia, Thessaly, W
Greece, and Peloponnese (Di Domenico et al. 2012). The combination of high genetic diversity in
the south, especially of the unique haplotypes that were detected in insular systems or in the
southernmost territories of the Mediterranean Peninsulas, coupled with the position of missing
haplotypes in the parsimony network, which are placed between lineages occurring in the
Mediterranean area, require careful consideration. It suggests that together with the disappearance
of many populations, a conspicuous portion the genetic diversity of Buxus may have been lost in
the last few thousand years. This hypothesis challenges the trite statement that the southern
Peninsulas are the most important areas for the long-term preservation of the modern genetic
resources of woody taxa. Also. it stress the importance of considering long-term population
56
dynamics based on fossil data to evaluate the vulnerability of modern fragmented plant populations
in view of conservation actions.
Lack of East-West phylogeographic break in western Palaearctic Buxus
The distribution of vascular plant species and landscape vegetation throughout the Mediterranean
basin show a profound biogeographic differentiation that is more significantly related to a
longitudinal east-west divide rather than a latitudinal split into northern-southern shores (Thompson
2005). This has been substantiated either by narrative and intuitive views or by numerical
phytogeographical approaches comparing the floristic affinities of selected territories (Junikka et al.
2006). Moreover, morphological east-west differentiation and vicariance has been suggested for
some putative, closely related sister taxa showing disjunct distributions (e.g. Pinus halepensis-P.
brutia, Quercus coccifera-Q. calliprinos, Cyclamen repandum subspecies; Thompson 2005).
Significantly, this organismal and biogeographic pattern is also mirrored by intraspecific molecular
genealogies documented from a number of strict Mediterranean plant species showing continuous
distributions. Contrasting molecular differentiation between western and eastern populations has
been reported not only for terrestrial species, including Ficus carica (Khadari et al. 2005), Olea
europaea subsp. europaea (Besnard & Bervillé 2000, Breton et al. 2006) and Laurus nobilis
(Rodríguez-Sánchez et al. 2009), but also for marine seagrasses like Posidonia oceanica (Micheli et
al. 2005). Furthermore, major genetic subdivisions reflecting a west-east phylogeographic break
throughout the Mediterranean have even been reported for submediterranean (e.g., Anthyllis
montana, Kropf et al. 2002) and widespread Euro-Asiatic inland (e.g., Frangula alnus, Hampe et al.
2003; Hedera helix, Valcárcel et al. 2003) or coastal taxa (e.g., Eryngium maritimum, Halimione
portulacoides, Kadereit et al., 2005). This biogeographic break might be intimately linked to the
extremely complex geological history, palaeogeography and palaeoclimatology of the
Mediterranean basin that fragmented and merged biotas as dispersal barriers appeared and
disappeared through time. Ultimately, the glacial episodes of the Quaternary (starting 2.6 Ma; Webb
& Bartlein 1992, Lambeck et al. 2002), and most recently the last glacial maximum, have had major
impacts on the present-day distribution of European species and their lineages (Hewitt 2004). The
long-term isolation of populations within geographically separate refugia in the southern territories
of the western Palaearctic area (Iberian, Italian, Balkan and Anatolian Peninsulas; Taberlet et al.
1998) might shaped differentially the spatial structuring and genetic differentiation of intraspecific
lineages in Mediterranean plants, further enhancing the distinctiveness of biogeographic breaks in
this area. Hence, empirical data drawn from many sources would predict the existence of a cpDNA
split encompassing the western and eastern populations of B. balearica. Contrary to expectations
and to the results obtained using nuclear ribosomal nuclear markers (Rosselló et al. 2007), this
west-east divide is not supported by our cpDNA data set. Interestingly, a new unrecognized
phylogeographic pattern emerged (Western islands vs. continental Mediterranean areas).
Conflicting genealogical signals retrieved by nuclear and cpDNA markers may be caused by several
processes, e.g. the inherent contrasting properties of the nuclear and cytoplasmic genomes analyzed
(e.g., different patterns of evolution, heritage, and recombination), stochastic factors (incomplete
lineage sorting), as well as the questionable utility of the ribosomal ITS region as a suitable marker
due to the presence of multiple copies and loci within the nuclear genome (Álvarez & Wendel 2003,
Nieto Feliner & Rosselló 2007). However, such conflict is generally explained by interspecific gene
flow resulting in more or less complex scenarios of hybridization and introgression. Regional
sharing of chloroplast haplotypes across morphospecies is a consequence of hybridization and could
be expected if directional introgression has occurred via asymmetric flow of pollen, (e.g. Rieseberg
& Soltis 1991, McKinnon et al. 2001, Petit et al. 2004, Albaladejo et al. 2005). Interestingly, B.
sempervirens and B. balearica shares some haplotypes at western (haplotype H) and eastern
(haplotype C) populations, where both species grow at close proximity. Therefore, ancient and
independent chloroplast capture (i.e., introgression followed by relatively rapid fixation of an
57
introgressed haplotype) of B. sempervirens haplotypes by B. balearica in both Western and Eastern
Mediterranean basin could be invoked as a likely explanation to reconcile the facts that the nuclear
multigene ITS family show species-specific markers, no signs of hybridization across the B.
balearica range, and a west-east phylogeographic break (Rosselló et al. 2007). Similar processes
involving extensive gene flow has been pointed out to blur the phylogeographic, but not the
phylogenetic signal, in other Mediterranean woody plants like Olea europaea L. (Rubio de Casas et
al. 2006). Additional support for the hybridization scenario here proposed for B. balearica comes
from morphology. Benedí (1997) reported that populations of this species from the Iberian
Peninsula and North Africa differ from those present in the Balearic Islands by the presence of hairy
shoots and a lower style/capsule length ratio, two of the discriminating features used to identify B.
sempervirens from B. balearica. In southern Turkey, the populations of B. balearica differ from the
western Mediterranean insular populations for hairy shoots and narrower leaves (Boissier 1853),
other characters used to discriminate B. sempervirens from B. balearica. Both nuclear and
chloroplast markers reject the hypothesis that, given the highly appreciated use of the Balearic box
wood by humans for several purposes in prehistoric times (e.g., charcoal, manufacture, ritual; Piqué
& Noguera 2002), anthropogenic introductions in historical times are responsible of the disjunct
pattern seen in B. balearica. Furthermore, the fixed presence of a private cpDNA haplotype in
Sardinia strongly suggests its wild origin in the island, in contrast with earlier views supporting an
alien status by anthropogenic introductions (Gennari 1864).
CONCLUSIONS
This first contribution to the knowledge of chloroplast variation in Western Palaearctic Buxus
revealed that the genetic resources of this genus are ‘trapped’ within population units, mimicking
the particular plant traits and ecological behavior of these woody taxa. The molecular divergence
between lineages located in the same territories and pairwise differences between southern
Peninsulas testify the existence of independent and heterogeneous population histories which
shaped chloroplast variation in a complex pattern. Results from the present study suggest that the
incorporation of palaeodemographic data from independent palaeobotanical assessments provide a
complementary tool to interpret modern genetic patterns in woody taxa. The present work also
testify the importance of ancient hybridization events in the evolution and genetic identity of closely
related species with contrasting distributions. Hybridization was shown to occur repeatedly in
disjunct areas and during different periods of Buxus history.
ACKNOWLEDGEMENTS
We are grateful to Charles Turner, Wolgang Willner, Rolf Sievers, Franz Essl, Frédéric Bioret,
Guillaume Decocq, Kris Verheyen, Olivier Honnay, Cristophe Girod, Errol Vela, Livia Zapponi,
Hamed Yosefzadeh, and Sandro Kolbaia for providing samples and help with field work.
58
REFERENCES
Aguinagalde, I., Hampe, A., Mohanty, A., Martín, J.P., Duminil, J., Petit, R.J. (2005) Effects of
life-history traits and species distribution on genetic structure at maternally inherited markers in
European trees and shrubs. Journal of Biogeography, 32, 329-339.
Albaladejo, R.G., Fuertes Aguilar, J., Aparicio, A., Nieto Feliner, G. (2005) Contrasting nuclearplastidial phylogenetic patterns in the recently diverged Iberian Phlomis crinita and P. lychnitis
lineages (Lamiaceae). Taxon, 54, 987-998.
Álvarez, I., Wendel, J.F. (2003) Ribosomal ITS sequences and plant phylogenetic inference.
Molecular Phylogenetics and Evolution, 29, 417-434.
Avise, J.C., Arnold, J., Ball, R.M., Bermingham, E., Lamb, T., Neigel, J.E., Reeb, C.A., Saunders,
N.C. (1987) Intraspecific phylogeography: the mitochondrial DNA bridge between population
genetics and systematics. Annual Review of Ecology and Systematics, 18, 489-522.
Batdorf, L.R. (2005) International Registration List of Cultivated Buxus L. The Boxwood Bullettin,
45, 98-109.
Benedí, C. (1997). Buxaceae. In: Castroviejo S., et al. (eds) Flora Ibérica. Vol. VIII, Real Jardín
Botánico. CSIC, Madrid, pp. 187-189.
Bennett, K.D., Tzedakis, P.C., Willis, K.J. (1991) Quaternary refugia of north European trees.
Journal of Biogeography, 18, 103-115.
Besnard, G., Bervillé A. (2000) Multiple origins for Mediterranean olive (Olea europaea L. ssp.
europaea) based upon mitochondrial DNA polymorphisms. Comptes Rendus de l’Académie des
Sciences, Sciences de la Vie, 323, 173-181.
Bessedik, M. (1983) Le genre Buxus L. (Nagyipollis KEDVES 1962) au Tertiaire en Europe
occidentale: évolution et implications paléogaographiques. Pollen et Spores, 25, 461-486.
Bhagwat, S., Willis, K.J. (2008) Species persistence in northerly glacial refugia of Europe: a matter
of chance or biogeographical traits? Journal of Biogeography, 35, 464-482.
Boissier, E. (1853) Diagnoses Plantarum Orientalium novarum. Nabu Press, Ginevra.
Breton, C., Tersac, M. , Bervillé, A. (2006) Genetic diversity and gene flow between the wild olive
(oleaster, Olea europaea L.) and the olive: several Plio-Pleistocene refuge zones in the
Mediterranean basin suggested by simple sequence repeats analysis. Journal of Biogeography,
33, 1916-1928.
Castelloe, J., Templeton, A.R. (1994) Root probabilities for intraspecific gene trees under neutral
coalescent theory. Molecular Phylogenetics and Evolution, 3, 102-113.
Clement, M., Posada, D., Crandall, K.A. (2000) TCS: a computer program to estimate gene
genealogies. Molecular Ecology, 9, 1657-1659.
Davis, P.H., Hedge I.C. (1971) Floristic links between N.W. Africa and S.W. Asia. Annalen Des
Naturhistorischen Museums in Wien, 75, 43-57.
59
Debussche, M., Lepart, J. (1992) Establishment of woody plants in Mediterranean old fields:
opportunity in space and time. Landscape Ecology, 6, 133-145.
Di Domenico, F., Lucchese, F., Magri, D. (2012) Buxus in Europe: Late Quaternary dynamics and
modern vulnerability. Perspectives in Plant Ecology, Evolution and Systematics, 14, 354-362.
Di Rita, F., Simone, O., Caldara, M., Gehrels, W.R., Magri, D. (2011) Holocene environmental
changes in the coastal Tavoliere Plain (Apulia, southern Italy): A multiproxy approach.
Palaeogeography, Palaeoclimatology, Palaeoecology, 310, 139-151.
Dupanloup, I., Schneider, S., Excoffier, L. (2002) A simulated annealing approach to define the
genetic structure of populations. Molecular Ecology, 11, 2571-2581.
Egea, J.M. & Alonso F.L. (1996) Patrones de distribución en la flora liquénica xerófila del sureste
de España. Acta Botanica Malacitana, 21, 35-47.
Excoffier, L., Lischer, H.E.L. (2010) Arlequin suite ver 3.5: a new series of programs to perform
population genetics analyses under Linux and Windows. Molecular Ecology Resources, 10,
564-567.
García-Plazaola, J.I., Esteban, R., Hormaetxe, K., Becerril, J.M. (2008) Seasonal reversibility of
acclimation to irradiance in leaves of common box (Buxus sempervirens L.) in a deciduous
forest. Flora, 203, 254-260.
Gennari, P. (1864) Flora medica sarda ossia descrizione della piante medicinali che crescono
spontanee in Sardegna. Sardegna Medica, Firenze.
Grivet, D., Petit, J.R. (2003) Phylogeography of the common ivy (Hedera sp.) in Europe: genetic
differentiation through space and time. Molecular Ecology, 11, 1351-1362.
Hacker, J., Neuner, G. (2007) Ice propagation in plants visualized at the tissue level by infrared
differential thermal analysis (IDTA). Tree Physiology, 27, 1661-1670.
Hampe, A., Arroyo, J., Jordano, P., Petit, R.J. (2003) Rangewide phylogeography of a birddispersed Eurasian shrub: contrasting Mediterranean and temperate glacial refugia. Molecular
Ecology, 12, 3415-3426.
Heuertz, M., Carnevale, S., Fineschi, S., Sebastiani, F., Hausman, J.F., Paule, L., Vendramin, G.G.
(2006) Chloroplast DNA phylogeography of European ashes, Fraxinus sp. (Oleaceae): roles of
hybridization and life history traits. Molecular Ecology, 15, 2131-2140.
Hewitt, G.M. (2004) Genetic consequences of climatic oscillations in the Quaternary.
Philosophical Transactions of the Royal Society of London, series B Biological Sciences, 359,
183-195.
Hormaetxe, K., Becerril, J.M., Hernández, A., Esteban, R., García-Plazaola, J.I., (2007) Plasticity
of photoprotective mechanisms of Buxus sempervirens L. leaves in response to extreme
temperatures. Plant Biologt, 9, 59-68.
60
Jalut, G., Dedoubat, J.J., Fontugne, M., Otto, T. (2009) Holocene circum-Mediterranean vegetation
changes: Climate forcing and human impact. Quaternary International, 200, 4-18.
Junikka, L., Uotila, P., Lahti, T. (2006) A phytogeographical comparison of the major
Mediterranean islands on the basis of Atlas Florae Europaeae. Willdenowia, 36, 379-388.
Kadereit, G., Yaprak, A.E. (2008) Microcnemum coralloides (Chenopodiaceae-Salicornioideae):
an example of intraspecific East-West disjunctions in the Mediterranean region. Anales del
Jardín Botánico de Madrid, 65, 415-426.
Kadereit, J.W., Arafeh, R., Somogyi, G., Westberg, E. (2005) Terrestrial growth and marine
dispersal? Comparative phylogeography of five coastal plant species at a European scale.
Taxon, 54, 861-876.
Khadari, B., Grout, C., Santoni, S., Kjellberg, F. (2005) Contrasted genetic diversity and
differentiation among Mediterranean populations of Ficus carica L.: A study using mtDNA
RFLP. Genetic Resources and Crop Evolution, 52, 97-109.
Kropf, M., Kadereit, J.W., Comes, H.P. (2002) Late Quaternary distributional stasis in the
submediterranean mountain plant Anthyllis montana L. (Fabaceae) inferred from ITS sequences
and amplified fragment length polymorphism markers. Molecular Ecology, 11, 447-463.
Kvaček, Z., Bůžek, Č. & Holý , F. (1982) Review of Buxus fossils and a new large-leaved species
from the Miocene of Central Europe. Review of Palaeobotany and Palynology, 37, 361-394.
Lambeck, K., Esat, T.M., Potter, E.-K. (2002) Links between climate and sea levels for the past
three million years. Nature, 419, 199-206.
Lang, G. (1992) Some aspects of European late- and post-glacial flora history. Acta Botanica
Fennica, 144, 1-17.
Lázaro, A., Traveset, A., Méndez, M. (2006) Masting in Buxus balearica: assessing fruiting
patterns and processes at a large spatial scale. Oikos, 115, 229-240.
Lenoble, F., Broyer, C. (1945) Sur la distribution de Buxus sempervirens L. en France. Bulletin de
la Société Botanique de France, 92, 118-131.
Leroy, S.A.G., Roiron, P. (1996) Latest Pliocene pollen and leaf floras from Bernasso palaeolake
(Escandorgue Massif, Hérault, France). Review of Palaeobotany and Palynology, 94, 295-328.
Liedloff, A.C. (1999) Mantel Nonparametric Test Calculator. Version 2.0. School of Natural
Resource Sciences, Queensland University of Technology, Australia.
Magny, M., Vannière, B., Calo, C., Millet, L., Leroux, A., Peyron, O., Zanchetta, G., La Mantia,
T., Tinner, W. (2011) Holocene hydrological changes in south-western Mediterranean as
recorded by lake-level fluctuations at Lago Preola, a coastal lake in southern Sicily, Italy.
Quaternary Science Reviews, 30, 2459-2475.
Magri, D. (2010) Persistence of tree taxa in Europe and Quaternary climate changes. Quaternary
International, 219, 145-151.
61
McKinnon, G.E., Vaillancourt, R.E., Jackson, H.D., Potts, B.M. (2001) Chloroplast sharing in the
Tasmanian eucalypts. Evolution, 55, 703-711.
Micheli, C., Paganin, P., Peirano, A., Caye, G., Meinesz, A., Bianchi, C.N. (2005) Genetic
variability of Posidonia oceanica (L.) Delile in relation to local factors and biogeographic
patterns. Aquatic Botany, 82, 210-221.
Montserrat, B.D., Cros i Matas, R.M., Casas, C. (1981) Contribució al coneixement de l'àrea
geogràfica d'alguns briòfits. Treballs de la Institució Catalana d'Història Natural, 9, 169-178.
Nieto Feliner, G. (2011) Southern European glacial refugia: A tale of tales. Taxon, 60, 365-372.
Nieto Feliner, G., Rosselló, J.A. (2007) Better the devil you know? Guidelines for insightful
utilization of nrDNA ITS in species-level evolutionary studies in plants. Molecular
Phylogenetics and Evolution, 44, 911-919.
Petit, R.J., Aguinagalde, I., de Beaulieu, J.-L., Bittkau, C., Brewer, S., Cheddadi, R., Ennos, R.,
Fineschi, S., Grivet, D., Lascoux, M., Mohanty, A., Müller-Starck, G., Demesure-Musch, B.,
Palmé, A., Martín, J.P., Rendell, S., Vendramin, G.G. (2003) Glacial refugia: hotspots but not
melting pots of genetic diversity. Science, 300, 1563-1565.
Petit, R.J., Bodénès, C., Ducousso, A., Roussel, G., Kremer, A. (2004) Hybridization as a
mechanism of invasion in oaks. New Phytologist, 161, 151-154.
Pigott, C.D., Pigott, S. (1993) Water as a determinant of the distribution of trees at the boundary of
the Mediterranean zone. Journal of Ecology, 81, 557-566.
Piqué, R., Noguera, M. (2002) Landscape and management of forest resources in the Balearic
Islands during the II -I millennium BCE. In: Waldren, W.H., Ensenyat, J.A. (eds.) World
Islands in Prehistory. International Insular Investigations. BAR International Series 1095: 292300.
Pons, O., Petit, R.J. (1995) Estimation, variance and optimal sampling of gene diversity. I. Haploid
locus. Theoretical and Applied Genetics, 90, 462-470.
Pons, O., Petit, R.J. (1996) Measuring and testing genetic differentiation with ordered and
unordered alleles. Genetics, 144, 1237-1245.
Raven, P.H. (1972) Plant species disjunctions: a summary. Annals of the Missouri Botanical
Garden, 59, 234-246.
Record, S.J. (1921) Boxwoods of Commerce. Bulletin of the Torrey Botanical Club, 48, 297-306.
Ribera, I., Blasco-Zumeta, J. (1998) Biogeographical links between steppe insects in the Monegros
region (Aragón, NE Spain), the eastern Mediterranean, and central Asia. Journal of
Biogeography, 25, 969-986.
Rieseberg, L.H., Soltis, D.E. (1991) Phylogenetic consequences of cytoplasmic gene flow in
plants. Evolutionary Trends in Plants, 5, 65-84.
62
Rodríguez-Sánchez, F., Guzmán, B., Valido, A., Vargas, P., Arroyo, J. (2009) Late Neogene
history of the Laurel Tree (Laurus L., Lauraceae) based on phylogeographical analyses of
Mediterranean and Macaronesian populations. Journal of Biogeography, 36, 1270-1281.
Rosselló J.A., Lázaro, A., Cosín, R., Molins, A. (2007) A phylogeographic split in Buxus balearica
(Buxaceae) as evidenced from nuclear ribosomal markers: when ITS paralogues are welcome.
Journal of Molecular Evolution, 64, 143-157.
Rubio de Casas, R., Besnard, G., Schönswetter, P., Balaguer, L., Vargas, P. (2006) Extensive gene
flow blurs phylogeographic but not phylogenetic signal in Olea europaea L. Theoretical and
Applied Genetics, 113, 575-583.
Salvesen, P.H., Kanz, B. (2009) Boxwood cultivars in old gardens in Norway. In: Morel, J.P. and
Mercuri, A.M. (eds.), Plants and Culture: seeds of the cultural heritage of Europe. Centro
Europeo per i Beni Culturali, Ravello.
Sanmartín, I. (2003) Dispersal vs. vicariance in the Mediterranean: Historical biogeography of the
Palearctic Pachydeminae (Coleoptera, Scarabaeoidea). Journal of Biogeography, 30, 1883-1897.
Stace, C.A. (1989) Dispersal versus vicariance - no contest! Journal of Biogeography, 16, 201-202.
Taberlet P., Gielly, L., Pautou, G., Bouvet, J. (1991) Universal primers for amplification of three
non-coding regions of chloroplast DNA. Plant Molecular Biology, 17, 1105-1109.
Taberlet, P., Fumagalli, L., Wust-Saucy, A.-G., Cosson, J.-F. (1998) Comparative phylogeography
and postglacial colonization routes in Europe. Molecular Ecology, 7, 453-464.
Thompson, J.D. (2005) Plant evolution in the Mediterranean. Oxford University Press, Oxford.
Thorne, R.F. (1972) Major disjunctions in the geographical ranges of seed plants. Quarterly
Review of Biology, 47, 365-411.
Valcárcel, V., Fiz, O., Vargas, P. (2003) Chloroplast and nuclear evidence for multiple origins of
polyploids and diploids of Hedera (Araliaceae) in the Mediterranean basin. Molecular
Phylogenetics and Evolution, 27, 1-20.
von Balthazar, M., Endress, P.K., Qiu. Y.-L. (2000) Phylogenetic relationships in Buxaceae based
on nuclear internal transcribed spacers and plastid ndhF sequences. International Journal of
Plant Sciences, 161, 785-792.
Walter, H., 1929. Die osmotischen Werte und die Kältenschäden unserer wintergrünen Pflanzen
während der Winterperiode. Berichte Der Deutschen Botanischen Gesellschaft, 47, 338-348.
Webb, T., Bartlein, P.J. (1992) Global changes during the last 3 million years: climatic controls
and biotic responses. Annual Review of Ecology and Systematics, 23, 141-173.
Wegmüller, S. (1984) Zur Ausbreitungsgeschichte von Buxus sempervirens L. im Spät- und
Postglazial in Süd- und Mitteleuropa. Dissertationes Botanicae 72, 333-344.
Wilcox, B.A., Murphy, D.D. (1985) Conservation Strategy: The Effects of Fragmentation on
Extinction. The American Naturalist, 125, 879-887.
63
Willis, K.J., Van Andel, T.H. (2004) Trees or no trees? The environments of central and eastern
Europe during the last glaciation. Quaternary Science Reviews, 23, 2369-2387.
Zanchetta, G., Borghini, G., Fallick, A.E., Bonadonna, F.P., Leone, G. (2007) Late Quaternary
palaeohydrology of Lake Pergusa (Sicily, southern Italy) as inferred by stable isotopes of
lacustrine carbonates. Journal of Paleolimnology, 38, 227-239.
64
65
CONCLUSIONS
In the past decades, a substantial amount of investigations were carried out to define the
distribution of Western Palaearctic woody species during the last glacial period. These studies had
two main practical purposes, that are possible suggestions for conservation strategies in view of
future climate changes (e.g. Willis and Birks 2006) and a better understanding of evolutionary
processes (e.g. Taberlet et al. 1998). In particular, the hypothesis that temperate trees may have
survived the last glaciation only in southern Europe (Bennett et al. 1991, Tzedakis et al. 2002) has
gradually shifted to the possibility of much wider distributions, including Central Europe (Willis et
al. 2000, Birks and Willis 2008) and even Northern Europe (Stewart and Lister 2001). Variations in
global ice volumes (Shackleton 1987) and fossil pollen records (Tzedakis et al. 2006) indicate that
the last glacial maxima has been one of the most severe of the whole Pleistocene. Thus, it is
reasonable to assume that the woody species surviving the last glaciation in central Europe may
have persisted during the other more or less equally harsh glacials (Magri 2010). If so, it is possible
that tree taxa present in Central Europe during the last glacial may have persisted over a large part
or even the whole of the Pleistocene. In this perspective, the effects of the last glaciation in shaping
the modern distribution of tree taxa and their genetic resources urge a reconsideration. As suggested
by Bennett and Provan (2008), the last glacial period should be considered as background noise
rather than as a signal when analyzing the origin and evolution of modern plant populations. On the
whole, these data advise caution in interpreting the modern distribution and genetic structure of tree
populations as the result of contraction and expansion events during the last glacial, and highlight
the need of a much longer perspective, which is still almost completely unexplored (Magri 2010).
In addition to the need of a much longer temporal perspective, there is one more point which
deserves a careful consideration. Geneticists have always been aware of the evolutionary effects of
historical population size (see Avise 2000 and references therein). However, the connection
between phylogeny and demography is usually based on genetic distances between individuals (e.g.
Rogers and Harpending 1992), which raises a problem for the risk of circular reasoning. Hence, it is
desirable to seek for alternative and independent data to infer historical population growths and
declines. The lack of fossil record at a given site for a tree taxa has been initially considered as a
proof of its absence in the surrounding areas. Comparison of pollen records from many sites
revealed that plant species show up diachronically, and not surprisingly this process was seen as a
spread from a site to another (e.g. Huntley and Birks 1983). Sparse, insignificant occurrences might
have reflected long-distance transportation of pollen rather than a taxon’s local presence. However,
attentive considerations lead to suggest that the appearance of a taxon at a given site may reflect a
population increase to a critical local density, above which a species is detectable by pollen analysis
(Bennett 1985). If the appearance of pollen records testify a population growth, pollen data may be
used as a proxy palaeodemographic information. Thorough reviews of the post-glacial literature
allow for a satisfactory collection of data that can be used to reconstruct the past demographic
histories of woody taxa. Comparing the record of a taxon from a considerable number of fossil sites,
overall spanning the whole post-glacial period and the taxon’s geographic distribution, allow to
detect patterns of population growth which may reflect the demographic history of a species in
space and time (Magri 2008). On these basis, it is possible to understand the varying modes and
rates of population growth and decline, showing different patterns from region to region and from a
period to another (Di Domenico et al. 2012). Also in this case, a much longer perspective is
desirable, because post-glacial population dynamics only reflect a minimal part of a taxon’s
demographic history. Nevertheless, the redundant behavior of tree taxa, witnessed by their cyclic
oscillations in long pollen records, suggest that some population dynamics may have been similar
during different stages of the Pleistocene. If so, the post-glacial patterns of population increase and
decline may be used to draw some conclusions on the modern genetic patterns and evolutionary
histories of tree taxa.
66
The present study is a first contribution to the knowledge of the genetic structure and a
deepening of the post-glacial history of Western Palaearctic Buxus. It also constitute an attempt to
depict a detailed distribution of Buxus. Despite the caveats associated to the amalgamation of
different distribution data from vegetation databases, Floras, herbarium specimens and field surveys,
this attempt provided some interesting information that were fundamental in the interpretation of
both fossil and genetic data of Buxus.
Referring to the aims proposed in the introduction, the following answers are given:

Which is the distribution of Buxus in the Western Palaearctic area? In a wide geographic
perspective, the genus Buxus show a distribution that is clearly linked with the major
Western Palaearctic mountains (Atlas, Baetics, Sistema Ibérico, Sistema Central,
Cantabrian, Pyrenees, Cévennes, Massif Central, Jura, Vosges, Alps, Apennines, Pindus,
Pontic, Taurus, Caucasus, Elburz). The distribution of Buxus is subdivided in two major
ranges, one in the central-western and one in the central-eastern areas of the Western
Palaearctic region. In the C-W range, Buxus is more frequent and the populations are
abundant and contiguous in Central Europe, whereas they are fragmented in the southern
territories. Buxus occurs in Western Mediterranean islands as well (Baleares, Sardinia,
Corsica). The C-E range of Buxus consist of several disconnected areas found in Greece
and the Middle East (Turkey, Syria, Caucasus, Azerbaijan, Iran).

Is the modern distribution of Buxus the result of a post-glacial migration from limited
glacial refugia? In addition to the sites located in the Mediterranean Peninsulas, fossil
records from the Jura Mountains between 30 and 12 ka BP document the location of
Buxus up to a latitude of at least 47◦N. The sparse early-Holocene occurrences (12-9 ka
BP) in central Europe confirm that Buxus may have persisted across most of its modern
distribution. This hypothesis is strongly supported by the plant traits of Buxus, which is a
poor pollen and seed disperser, and is able to withstand extremely low temperatures. In
any case, the modern distribution of Buxus in central Europe is not the effect of a
postglacial migration from southern refugia, where Buxus had only a sparse distribution
during the last glacial period.

Are there regional differences in the Holocene dynamics of Buxus populations in Europe?
The fossil data indicate that between 30 and 12 ka BP the populations located in the
Temperate bioclimatic zone were relatively dense. Their seventy-fold exponential
increase in the last 12.000 years resulted in a continuous modern distribution area. By
contrast, the Mediterranean bioclimatic zone hosted far less robust populations, which
increased only moderately during the first half of the Holocene and resulted in a modern
highly fragmented distribution area. The populations located in Minorca, Ibiza, Sardinia,
Sicily, Apulia, Basilicata, Croatia, Thessaly, W Greece, and Peloponnese have been
completely extirpated.

Which are the different roles played by natural population dynamics, climate change and
human impact in shaping the European distribution of Buxus in the last 15,000 years? In
the Temperate regions, the exponential population increase of the postglacial period
points to a typical biological process, not significantly affected by either climate or
human activity. In the Mediterranean regions, the natural population growth is limited to
the first half of the Holocene. Afterwards, an aridification process may be considered the
most likely cause for the population decline that fragmented the Buxus distribution,
mainly reducing it to moist habitats. There is little support for alternative causes (human
67
activities, fires, phytopathogens, and competition), which would have equally affected
the Buxus populations of both the Temperate and the Mediterranean areas.

Is Buxus a vulnerable genus in Europe? Most of the Buxus populations located in the
Temperate zone are vigorous, so that they do not require specific conservation actions.
However, the populations in the northern part of the Temperate range (England, NW
France, Belgium, Luxembourg, and Germany) are very fragmented, which indicates a
general weakness. These populations may not be vigorous enough to survive future
glacial reductions. In the Mediterranean bioclimatic area, Buxus populations are severely
fragmented and the undergoing reduction is expected to produce an even more dramatic
fragmentation. The combination of high genetic diversity in the south, especially of the
unique haplotypes that were detected in insular systems or in the southernmost territories
of the Mediterranean Peninsulas, coupled with the position of missing haplotypes in the
phylogenetic network, which are placed between lineages occurring in the Mediterranean
area, require careful consideration. It suggests that together with the disappearance of
many populations, a conspicuous portion the genetic diversity of Buxus may have been
lost in the last few thousand years. Thus, conservation actions of natural populations of
Buxus are of primary importance in the southernmost territories of the Mediterranean
Peninsulas.

How many Buxus species are found in the Western Palaearctic area? From molecular data,
it appears that only B. sempervirens is a distinct genetic species, whereas B. colchica and
B. hyrcana most likely represent locally adapted and isolated populations of
B. sempervirens. The genetic differentiation between B. balearica and B. longifolia is
difficult to interpret, as nuclear data suggest that this is a single-species complex with two
quite divergent, historically sundered groups of populations. However, the absence of
Tertiary and quaternary fossil of Buxus along N Africa questions the origins of this
disjunction.

Does the phylogeographic pattern of Buxus shows some commonalities with other woody
taxa of the Western Palaearctic area? As in many other woody taxa, Buxus show that all
the genetic diversity is constrained to the southern territories of Europe, whereas a single
haplotype occurs throughout Central Europe. However, the organization of plastidial
diversity is unusual and shows maximum levels of population differentiation and a lack
of intrapopulational variation. These genetic features seem to be a characteristic of the
genus Buxus and are linked to historical and ecological processes.

Does the molecular differentiation of Buxus mirrors the extreme disjunction of Western
and Eastern Mediterranean populations? The extreme disjunct distribution of Buxus in the
Mediterranean basin is reconciled by the lack of a phylogeographic break between the
Western and Eastern domains. Haplotype sharing between parapatric populations of
B. sempervirens and B. balearica and some morphological insights suggests that
hybridization has occurred during the history of Buxus in this area. This event has
occurred twice, the first to the east and a second in the west.

Does the comparison of genetic and palaeobotanic data provides new insights on the
historical causes responsible for the distribution of genealogical lineages of Buxus in the
Western Palaearctic area? Comparing the fossil and genetic data in the light of the
modern distribution of Buxus suggest that diversity occurs where the populations are
fragmented and experienced a severe reduction. By contrast, genetic evenness is detected
where the populations are continuous and experienced an exponential growth throughout
68
the Holocene. These findings question the role of Mediterranean refugia for Buxus and
indicate that genetic differentiation occurs where the populations are declining, whereas
genetic evenness characterize expanding populations.
69
REFERENCES
Avise, J.C. (2000) Phylogeography. The history and formation of species. Harvard University Press.
Bennett, K.D. (1985) The spread of Fagus grandifolia across eastern North America during the
last 18,000 years. Journal of Biogeography, 12, 147-164.
Bennett, K.D., Provan, J. (2008) What do we mean by ‘refugia’? Quaternary Science Reviews, 27,
2449-2455.
Bennett, K.D., Tzedakis, P.C., Willis, K.J. (1991) Quaternary refugia of north European trees.
Journal of Biogeography, 18, 103-115.
Birks, H.J.B., Willis, K.J. (2008) Alpines, trees, and refugia in Europe. Plant Ecology & Diversity,
1, 147-160.
Di Domenico, F., Lucchese, F., Magri, D. (2012) Buxus in Europe: Late Quaternary dynamics and
modern vulnerability. Perspectives in Plant Ecology, Evolution and Systematic, 14, 354-362.
Huntley, B., Birks, H.J.B. (1983) An atlas of past and present pollen maps for Europe: 0–13 000
years ago. Cambridge University Press, Cambridge.
Magri, D. (2008) Patterns of post-glacial spread and the extent of glacial refugia of European beech
(Fagus sylvatica). Journal of Biogeography, 35, 450-463.
Magri, D. (2010) Persistence of tree taxa in Europe and Quaternary climate changes. Quaternary
International, 219, 145-151.
Rogers, A.R., Harpending, H. (1992) Population growth makes waves in distribution of pairwise
genetic differences. Molecular Biology and Evolution, 9, 552-569.
Shackleton, N.J. (1987) Oxygen isotopes, ice volume and sea level. Quaternary Science Reviews, 6,
183-190.
Stewart, J.R., Lister, A.M. (2001) Cryptic northern refugia and the origins of the modern biota.
Trends in Ecology & Evolution, 16, 608-613.
Taberlet, P., Fumagalli, L., Wust-Saucy, A.-G., Cosson, J.-F. (1998) Comparative phylogeography
and postglacial colonization routes in Europe. Molecular Ecology, 7, 453-464.
Tzedakis, P.C., Hooghiemstra, H., Pälike, H. (2006) The last 1.35 million years at Tenaghi
Philippon: revised chronostratigraphy and long-term vegetation trends. Quaternary Science
Reviews, 25, 3416-3430.
Tzedakis, P.C., Lawson, I.T., Frogley, M.R., Hewitt, G.M., Preece, R.C. (2002) Buffered tree
population changes in a Quaternary refugium: evolutionary implications. Science, 297, 20442047.
Willis, K.J., Birks, H.J.B. (2006) What is natural? The need for a long-term perspective in
biodiversity conservation. Science, 314, 1261-1265.
70
Willis, K.J., Rudner, E., Sümegi, P. (2000) The full-glacial forests of central and southeastern
Europe. Quaternary Research, 53, 203-213.
71