ACTA PHYTOGEOGRAPHICA SUECICA 80 EDIDIT SVENSKA VAXTGEOGRAFISKA SALLSKAPET

Martin Diekmann

Deciduous forest vegetation in Boreo-nemoral Scandinavia

UPPSALA 1994

ACTA PHYTOGEOGRAPHICA SUECICA 80 EDIDIT SVENSKA VAXTGEOGRAFISKA SALLSKAPET

Martin Diekmann

Deciduous forest vegetation in Boreo-nemoral Scandinavia

OPULUS PRESS AB UPPSALA 1994 Doctoral thesis at Uppsala University

ISBN 91-721 0-080-X (paperback) ISBN 91-7210-480-5 (cloth) ISSN 0084-5914

Editor: Erik Sj ogren

Editorial Board:

A. W .H. Darnman, Storrs, CT F.J.A. Daniels, MUnster L. Ericson, Umea D. Glenn-Lewin, Ames, lA 0. Hamann, Copenhagen H. Sj ors, Uppsala H. Trass, Tartu

Technical Editor: Marijke van der Maarel-Versluys

© Martin Diekmann 1994

Cover illustration: Reno Lottmann

Edidit: Svenska VaxtgeografiskaSallskapet Villavagen 14, S-752 36 Uppsala

DTP: OPULUS PRESS AB Printed in Sweden, 1994 by Eklundshof Grafiska AB, U ppsala

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 3

Abstract. Martin Diekrnann. 1994. Deciduous forest vegetation The underlying environmental factors for the vegetational in Boreo-nemoral Scandinavia - Acta Phytogeogr. Suec. 80, differentiation on the forest type level were a complex-gradient Uppsala. 112 pp. ISBN 91-7210-080-X. (91-7210-480-5). in nutrient status, connected with variation in light conditions, and a moisture gradient. On the community level, mainly geo­ This study aimed at an investigation of the vegetation ecology of graphic-climatic factors were operating, in particular an East­ deciduous hardwood forests in the Boreo-nemoral zone of Scan­ West and a humidity gradient. Anthropogenic factors were less dinavia. Different community types were described with regard important. to their species composition and structure, differentiation, geo­ Ellenberg's central-European indicator values for light, graphic distribution, ecological conditions and dynamics, as moisture, nitrogen and reaction proved to be useful in character­ well as their affinities to community types in other vegetation izing the Scandinavian forests, although some species seem to zones in northern and central Europe. occur under environmental conditions which differ from those Field work was conducted in S Sweden and SE Norway. In in central Europe. The Boreo-nemoral forests are as species-rich total 367 releves were made. In each stand, structural character­ as the Nemoral forests; the absence of several southerly distrib­ istics as well as cover-abundance values of all vascular plants uted taxa is 'compensated' for by comparatively higher frequen­ and bryophytes were recorded. The releves were analyzed by cies of other species. Community types on soils with an interme­ means of cluster analysis (program TWINSPAN) and ordination diate to moderately high fertility have the highest species rich­ (program Correspondence Analysis from the program package ness. CANOCO). Vegetation-environment relationships were analyzed Most stands have been influenced strongly by human activ­ by correlating the explanatory variables with releve scores on ity and are compositionally unstable. In eutrophic elm-ash for­ the ordination axes. ests, Quercus spp. and Fraxinus excelsior were better repre­ In 69 stands of two areas (Oland and the mainland of eastern sented with larger size classes, whereas Acer platanoides and Sweden), tree size was analyzed in order to reveal the population Ulmus spp. had higher frequencies in smaller size classes. It is structure of the most important tree species in the main forest suggested that this patterncan be interpreted in terms of a future types. Successional trends were inferred from the relative densi­ replacement series: in a succession following termination of ties of species in different DBH (diameter at breast height) human impact, Quercus and Fraxinus will be partly replaced by classes. Ulmus and Acer. The population structure in mesotrophic for­ Environmental studies were carried out in 17 stands of ests revealed that Quercus will decrease, whereas Ulmus and mesotrophic and eutrophic forests in the vicinity of Uppsala, Fraxinus and, in particular, Acer and Tilia, will increase. In including inclination, heat index, maximum temperature, mini­ oligotrophic forests, however, Quercus will maintain a domi­ mum temperature, temperature range, evaporation, wind expo­ nant position, while Picea may increase in importance. The tree sure, canopy cover, light, dry bulk density, organic matter, species dynamics will also lead to increasing canopy closure moisture, pH(H20), pH(KCl) and nitrogen. Canonical Corre­ and, subsequently, compositional changes in the other vegeta­ spondence Analysis (CANOCO program package) and correla­ tion layers. tion analysis were used to reveal the relationships between The environmental studies in the selected forest stands re­ vegetation and environment. vealed that pH (together with nutrients) and light were the Four forest types were described, with in total nine commu­ primary factors related to vegetational variation in the field and nities: (1) oligotrophic oak forests: Quercus petraea-Frangula bottom layers. Other properties of the soil and the climate of the alnus community, Quercus robur-Betula pendula community; forest as well as physiographic factors, were of minor signifi­ (2) mesotrophic mixed deciduous fo rests: Quercus robur-Tilia cance. Mesotrophic forests had significantly lower soil reaction, cordata community, Quercus robur-Euonymus europaeus com­ higher light intensities and higher maximum temperatures than munity, Quercus robur-Fraxinus excelsior community; (3) eutrophic forests. They also tended to have a lower soil mois­ eutrophic elm-ash forests: Ulmus glabra-Fraxinus excelsior ture, a wider temperature range, lower minimum temperatures community, Ulmus minor-Fraxinus excelsior community; (4) and a higher evaporation. eutrophic alder-ash forests: Fraxinus excelsior-Prunus padus As a whole, deciduous hardwood forests in Boreo-nemoral community, Fraxinus excelsior-Alnus glutinosa community. Scandinavia have become rare and threatened, particularly the Mesotrophic forests represent the most widespread and charac­ eutrophic forests. Existing forests, especially those with a long, teristic forest type of the Boreo-nemoral zone, without floristically uninterrupted history, should therefore be maintained and pro­ similar counterparts in the Nemoral zone. tected.

Martin Diekmann, Department of Ecological Botany, Uppsala University, Villaviigen 14, S-752 36 Uppsala, Sweden.

Acta Phytogeogr. Suec. 80 Contents

1 Introduction 5 1.1 Aims 5 1.2 Nature conservation aspects 5 1.3 Deciduous hardwood forests in northern Europe: a brief literature review 6 1.4 Vegetation and forest history 8

2 Study area 12 2.1 The Boreo-nemoral zone and its delimitation 12 2.2 Geology and Soils 13 2.3 Climate 15

3 Analysis of vegetation and vegetation-environment relations 18 3.1 General 18 3.2 Sampling procedure 18 3.3 Data treatment 18 3.4 Tables 20 3.5 Nomenclature of fo rest communities 20 3.6 Nomenclature of species and comments on difficult taxa 20

4 Forest communities 22 4.1 Introduction: Survey of clusters 22 4.2 Oligotrophic oak fo rests 23 4.3 Mesotrophic mixed deciduous fo rests 32 4.4 Eutrophic elm-ash fo rests 48 4.5 Eutrophic alder-ash forests 55 4.6 Syntaxonomy 64

5 The population structure of trees and fo rest dynamics 69 5.1 Introduction 69 5.2 Methods 70 5.3 Results 71 5.4 Discussion 74

6 Environmental studies 79 6. 1 Introduction 79 6.2 Soil analysis and climatic measurements 80 6.3 Results 82 6.4 Discussion 89

7 General discussion and conclusions 95 7. 1 Species 95 7.2 Communities and environment 98 7.3 Geographic distribution of communities and the Boreo-nemoral zone 10 1 7.4 Structure and dynamics 102 7.5 Implications for nature conservation 103

8 Acknowledgements 104

9 References 105 1 Introduction

1.1 Aims composition in order to reveal the primary causes for community differentiation. The Boreo-nemoral commu­ This study aims at a floristicand ecological description of nities will also be discussed with respect to their central deciduous forest communities in the Boreo-nemoral zone and western European counterparts. of Scandinavia. Field investigations have mainly been carried out in Sweden, partly for practical reasons but also because the deciduous forests in Sweden are less well 1.2 Nature conservation aspects known than their Norwegian counterparts. The study deals with forests of a type called 'adellovskog' in Swedish Climatic deterioration as well as millennia of cultivation (broad-leaved hardwood forest), composed of Acer and exploitation have diminished the area of deciduous platanoides, Fraxinus excelsior, Quercus spp., Tilia forest in Scandinavia to a small fraction of its prehistoric cordata and Ulmus spp. The term 'adel' ('noble') refers to size. The majority of sites suitable for deciduous forest are the species' relatively high demands for nutrients and now covered by arable land and spruce plantations. The temperature, as well as their greater economic importance remaining stands have been influenced and changed by and higher status, compared with 'trivial' species such as land use practises such as logging, pollarding and grazing, Betula spp. and Populus tremula. Also Fagus sylvatica earlier also thinning for haymaking in wooded meadows. and Carpinus betulus are often treated as 'noble' tree Deciduous forests seem to be common in many parts of species. However, forests composed of these two species southernand central Sweden. However, this impression is have their main distribution in the Nemoral zone and form delusive, since they usually are confinedto open areas or only minor stands within the study area. Beech forests fringes around cultivated land, lake-shores and hillsides, will therefore not be dealt with, and particularly since they and in the vicinity of old roads and villages. Only occa­ have been described in detail by Lindquist (1931, 1932), sionally, stands of deciduous forest are surrounded by an Lindgren (e.g. 1970, 1975) and recently by T. 0kland extensive area of coniferous forest. (1988). Neither will young successional forests with e.g. The changes in agriculture and forestry (planting of Betula spp. be considered, or wet swamp forests with conifers and decreasing importance of grazing in forests Alnus glutinosa, which have a different species composi­ and former wooded meadows) since the end of the last tion. The study thus comprises forests which phytosocio­ century have resulted in a slight increase of the deciduous logically can be arranged in the class Querco-Fagetea. forest area. Many of the stands originating from older Since the study only deals with fairly natural forests with times have become mature and economically attractive a more or less closed canopy, the term 'forest' will be used for selective logging or complete deforestation, the more throughout instead of 'woodland', which is a term with a so as the value of timber from broad-leaved tree species wider meaning, including both forests and parks, wooded has increased. This has only rarely, as yet, led to replant­ meadows, wooded pastures, etc. ing with broad-leaved trees. During the course of the More specifically, the study foremost aims at a classi­ study, the author has witnessed the destruction of many fication of the deciduous forest communities in Scandina­ stands. In the Mittlandsskogen areaon Gland, a large part via. They will be described with respect to their species of the mature forest has been completely or partly cut composition, structure, geographic distribution, differen­ down during only five years. tiation and ecological conditions. Ordination in connec­ Motives for the preservation of deciduous forests are tion with correlation analysis will be used to reveal the manifold, e.g. their high floristic and faunistic diversity, importance of physiographic and macroclimatic param­ as compared with conifer forests (especially plantations), eters for community differentiation. Species indicator the occurrence of rare plant and animal species bound to values will be applied to evaluate the importance of edaphic deciduous forests, their importance for recreation and, in and climatic factors for species responses. Emphasis is some cases, as monuments of former land use and cultural also placed on the successional trends of the main com­ history. munities, derived from the population structure of tree species. Measurements of soil and climatic parameters, carried out in a selection of forest stands belonging to different forest communities, will be related to species

Acta Phytogeogr. Suec. 80 6 M. Diekmann

1.3 Deciduous hardwood fo rests in north­ Several studies cover fairly large areas, treating the (forest) vegetation of a parish or larger district. In Swe­ em Europe: a brief literature review den, Ivarsson (1962) and Wallin (1973) described the deciduous vegetation in parts of BohusHi.nand Vastergot­ Among the first to call attention to the deciduous forest land, respectively. Both give information on community vegetation in Sweden was Linnaeus (1745) who, in his differentiation, forest history and successional aspects. book on the journeyto bland and Gotland in 1741, wrote Sjogren (1961, 1964) described the different forest types as follows: and bryophyte communities of deciduous forests on bland. " ...The road passed through the most beautiful groves Olsson (1974, 1975) studied the south Swedish sand one could ever see, which for beauty surpassed all other vegetation, including oak forests on acid soils. Here, the places in Sweden and competed with all in Europe; they classification was accompanied by extensive soil analyses. consisted of linden, hazel and oak, with the ground smooth In Finland, literature on deciduous forest in the Boreo­ and green, without rocks or moss ... " ( transl. by Asberg & nemoral zone is almost lacking, probably due to the Steam 1973). restricted extension of the zone and the rareness of this vegetation type. An exception is the monograph on eutrophic deciduous woods in the SW archipelago by General description Hinneri (1972). However, many descriptions of wooded Until the beginning of the 20th century, botanical interest meadows exist, e.g. by Palmgren (1915-1917), Cedercreutz was mainly focused on plant geography and systematics. (1927) and Hregstrom (1983). Several publications from The increasing importance of industrial forestry led to a Finland deal with forest communities of more northern growing interest in forest research, which, however, mainly but adjacent areas belonging to the southern Boreal zone, aimed at a high productivity of certain favoured tree including mixed thermophilous forests with Tilia cordata species, i.e. Picea abies and Pinus sylvestris. One of the andAlnus-richforests, e.g. Tapio (1953), Koponen (1967) first detailed ecological descriptions of deciduous vegeta­ and Makirinta (1968). Compared with Sweden, the de­ tion in the Boreo-nemoral zone was the study of wooded ciduous forest vegetation in Norway is much better known. meadows on Aland by Palmgren (1915-1917). A mono­ Especially since the beginning of the seventies, many graph on the floraand vegetation in Vpp land by Almquist studies from different parts of Norway have been pub­ (1929) contains a classification of deciduous forests that lished, often using the Braun-Blanquet approach, e.g. A. is mainly based on structural characteristics and domi­ Bj�rnstad (1971), Aune (1973), Fremstad (1979) and nances of species. During the following decades, several 0vstedal (1985). Kielland-Lund (1981) made a compre­ books and papers on deciduous forests were published. hensive study of the forest communities in SE Norway. However, a few attempts were made to give a comprehen­ For each community, information is given on the syntaxo­ sive account of these forests, e.g. by Selander (1955). At nomy, general species composition, differentiation, eco­ least for Boreo-nemoral Sweden, deciduous forests were logical conditions, dynamics and chorological aspects. rarely dealt with in detail, in contrast to other vegetation Other publications treat the widely distributed, Alnus types of a much wider distribution, such as mires. incana-rich forests on river banks, e.g. Fremstad & Deciduous forests are of a limited extension, but often 0vstedal (1978) and Klokk (1980, 1982), having counter­ very striking in the coniferous or cultivated landscape. In parts in Sweden and Finland mainly in the Boreal zone, many publications, single forest stands have been de­ but to some extent also in the western partsof the Boreo­ scribed, particularly from some provinces in southeastern nemoral zone. Sweden, for example bland (e.g. Du Rietz 1917; Sterner 1926; 0. Johansson 1982 and K. Larsson 1982) and Classificationsystems and nomenclature of forest com­ Sodermanland (e.g. Halden 1950; Ryberg 1956, 1971). In munities Norway, Skogen (1971) made an ecological and plant geographic study of the world's northernmost oak forest. Vegetation science in Fennoscandia has not followed one In many areas, inventories of valuable forest stands in particular approach, either with respect to methodology, connection with nature conservation projects were carried or in syntaxonomical classification(Trass & Malmer 1973; out, e.g. by Sjogren et al. (1974) and Ekstam (1979) on Malmer 1974). Swedishphytosociology, and the 'Uppsala bland, and Korsmo (e.g. 1974) in Norway. These publi­ School' in particular, is more a tradition than a fixed cations often contain forest classification systems for School. There is no generally accepted system of hierar­ mapping purposes. With regard to nature conservation chically structured vegetation units, as exists in central aspects, the floraand vegetation of deciduous forests have Europe where the Braun-Blanquet approach is used by been treated by e.g. Bergendorff et al. (1979), IngelOg practically all vegetation scientists. Deciduous forest com­ (1981), Anon. (1982), Almgren et al. (1984) and IngelOg munities in Sweden have been described using different et al. (1984). methods and nomenclatural systems, e.g. by Julin (1948),

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 7

Ivarsson (1962), Sjogren (1964), Wallin (1973) andKlotzli either particular forest types or limited areas. A compre­ (1975a). Several Norwegian authors have followed or hensive treatment of deciduous forests in Sweden, using adapted the Braun-Blanquet approach, e.g. Bj�rnstad multivariate methods, has not been carried out so far. (1971) and Kielland-Lund (1981). Only a few attempts have been made to work out a Productivity classification for the whole of Scandinavia or the Boreo­ nemoral zone, respectively. Kielland-Lund (1971, 1973) Apart fromclassification and description of communities presented the first schemes in connection with the Inter­ and their relationship with environmental factors, the national Biological Programme. Each community is char­ productivity of deciduous forest species has been ad­ acterized by a short general description and a species list, dressed. In the framework of the International Biological including dominant and differential species. Vevle (1983) Progr e, Hyttebom (1975) and H. Persson (1975) amm published a preliminary survey of higher syntaxa for investigated the above-ground woody production and field­ Norway, and a synopsis of Norwegian forest communities layer and below-ground production, respectively, of a was recently given by Kielland-Lund (1994). Classifica­ deciduous woodland in the province of Uppland. The tions were also presented by Bergendorff et al. (1979) and latter study also presents many phenological data on the Anon. (1984). As the most important criterion for the seasonal development of herbs and grasses. distinction of forest types, the dominance of tree species has been used. Both authors give valuable lists of syno­ Population biology and forest dynamics nyms of communities from the North European literature. None of the systems mentioned is based on a comprehen­ The population ecology of deciduous forest species has sive tabular comparison of releves from different regions. not gained much attention up to now. An exception is Klotzli (1975a) proposed a classification derived from formed by the thorough demographic study of theperen­ such a comparison, but his system is based on a rather nial herbs Hepatica nobilis and Sanicula europaea by small number of releves. Comparisons of central Euro­ Inghe T (1985), where the authors demonstrate the & amm pean and Scandinavian forest communities have mainly importance of weather conditions, particularly summer dealt with conifer forests (Matuszkiewicz 1962; Kielland­ drought, for flowering and mortality of the investigated Lund 1967 and Neuhausl 1969). However, North Euro­ species. Among other studied herbs are Primula veris pean wet alder forests have been included in a monograph (Tamm 1972), Fragaria moschata (Ryberg 1986) and on Alnus incana-rich communities in Europe by Schwabe Lathyrusvemus (Ehrlen 1992). Brunet (1994) studied the (1985). demography of the woodland grasses Festuca altissima, Hordelymus europaeus, Bromusramosus and B. benekenii. With respect to deciduous tree species, Fraxinus excelsior Multivariate analysis was treated by Hulden (1941) and Tapper (1992), and Modem multivariate techniques, such as cluster analysis Quercus robur by C. Andersson (1994). Much more is and ordination, have only exceptionally been applied to known about the population biology, as well as popula­ Fennoscandian deciduous forest vegetation. One of the tion genetics, of conifers such as Picea abies and Pinus first studies was performed by Ammar (1978) on the sylvestris. vegetation (and its underlying environmental factors) of With respect to forest dynamics (both natural and shore ridges on bland. Rlihling& Tyler (1986) and Brunet anthropogenic), attention has foremost been given to three (e.g. 1991) use cluster analysis for the description and different subjects: firstly, the acidification of forest soils comparison of deciduous forest communities in southern­ and its impact on species composition; secondly, the most parts of Sweden. In Finland, Makirinta (1990) and succession of deciduous forests, particularly concerning Tonteri et al. (1990) have used both classification and woody species; thirdly, the impacts of climatic change on ordination techniques for their analyses of forests in the forest ecosystems. southern part of the country. Their studies include a The negative consequences of soil acidification due to comparison of modem techniques with the traditional acid precipitation have first been described for central Finnish approach, i.e. Cajander's forest site type classifi­ Europe, but have recently become a major topic in Swe­ cation. Cluster analysis and ordination were also applied den as well. Research has concentrated on S and SW by Heikkinen (1991) in order to analyse the esker vegeta­ Sweden where the acidification is most pronounced due tion in an area in S Finland. His study shows the impor­ to a comparatively high amount of rainfall. The first tance of physiographic and microclimatic factors for the results of these long-term changes of soil acidity and its differentiation of forest communities. In Norway, fern­ impact on soil nutrient contents and species composition rich vegetation has been analyzed with the aid of multi­ have been presented by Hallbacken T (1986), & amm variate methods by Odland et al. (1990) and Odland Falkengren-Grerup (e.g. 1986, 1987), T Hallbacken amm& (1991, 1992). Thus, most studies have concentrated on (1988) and O.N. Bj�rnstad (1991). Norden (1992)

Acta Phytogeogr. Suec. 80 8 M. Diekmann

compared the impact of different deciduous forest species sition of these forests, it is necessary to elaborate the on soil acidificationand element fluxes in Skane. forest history of the study area up to the present. Much of As most deciduous forests in Scandinavia are in a the information given in this chapter is based on the highly dynamic state, successional trends and the poten­ comprehensive studies of the vegetation and forest his­ tial 'climax' vegetation have been addressed by many tory by Fries (1965) and Huntley (1988), and of the authors (e.g. Lindquist 1938 and Ivarsson 1962). How­ cultural history and former land use by Sjobeck (1931, ever, the discussions are seldom based on field observa­ 1933), Selander (1955) and Ekstam et al. (1984). tions. Ryberg (1971) studied the diameter and age distri­ bution of tree species with regard to forest history and Vegetation history succession. Through repeated analysis of permanent plots established between 1925 and 1935, long-term changes in After the last glaciation, trees started to re-invade north­ the composition of the tree layer could be demonstrated ern Europe from the south during the early Post-glacial for Dalby Soderskog in Skane, situated in the Nemoral time (since about 8000 B.C.). Among the first woody zone (Lindquist 1938; Malmer et al. 1978; S. Persson species to appear were Betula spp., Populus tremula, 1980). This forest stand has been heavily disturbed during Pinus sylvestrisand Corylus avellana. Otherthermophilous the past and is in a successional state towards a more species than Corylus, such as Alnus glutinosa, Ulmus natural species composition. In a study of Vardsatra Na­ glabra, Fraxinus excelsior, Quercus robur and Tilia ture Reserve near Uppsala, Hytteborn (1986) compared cordata, first immigrated at the beginning of the Late densities and basal areas of tree species between 1912 Boreal (ea. 6800 B.C.) due to a climatic amelioration. (measured by R. Sernander) and 1985. The importance of During the Atlantic time (Older Stone Age, culmination forest history and former land use for the structure and about 4000 B.C.), a period of climatic optimum, rich species composition of a deciduous woodland on bland deciduous forests covered extensive areas of Sweden. were emphasized by Ekstam & Sjogren (1973). Julin These so-called mixed oak forests (Quercetum mixtum) ( 1948) treated the ecology of wooded meadows after occurred further to the north at that time, compared with abandonment. The roles of particular species in forest presentday distribution. Evidence of this is found in pol­ succession were discussed for Alnus glutinosa by Fremstad len spectra, fossil nuts of Corylus avellana, and isolated ( 1983) and for both Fraxinus excelsior andAlnus glutinosa present occurrences of Corylus, Tilia and Ulmus glabra by Tapper (1992). Andersson (1994) studied the patterns ssp. montana in Norrland. In southern Sweden, less de­ of seedling emergence and early survival of Acer plata­ manding trees such as Betula and Pinus were probably noides, Quercus robur and Tilia cordata at a site in a restricted to either dry and shallow or very wet soils. The transitional state between open grassland and deciduous early Sub-boreal time (since about 3000 B.C.) is charac­ forest. However, general trends of deciduous forest suc­ terized by a deterioration of the climate and an increasing cession in the Boreo-nemoral zone, as well as trends of human impact on the vegetation. As a result, the area of particular species or forest communities, are still insuffi­ deciduous forest decreased again in the whole of Sweden. ciently known. Due to the great but variable longevity of At the beginning of the Sub-atlantic time (transition from most tree species, predicting the future from present trends Bronze to Iron Age, ea. 500 B.C.), the climate deterio­ is an uncertain procedure. rated once more towards cooler and moister conditions Successional trends as described above may be super­ and caused a further decrease of the Quercetum mixtum. imposed by long-term vegetation dynamics due to cli­ The present northern distribution of, e.g. Tilia cordata, matic changes, induced by an increasing carbon dioxide reflects a relict status from the mid-Holocene. Besides, level and subsequent global warming. Modelling of the new forest tree species immigrated to northern Europe, responses of Nordic forests to these climatic changes Pieea abies from the northeast and Fagus sylvatica anJi indicates that many tree species eventually may expand Carpinus betulus from the south. Pinus and Picea gradu­ (Fagus sylvatica) or reduce (Picea abies) their distribu­ ally became the most important constituents of North tion areas, and that their competitive· relations may change European forests, apart from Fagus and Quercus in the (Sykes & Prentice 1993). As a consequence, the Boreo­ southwest. nemoral zone in Sweden would 'move' northwards over a considerable distance. Cultural history

During pre-agricultural time, human impact on the North 1.4 Vegetation and forest history European forest landscape was probably negligible. Set­ tlements of the Younger Stone Age and Bronze Age were All forests and forest sites have in some way been influ­ concentrated to areas in S Sweden that were naturally rich enced and changed by the activity of man. In order to in deciduous forests. Farmingplayed a less important role understand and interpret the structure and species compo- during these times than cattle-breeding, and the people

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 9

Fig. 1. Wooded meadow, spring aspect. Scattered trees and shrubs alternatewith open areas. Conspicuous are a large individual of Quercus robur and several Corylus avellana bushes. In the foreground, a traditional fire place can be seen, where the litter and small branches of the previous year have been burnt. Gotland, May 1989. Photo M. Diekmann.

burnt the woodland extensively in order to create clear­ alternatedwith open meadow areas without trees (Fig. 1). ings for pastures and later hay-meadows. Losses of nutri­ The maintenance of wooded meadows was very time- and ents and minerals due to constant grazing and burning energy-demanding. Work consisted in removing and burn­ caused a significant deterioration of the soil, except on the ing litter, cutting of branches and twigs, collecting and most fertile morainic tills and in wetlands which were drying their leaves, mowing of the grass and herbs and naturally manured by periodical flooding. During the removing invading trees and shrubs. Following the mow­ Middle Ages and later periods, the forest area decreased ing, cattle were usually allowed to graze during late continuously, caused by an increasing population and its summer and autumn. All this implied a constant loss of high demands for cultivated land, timber and firewood. In plant material which was not compensated for by fertiliz­ parts of northernEurope, forests had vanished completely ing. In the beginning, however, wooded meadows did not already before Modem times, for example in the west suffer so much from an impoverishment of the soil, be­ Norwegian coastland (Behre 1988). When the climate cause trees with their extensive root systems permanently turnedcooler and wetter at the beginning of the Iron Age, transferred nutrients from deeper soil layers to the ground. supply of winter fodder for cattle became increasingly They were bound to old settled areas and therefore con­ difficult. This period gave rise to the partitioning of farm­ centrated to S Sweden and the provinces along the Baltic land into different structural and functional units, and the Sea, from northern Skane and Blekinge up to Uppland. creationo,of hay-meadows. However, wooded meadows could also be found in cen­ The former agrarian landscape was divided into 'inago­ tral and northern Sweden, for example in Dalarna, where mark', the infield, infenced land, and 'utagomark' , the trivial broad-leaved trees (e.g. Betula spp., Populus outfield, unfenced land. The latter comprised nutrient­ tremula, Sorbus aucuparia, Salix spp. and Alnus incana) poor heath and woodland and was used as common pas­ replaced the more demanding, thermophilous species (Sjors ture. The infield land was made up of the rich soils with 1954 ). The extension of wooded meadows culminated arable land and hay-meadows. The meadows provided during the 18th century. From this time on, but particu­ winter fodder for the cattle, which in turn provided ma­ larly during the 19th century, changes in agricultural nure for the fields. The harvest of grasses and herbs was economy and implements entailed a major change in the difficult because tools for cutting, such as sickles and agrarian and forest landscape. scythes, first became available during the course of the In general, the relocation into larger blocks of farm­ Iron Age. Instead, the farmers cut and broke off branches land, resulting from three land reform events ('storskifte', of trees of deciduous species, in order to dry the leaves. 'enskifte' and 'lagaskifte'), led to a more effective agri­ Thus, probably about 1500 years ago, the first wooded culture. The former outfield land was now private-owned meadows were created. They represented a mosaic of and subjected to exploitation, especially from the copper­ different vegetation types: small groves or groups of trees and ironworks which required great quantities of char-

Acta Phytogeogr. Suec. 80 10 M. Diekmann

coal. Due to milder forest laws and an increasing demand Most important for the supply of winter fodder in the for timber, the total forest area decreased rapidly in many wooded meadow was Fraxinus excelsior. Its ability to provinces during the 19th century, e.g. on bland (Daniels­ resprout after repeated cutting of branches and twigs son 1918). Parts of BohusHin were almost completely made it very popular, not only in northern Europe, but also deforested in 1850, as a result of the high demand for in central Europe (Pettersson 1958; Ellenberg 1986). Pol­ timber, smallwood and fuel for agriculture, and for mak­ larded ash trees can still be seen as remnants in forests and ing herring oil (Fries 1958). Also in Halland, the forest meadows in many parts of the continent (Behre 1988). area reached its minimum in the middle of the 19th Hence, the species is nowadays often found on the former century (Malmstrom 1939). At the same time, grazing infield land in the neighbourhood of old villages and increased considerably, and many closed forests were farms, especially in Blekinge and Smclland. On the other turned into wooded pastures. In this stage, more or less hand, Fraxinus was greatly reduced on the former outfield natural, primary forests had become quite rare and were land due to its sensitivity to grazing. restricted to special sites. Forests on steep slopes, e.g. at Ulmus glabra was not particularly favoured as a edges of the plateau hills, were less accessible to grazing wooded meadow tree. However, elm was often planted animals and unsuitable for cultivation, and probably only close to large estates and castles. According to Lindquist subjected to selective logging. Woodland belonging to (1932), many of the planted trees belonged to introduced big estates or the Crown was sometimes preserved for provenances of southern origin. Ulmus minor could be hunting reasons. found in wooded meadows on bland and Gotland, and its After the introduction of more effective ploughs and ability to resprout as root suckers made it a fast colonizer artificial fertilizers, arable lands, including leys and ferti­ on abandoned pastures (Pettersson 1958). lized meadows, extended their areas at the expense of Another selected tree species was Tilia cordata. It wooded meadows, which suffered from deterioration of provided leaf fodder, bast fibers for ropes, and valuable soil and were either cut down and turned into arable land timber which was easy to process (Anon. 1982). Linden and pastures, or allowed to regrow into forest. Many of the played an important role in areas of bee-keeping and bast present deciduous forests originate from former wooded production. However, on Gotland it was repressed through meadows, often through an intervening stage of wooded human influence (Pettersson 1958). Besides, it generally pastures. The rapid decrease of the wooded meadow area suffered from grazing in the former outfield land, but can be illustrated by some figures from Gotland, where often survived among large boulders. about 32 000 ha in the year 1900 had become reduced to Quercus robur and Q. petraea were the tree species only 284 ha in 1983. that were considered the most desirable, both in northern At the turn of the century, grazing in forests and and central Europe (Ellenberg 1986; Behre 1988). They wooded pastures decreased due to a more intensified were usually the only species to be allowed to grow tall. agriculture (Anon. 1982). The need for timber decreased Oak hardly provided any leaf fodder, but raising of pigs too, as other materials and energy sources became more was very dependent on a good supply of acorns. For this important. Many pastures, wooded pastures and meadows purpose, also beech (Fagus sylvatica) was allowed to turned into forest again, and the deciduous forest ex­ form large stands on the estates in the south. Besides, both tended its area considerably. This extension has been oak and beech gave valuable timber, and oak bark could most pronounced in central bland which was almost be used for tannery. Land owners kept oak forests for the devoid of trees in the middle of the 18th century (Daniels­ purpose of hunting, and the Crown preserved and even son 1918), but now carries the largest continuous area of planted oak to ensure a supply of timber for warships. deciduous forest in S Sweden, Mittlandsskogen. How­ Already in medieval times (14th century), a law was ever, large areas suitable for deciduous forest were planted enacted against the cutting of oak and hazel. A decree with conifers, mainly spruce. This planting activity started from 1569 prohibited the cutting of oak, hazel and apple already in the 19th century (e.g. Malmstrom 1939; Fries on bland (Danielsson 1918). Due to its low demand for 1958). nutrients, Quercus probably also had a good chance of survival in pastures of the outfield land with its deterio­ rated soils. However, grazing must have prevented regen­ Tree species composition of wooded meadows eration of oak. As many deciduous forests have arisen from wooded Even other 'fruit-bearing' species such as Malus meadows, the management of the latter must be given sylvestris and Corylus avellana (in the south locally also some consideration. Certain trees and shrubs of high Fagus sylvatica) were selected. Hazel gave valuable wood 'quality' and usefulness were selected and favoured, oth­ and nutrient-rich leaves for winter fodder. ers repressed or eliminated. As a consequence, the natural Acer platanoides was neither selected for in wooded frequencies of important tree species of the Boreo-nemoral meadows, nor needed as timber or fire wood. On Gotland, zone were altered. human activity eliminated it from major parts of the island

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 11

(Pettersson 1958). However, it was popular as an orna­ species do not suffer in the same way, as several species mental tree close to farms and villages. are thorny or unpalatable, e.g. Rosa spp., Crataegus spp., Conifers, especially Picea, were usually considered Prunus spinosa and Juniperus communis. A high fre­ undesirable in the wooded meadows and were removed quency of these species in deciduous forests may indicate (Hregstrom 1983). former grazing. Summarizing these effects of management, Quercus A moderate grazing intensity often results in a higher robur, Fraxinus excelsior and Corylus avellana (locally species diversity due to a more open, lighter forest struc­ also Tilia cordata) were the most favoured species. In ture and the constant removal of large, competitive species. contrast, Betula spp., Ulmus spp. and Acer platanoides Releves in a grazed, open oak-ash forest on bland had were not favoured, and conifer trees were usually species numbers up to 84 species/200 m2, including mosses disfavoured. This selection of species has had a great (Diekmann 1988). On the other hand, intensive grazing impact on the composition of deciduous forests until pressure causes a floristicimpoverishment. In general, the recent times (Ekstam & Sjogren 1973). number of herbs decreases, whereas grasses and certain unpalatable herbs increase. With respect to structure, the field layer becomes lower and denser. As grazing pressure Effects of grazing increases, animals behave less selectively. Forests and wooded pastures of the former outfield land In the long run, continuous grazing necessarily results have been grazed by cattle, pigs, horses, goats and sheep in a deterioration of the soil, even if droppings and urine, for many centuries. After the land reforms in the first half in limited spots, may cause a much higher fertility. It may of the 19th century, also many parts of the infield land also change the morphology and density of the soil (Steen suffered from a hard grazing pressure. The effects of this 1958). grazing have been numerous, depending on its intensity, Many centuries of intensive land use in the form of duration and character. logging, mowing and grazing have altered forest struc­ Foraging of acorns by pigs had a largely positive ture, species composition and soil conditions. Apart from impact on oak and beech forest. Trampling might have the effects described above, the geographical isolation caused some damage, but greatly facilitated regeneration and historical discontinuity of present deciduous forests of trees due to uprooting of the soil. Grazing, e.g. by are important factors in terms of species richness. Studies cattle, had a strong direct impact on the vegetation. Even in several parts of Europe have shown that continuous, a moderate grazing intensity can prevent regeneration of ancient forests have a higher species richness than discon­ trees (Steen 1958), as shown for, e.g. Quercus robur tinuous, recent forests, and that isolation causes a de­ (Brenner 1921) and Fraxinus excelsior(Hulden 1941). As crease in species (e.g. Peterken & Game 1984; Dzwonko a consequence, in the long run, and with some logging, the & Loster 1990; Brunet 1993). Thus, compared with natu­ tree layer becomes more open, and the closed forest turns ral woodland of pre-historical time, many deciduous for­ into a wooded pasture. Eventually, trees may totally dis­ ests in northernEurope may have suffered from a floristic appear, as on the island Stora Karlso close to Gotland after impoverishment of native forest species. centuries of intensive sheep grazing (Froman 1946). Shrub

Acta Phytogeogr. Suec. 80 2 Study area

2. 1 The Boreo-nemoral zone and its de­ A much larger area of northern Europe, compared limitation with the Nemoral zone, belongs to the Boreo-nemoral zone (Hemiboreal zone according to Ahti et al. 1968). The In the same way as any extensive area of the earth, boundary between these two zones is usually defined as northern Europe can be divided into different vegetation the natural southwesterndistribution limit of Pieea abies. zones. These are primarily defined according to prevail­ Because of the intensive planting of this species, as men­ ing types of vegetation. The criteria for the division and tioned previously, this limit cannot be accurately recon­ delimitation of such zones fall into three categories, structed any longer (Sjors 1965). Due to the scarcity of bioclimatic, edaphic-topographic and botanical (Ahti et Picea on southernmost bland this part of the island is al. 1968). Macroclimate is of crucial importance as an usually included in the Nemoral zone. The higher el­ underlying factor. The climatic approach normally uses evated plateau in the province of Smaland ( 'sydsvenska thermal values, e.g. mean temperature, length of growing hoglandet') is treated as a southern outlier of the Boreal season, and temperature sums (Tuhkanen 1984). Edaphic­ topographic divisions are based on soil types and/or the general topography of an area. A botanical division makes use of distribution patternsof single species or the occur­ rence and dominance of certain plant communities. Apart from the criteria mentioned above, a division of zones can also be based on ecological criteria such as phenological data or productivity. Besides, even cultural and agricul­ tural features are usually specific for certain vegetation zones. Attempts using different criteria often come to similar results. A vegetation zone can be further divided into vegetation sections based on characteristic features caused by the oceanity/continentality of the climate (Ahti et al. 1968). Vegetation belts designate altitudinally sepa­ rated regions. Several systems of vegetation zones in northern Eu­ rope have been proposed by, e.g. Zoller (1956), Hustich (1960), Sjors (1963) and Ahti et al. (1968). A detailed map of the vegetation regions in Norway was published by Moen (1987). The vegetation zones of the Nordic countries according to Sjors are shown in Fig. 2. Rich deciduous forests with tree species of the genera Quercus, Acer, Tilia, Ulmus and Fraxinus are confined to the south­ em- and westernmost parts of northern Europe. The Nemoral zone (Temperate zone according to Ahti et al. 1968) embraces Skane, the western parts of Halland and BohusHin, SE Blekinge and S bland as well as parts of the south coast of Norway. The forests of this zone are almost exclusively deciduous. Beech (Fagus sylvatica) is the dominating tree species on mesic soils throughout the Fig. 2. Vegetation zones in northern Europe. 1. Arctic zone; 2. zone except on bland and in the coastal parts of northern Alpine belt; 3-6. Boreal zone (3. Sub-alpine birch woodland BohusHi.n as well as most of the Norwegian part. How­ belt; 4. Sub-arctic and Boreo-montane sub-zone; 5. Main Boreal ever, on very dry or wet soils other broad-leaved species zone; 6. Southern Boreal sub-zone); 7. Boreo-nemoral zone; 8. play a more prominent role. Norway spruce (Picea abies) Nemoral zone; 9. Western broad-leaved and pine forest region originally was absent from the Nemoral zone, but can (North Atlantic pine-birch woodland and heath region). Accord­ nowadays be found almost everywhere due to planting. ing to Sjors (1963), from Anon. (1977), revised.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 13

zone by Ahti et al. (1968). According to Ahti et al. (1968), the southern part of this The Boreo-nemoral zone is dominated by conifer for­ area belongs to the Nemoral zone, and the northernpart to ests consisting of Picea abies and Pinus sylvestris. De­ the Boreo-nemoral zone. Moen (1987) distinguished a ciduous forests depend on favourable climatic and/or Coastal section as a major phytogeographical unit of its edaphic conditions and have a more restricted distribu­ own, characterized by a general scarcity of forest due to tion. Nowadays, deciduous forests are more or less un­ logging and subsequent burning, grazing and hay-mak­ common in most parts of the Boreo-nemoral zone because ing. Sjors (1963), however, defines it as a separate North man has turnedthem into agricultural land. Most Nemoral Atlantic pine-birch woodland and heath region where tree species are still represented, namely Quercus robur, both Picea abies and Fagus sylvatica are nearly absent. Ulmus glabra, Fraxinus excelsior, Acer platanoides and Here Pinus sylvestris and some Nemoral broad-leaved Tilia cordata. Only Acer campestre and Tilia platyphyllos tree species play an important role. Forests of this region are absent, and Carpinus betulus is restricted to a narrow will therefore be included in the comparison of forest fringe close to the Nemoral zone. Fagus sylvatica forms vegetation from the Boreo-nemoral zone. stands in the southern and southwestern parts, with scat­ tered occurrences further to the north. Quercus petraea has a western distribution, whereas Ulmus minor only 2.2 Geology and Soils occurs on the Baltic islands of Oland and Gotland. Ulmus laevis is confined to Oland and southernFinland. Topography and Geomorphology The northern boundary of the Boreo-nemoral zone is usually definedas the distribution limit of Quereus robur. In S Sweden, altitude does not vary much. Most areas, It is well known as the 'limes norrlandicus'. In Sweden, particularly in the east, lie below 100 m a.s.l. The interior this border line can be drawn from just north of the lower parts of SmiHand, the so-called south Swedish Highland, course of the river DaHilven, through central Vastmanland only reach about 300 m, but they function as a barrier for and Narke and north of Lake Vanern to the Norwegian the predominant westerly winds, resulting in a rain-shadow border. The fairly abrupt change in the floristic and situation in SE Sweden. Some areas in Vastergotland and vegetational composition corresponds to a similar abrupt bstergotland have altitudes exceeding 250 m, the most change of both topographic, edaphic and climatic condi­ conspicuous being Omberg (263 m) at the eastern side of tions and, subsequently, of land use (Fries 1948; Fransson Lake Vattern and the plateau-like stratified hills in 1965). In the Boreal zone, many Nemoral tree species are Vastergotland, rising from the surrounding lowland, e.g. totally absent or restricted to a narrow zone in the south. Billingen (299 m), Mosseberg (327 m) and Kinnekulle The reason for this has often been considered to be the (306 m). Whereas SW Finland is rather level throughout length of the growing season (or temperature sum) which the study area, S Norway is mountainous, and only the decreases with increasing latitude. Generally, this favours coastal areas and deep inland valleys lie below 500 m. the 'evergreen-strategy' of conifers (Schroder 1983). As to the relative relief, large areas of the Boreo­ Nemoral outposts in the form of small stands of ash, nemoral zone in S Sweden can be described as plains or linden, maple and hazel are often bound to south-facing joint-valley landscape (Rudberg 1987). The central parts slopes with a favourable local climate and soil chemistry. form an undulating hilly landscape of varying altitude. The western race of elm (Ulmus glabra ssp. montana) extends almost to the polar circle in Sweden and even Pre-Quaternary bedrock farther in Norway. Apart from Sweden, the coastal region of southemmost S Sweden, like northern Europe as a whole east of the Finland, including Aland, belongs to the Boreo-nemoral Scandes, is characterized by predominantly very ancient zone. Even here, the distribution limit of Quercus robur bedrock, part of the extensive Baltic Shield. Precambrian forms the border line to the adjacent zone, again due to a crystalline rocks, mainly granites and gneisses, cover a fairly abrupt change of topographic and climatic condi­ major part of the area (Rudberg 1987). These are rather tions (Jalas 1957). On the other side of the Baltic sea, the acid and resistant to weathering. Basic (alkaline) crystal­ Boreo-nemoral zone includes the Baltic states and adja­ line rocks appear at only a few places, e.g. close to cent regions mainly to the east. Almunge in the province of Uppland. The Boreo-nemoral zone comprises also SE Norway, The Precambrian stratum forms the basement for lo­ namely the Oslofiord region as far north as Lake Mjfbsa. calized Palaeozoic sediments of varying thickness, mainly According to Moen (1987), also some small areas along Cambrian sandstone and alum shale, Ordovician lime­ the western fiords as far north as the Trondheim region stone and Silurian shales. These soft, often calcareous, belong to this zone. The phytogeographical position of the rocks are preserved only in a few regions, with their hyperatlantic southernand southwesterncoast of Norway largest extension in the province of Jamtland (Fig. 3). Due up to the Trondheim region has been much discussed. to their physical and chemical properties, weathering

Acta Phytogeogr. Suec. 80 14 M. Diekmann

- bland and Gotland, where the layers form the visible portion of a continuous sequence in the Baltic region; � Rocks rich in lime - some parts of Skane. � Soils rich in lime Jotnian and Carnbrian sandstones occur in some areas, 1 1 1 Transport direction of I I disintegrated rocl

Quaternary deposits

Because Scandinavia has been a land area from the late Silurian up to the present, it has a long history of erosion and denudation (Rudberg 1987). Of major importance are \ the repeated glaciations during the Pleistocene which . .·· �- . resulted in a total (or almost total) coverage by ice. Glacial P:lt. ·-- and glaciofluvial erosion had a great geomorphological � ..: impact on the form of the landscape. After the last glaciation, most of the present land surface of Sweden was covered by till, mostly rich in boulders, glacial or marine clay and coarse-grained glaciofluvial sediments. These deposits vary as to type and distribution, depending on the form of the ice sheet and its movement, type of deglaciation, etc. A large part of S Sweden was under water during the retreat of the ice sheet. When the land emerged from the sea, the glacial material was sorted and partly transported due to wave action (Sjors 1965a). As a consequence, the till cover has been removed from hills and exposed sites and re-depos­ ited in lower, sheltered areas. It fills depressions and softens the irregularities of the basal bedrock. Among the l·, apparent surface features are radial and terminal moraines lOO 200 km as well as drumlins (G. Lundqvist 1959). Originating from the areas with Palaeozoic sediments, calcareous material has been dispersed southwards by glacial drift Fig. 3. Distribution of calcareous rocks and soils in Sweden. into large areas of Vastergotland, Smaland, bstergotland Arrows show the dispersal of disintegrated limestone by glacial and Narke (Fig. 3, Magnusson et al. 1963). Calcareous drift. From Magnusson et al. (1963), revised. soils were also transported to eastern Uppland and Sodermanland, originating from the Ordovician bedrock on the bottom of the Bothnian Sea (Sjors 1965a). How­ ever, large parts of the area are no longer covered by any results in favourable soils suitable for growth of decidu­ Quaternary deposits, but show exposed bedrock. This is ous forest. Within the study area, Carnbro-Silurian rocks particularly apparent in Bohuslan on the Swedish west can be found in: coast, in places on the east coast and also on the Baltic - bstergotland (between Lake Vattem and Lake Roxen); islands of bland and Gotland (G. Lundqvist 1959). - Vastergotland (Billingen, hills of Falbygden, Halleberg, Apart from glacial drift, we find glaciofluvial deposits Hunneberg, Kinnekulle and Lugm1s hill; sedimen­ transported by melt water streams which are usually built tary rocks often covered by Permian dolerite sills); up of coarse-grained material. The eskers, e.g. the Uppsala - Narke (area west of Lake Hjalmaren); esker, are quite striking features on the flat, cultivated

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 15

plains. Particularly in the lower elevated parts of the study area, fine-grained sediments have been deposited due to the influence of both standing (fresh or brackish) and running water. Aeolian sands, both early Post-glacial and recent, are only locally frequent and, within southern Sweden, concentrated to the province of Vastergotland and the Baltic islands of bland and Gotland (G. Lundqvist 1959). Peatlands in S Sweden only locally cover large areas, e.g. in Vastergotland and Sm:lland.

Soils

In S Sweden, brown earths play an equally important role as podsols (Troedsson & Nykvist 1973). They are more or less restricted to areas with alkaline bedrock and/or cal­ careous Quaternary deposits. The vegetation formerly connected with brown earth was deciduous forest, but the brown forest soils have been transformed into agricultural land to a great extent. As to the humus form, podsols usually are combined with mor and brown earths with mull, respectively.

Growing season 2.3 Climate (days)

400 km Scandinavia enjoys a higher annual temperature than any other area at a comparable latitude. With its location at the western edge of the big Eurasian landmass, it is influ­ enced by warm westerly and south-westerly winds and Fig. 4. Length of the growing season (threshold +5 °C) in the Gulf Stream. Due to the proximity of the Atlantic Sweden, Finland and northernNorway. FromTuhkanen (1980). Ocean and Baltic Sea, the climate is fairly maritime, and the winters are relatively mild.

the values for Blekinge and bland vary between - 1.0 oc Temperature and - 2.0 °C, they drop off to - 4.0 oc to - 5.0 oc in the Within the study area, the annual mean temperature is inner parts of Sm:lland and the northern provinces of the mainly a function of latitude. Besides, increasing altitude Boreo-nemoral zone, e.g. Uppland. The mild winters of as well as increasing distance to the sea or extensive the coastal areas of S Sweden (and W Norway) are of inland waters result in decreasing annual mean tempera­ great importance for the vegetation and allow frost-sensi­ tures (Eriksson 1982). The values vary between 5.0 °C tive species to extend their distribution far to the north and 7.0 °C, compared with 7.0 oc to 8.0 oc in southemmost (Sjors 1965a; Hulten 1971). Sweden. In the surroundingsof the big lakes (Lake Vanern, The risk of frost during the growth period of plants is Lake Vattern and Lake Malaren), the annual mean tem­ also of botanical importance. It is particularly low in perature is comparatively high. In contrast, the interior coastal areas and in the surroundings of the big lakes, both parts of Smaland show comparatively low values (down in spring and autumn (Angstrom 1974). to 4.8 °C). Despite the great latitudinal extension of the area, the Growing season mean temperatures of the warmest month (July) are fairly equal, ranging from about 17 oc to 15 °C, with the lower The growing season is defined as the time period with a values at higher altitudes. The value for Uppsala (16.5 °C) mean diurnal temperature above a certain threshold (usu­ is about the same as for most climate stations in Skane, ally 5 °C). Fig. 4 shows the length of the growing season situated 600 km to the south (Eriksson 1982). in Sweden and Finland according to Tuhkanen (1980). The mean temperature of the coldest month (usually The isoclines correspond well with the borders of the February) shows a much higher variability. In general, it main vegetation zones. The southemmost parts of Swe­ decreases with latitude and distance to the sea. Whereas den, belonging to the Nemoral zone, enjoy the longest

Acta Phytogeogr. Suec. 80 16 M. Diekmann

de Martonne's index

70

56

l:r2500

Annual precipitation

Fig. 5. Annual precipitation in mm in 1931-1960 in northern Europe. From Nordseth (1987), revised. growing season, i.e. Skane and the southern parts of the Swedish west and east coast. The Baltic islands of bland and Gotland show comparatively high values. The 180- days isocline is well in accordance with the distribution limit of Que reus robur both in Sweden and Finland, and, hence, with the borderline of the Boreo-nemoral zone. The interior elevated parts of Smaland have a growing season shorter than 180 days which supports the decision 0 100 200 km of Ahti et al. (1968) to include this region in the Boreal zone. The use of temperature sum (with a 5 oc threshold) instead of growing season reveals a very similar isocline pattern (Tuhkanen 1980). Fig. 6. de Martonne's index in Sweden, applied to the growing season only (based on a 3 oc threshold). From Angstrom (1974), revised. Precipitation

Due to the prevailing wind directions and orientation of to 400-500 (600) at the Swedish east coast and the mm the Scandes, northern Europe shows a strong West-East Baltic islands of bland and Gotland. There is an appreci­ gradient in precipitation. Whereas the annual amount of able increase with altitude. On the other hand, areas in the rain in western Norway exceeds 2000 it may fall vicinity of the big lakes show comparatively low values. mm, below 400 in areas situated in the rain-shadow of With respect to the seasonal distribution of rain, maxi­ mm mountains, e.g. in valleys of south-central Norway and in mum values almost always occur in late summer (July and Finnmark (Fig. 5, Nordseth 1987). In accordance with this August). A second maximum is shown in autumnby some gradient, the highest values within the study area are stations at the west coast, due to the increasing frequency shown by the westernprovinces of Sweden. The average of south-westerly winds during this season. The lowest annual precipitation amounts to 900-1100 in parts of values are generally found in late winter/early spring mm Halland, BohusHin, Smciland and Vastergotland, but only (Angstrom 1974).

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 17

Snow keeps the temperature at ground close to the freezing point, even when the air temperature is much below 0 °C. This is of great importance for plant growth. The average duration of snow cover varies between ea. 40 days in the southernmost parts of Sweden and ea. 120 days at the northern border of the Boreo-nemoral zone. However, the actual duration fluctuates greatly from year to year.

Climatic indices

Moisture conditions In order to characterize the relative climatic humidity or aridity of an area, one usually relates precipitation and temperature (or evaporation) to each other in a formula (Tuhkanen 1980). One of the most widespread indices is de Martonne's 'indice d'aridite' (better called humidity index because values increase with humidity) which has the form: H P/(T + 1 0), where P is the annual precipita­ = tion and T the annual mean temperature in °C (de Martonne 1926). In a slightly different form, the index can also be calculated for shorter time periods. Fig. 6 (Angstrom Continentality index 1974) gives de Martonne' s index based on the growing season for Sweden. It is in quite good accordance with the distribution of rain as given above, with the highest values 400 km in SW Sweden and extreme low values in a narrow band along the Swedish east coast and on bland and Gotland. The surroundings of the big lakes show comparatively low humidity, and the increase with altitude is even clearer Fig. 7. Conrad's continentality index in Sweden, Finland and than for precipitation alone. The moisture conditions can northern Norway. From Tuhkanen (1980). be graphically displayed in the climatic diagrammes of Waiter (Waiter & Lieth 1960- 1967).

Continentality/maritimity Continentality indices can be based on both thermal and of southern Sweden and increase towards the interior and hygric properties of the climate. Often we make use of the towards the north (Fig. 7, Tuhkanen 1980). A similar annual temperature range, corrected for the effect of lati­ pattern of relative continentality and maritimity results tude (Tuhkanen 1980). In Conrad' s continentality index from a survey of temperature anomalies, a measure of the deviation of A and annual mean temperature from the [C = 1,7 X A/sin (% + 10 °C)], A denotes the annual temperature range and % the latitude (Conrad 1946). The mean for the latitude in question (Angstrom 1974). W higher the value of C, the more continental is the climate. Norway has an extreme oceanic climate with high pre­ Within Sweden, C-values are lowest in the coastal areas cipitation and a small annual temperature range.

Acta Phytogeogr. Suec. 80 3 Analysis of vegetation and vegetation-environment relations

3.1 General one plot was laid out, usually in its centre to avoid mar­ ginal effects. The localization of deciduous forest stands suitable for Plots were usually of a square form and had a size of 2 investigation was based on literature, inventories for na­ usually 15 x 15 m , which is within the recommended ture conservation purposes, topographic and vegetation limits for the minimal area of temperate deciduous forests maps and personal communications. Where no informa­ (Ellenberg 1956; Westhoff & van der Maarel 1973). Ex­ tion was available, some reconnaissance was carried out. ceptionally, however, a rectangular form and a somewhat Efforts were concentrated on regions known to have a smaller size had to be accepted, e.g. in narrow forest comparatively high proportion of deciduous forest, i.e. stands on slopes and along rivers. Because of the strong the coastal parts of SE and SW Sweden, bland and floristic seasonality, each plot had to be visited twice: Gotland, the plateau-like hills in Vastergotland, the east­ once in spring to sample the early-fading spring geophytes, em side of Lake Vattem in Smaland and Ostergotland as the second time in summer to carry out the main vegeta­ well as the surroundings of Lake Malaren. tion analysis. For each plot were noted: location, plot size, Vegetation analyses were only carried out when the aspect and inclination, as well as humus form. For the forest stand or part of it fulfilled some basic criteria: description of structure, the stand was divided into differ­ - predominance of hardwood species in the canopy ent strata: (Acer, Carpinus, Fraxinus, Quercus, Tilia, Ulmus). - tree layer (T), including all tree species taller than 5 m, Fagus forests and forests dominated by Betula spp. [when appropriate, split into upper (T1) and lower tree and Populus tremula were not included (see Chapter layer (T2)]; 1); - shrub layer (S), including all woody species with a 2 - a minimum size of about 1000 m in order to avoid height between 0.5 and 5 m (Corylus avellana was edge effects; consistently treated as part of the shrub layer though it - a minimum height of the upper tree layer of at least locally reached 10 m height); 20 m, with the exception of naturally low growing oak - fieldlayer (F), including all non-woody vascular plants, forests on very acid soils; as well as woody species not taller than 0.5 m; - a certain 'naturalness' which implies absence or rare­ - bottom layer (B), including all bryophytes and lichens. ness of exotic tree species; For the vascular plant layers, the cover was estimated - the absence of any signs of recent logging and other in 10-% intervals, whereas the cover of the bryophytes heavy disturbances (i.e. during the last 20-30 years); often was low and, when lower than 10 %, was estimated - the absence ofgrazing by domestic animals for at least on a 1-% scale. During the examination of the forest some years. structure, signs of former human impact were noted, such as pollarding, logging, planting, grazing and draining. For each layer, a species list was made. Epiphytic 3.2 Sampling procedure species were not included. For estimation of the cover degree and/or abundance of species, the combined 7- Sampling followed the Braun-Blanquet approach (Braun­ degree cover/abundance scale was used (see Braun­ Blanquet 1964; Westhoff & van der Maarel 1973; Blanquet 1964; Westhoff & van der Maarel 1973). Dierschke 1994). To begin with, a general survey (recon­ naissance) of a forest was made in order to get information on habitat and vegetation differentiation within the forest. 3. 3 Data treatment Relative homogeneity of structure and species composi­ tion and uniformity of habitat were necessary conditions All 367 releves were arranged in a table, separated into the for establishing a plot (Mueller-Dombois & Ellenberg different layers mentioned above. Tree species may thus 1974). Larger stands often included different forest com­ appear up to four times (T1, T2, S, F) in a table. Before munities: here, several analyses were made representing starting data treatment, cover-abundance values were trans­ these different communities. Within a more or less homo­ formed into the 1-9 ordinal scale according to van der geneous part of the stand, the plot was chosen at random. Maarel (1979). In stands just exceeding the required minimum size, only

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 19

Classification of Ellenberg (Ellenberg et al. 1991) were used for the factors light (L), temperature (T), continentality (C); soil Clusters were obtained with the program TWINSP AN moisture (M), soil reaction (R) and soil nitrogen (N). An (Hill 1979). Usually, the default options of the program indicator value on a 1 to 9 scale ( 1 to 12 for moisture) were chosen. As a 7 -degree scale had been used for cover indicates the ecological optimum (not the physiological estimation, seven appropriate pseudospecies cut levels optimum) of a species with regard to a particular factor. were used, resulting in a 1-7 scale in the final tables: For example, a species score of 1 for light means that this 1 one or a few individuals; species is mainly found in deep shade, a score of9 implies 2 occasional and < 1 % cover; that it is mainly found in completely open, light-exposed 3 1-5 % cover; 4 6-25 % cover; habitats. Indifferent or insufficiently known species are 5 26-50 %; disregarded and not assigned an indicator value.

6 51-75 % cover; The value of an environmental variable at a site (releve) 7 > 75 % cover. is estimated through the method of weighted averaging Values for 'Minimum group size of division' and (Jongman et al. 1987). The weighted average (WA)for a 'Maximum level of division' were determined depending site is the average of all indicator values of those species on the total number of releves included in the analysis. In which are present at this site, calculated as: a few cases, obvious outliers were removed from clusters. In order to be able to reproduce single releves, the TWINSPAN output table was split into a few small tables, each containing the releves of one or several clusters. where y., y 2, ..., y m are the species responses at the site, In order to get a general comparison of different forest and ul u2, , urnare the species indicator values. When ' .•. communities, a synoptic table was constructed. Taxa are presence/absence values are used as species responses, represented with their presence degree (Westhof f & van WA is simply the mean of the indicator values. If species der Maarel 1973), calculated in percent as the number of cover/abundance values are taken into account, indicator releves in a cluster in which a taxon occurs, divided by the values are weighted in proportion to species abundance. total number of releves in that cluster. In species-rich vegetation (as deciduous forests), the two Releves of similar forest types found in the literature ways of calculating W A lead to similar results (Backer et have been used for comparison in the following way: On al. 1983). In this study, weighted averages are based on the basis of the synoptic table mentioned above, literature presence/absence, as recommended by Kowarik & Seidling releves were arranged together with similar ones in a new (1989), and woody species are considered only once and table. Because literature data were based on different not for each layer. In the following, weighted averages for methods with respect to, e.g. plot size, separation of layers releves will be named INDICATOR FIGURES, e.g: light and cover degree scale, these new tables had to be modi­ (L-) figures or moisture (M-) figures, whereas the term fied: All cover-abundance values were transformed into 'indicator values' refers to species only. presence/absence values. Moreover, all tree and shrub species are listed only once, without differentiation of Correlation analysis strata. When literature data did not contain information on the floristic compositionof the bottom layer, bryophytes The relationship between sample plot on an ordination and lichens had to be excluded from further analyses. axis and explanatory variables (i.e. physiographic vari­ TWINSPAN analyses of these tables were carried out in ables, climatic variables and indicator figures) was ascer­ the same way as described above. tained by means of correlation analysis. The strength of the correlation expresses the probability that a variable is part of the gradient underlying the axis (R.H. 0kland Ordination 1990). As the variables often did not show a normal For the ordination of species-releve tables, the program distribution, the non-parametric Spearman rank correla­ Correspondence Analysis (CA), part of the program pack­ tion test was applied (Siegel & Castellan 1988) .. Since the age CANOCO (ter Braak 1987, 1990), was used. CA indicator figures were based on presence/absence of spe­ constructs eigenvectors as ordination axes which repre­ cies, the explanatory variables were correlated with sam­ sent the floristic variation. Usually, the default options of ple scores on CA-ordination axes which were also based the programs were chosen. on presence/absence data (which are somewhat different from the axes of the CA-ordination diagrams based on cover/abundance data). Indicator values Apart from indicator figures, the following macro­ In order to estimate the importance of climatic and edaphic climatic (geographic) and physiographic variables were variables for species composition, species indicator values used (abbreviations used in tables and main sources given

Acta Phytogeogr. Suec. 80 20 M. Diekmann

in brackets): separately for the different strata, because (see Westhoff longitude (LON) and latitude (LAT); & van der Maarel 1973): annual mean temperature (TEM YEAR), as well as all strata are rooting in a common substratum; mean temperatures of the warmest month (TEM JULY) there are strong ecological interactions between the and coldest month (TEM FEB) (Eriksson 1982; Walter different strata; & Lieth 1960-1967); plants of all strata start their life in the ground layer length of the growing season, defined as the period and are then subjected to its ecological conditions May-August (GRO PER) (Walter & Lieth 1960- 1 967); (which may, however, have been quite different annual precipitation (PRE YEAR) and precipitation from present conditions). during the growing season (PRE VEG) (Eriksson b. All units are arranged in a hierarchical system. The 1983; Waiter & Lieth 1960-1967); basic units are called 'community', a non-committal term. de Martonne's index (MAR IND) and Conrad's A community name is composed of two names of woody continentality index (CON IND), calculated with the species, the first one being a tree species, the second one aid of the variables given above (see Chapter 2.3); being a tree or shrub species. The sub-units of first and aspect (ASP), converted to a discrete scale in the second order are called sub-community and form. For the following way: 2-SW; 1-S and W; 0-SE and NW; denomination of these sub-units, names of differential -1 - E and N; -2: NE; species (not necessarily the most frequent or abundant inclination (INCL); species) were used. The term 'Typical' denotes sub-units heat index (HEAT IND): cos (aspect - 225)·tan without their own differential species and which, in gen­ (inclination), according to K.C. Parker (1988). eral, are often relatively poor in species (so-called 'inops'­ The variables chosen for use, depended on the avail­ units, cf. Westhoff & van der Maarel 1973). The nomen­ able data and the forest type in question. clatural rank of a unit is not necessarily congruent with the cluster level of the TWINSP AN analysis.

3.4 Tables 3.6 Nomenclature of species and Releve tables are organized in the following way: The comments on difficult taxa table head shows the cluster codes, the cover percentages of different layers and the total number of species (N­ The nomenclature of vascular plant species follows Flora TOT), as well as the number of vascular plants (N-VAS) Europaea (Tutin et al. 1964- 1980). However, some devia­ and bryophytes (N-BRY). In general, also the location of tions and modifications were applied, mainly because of the sample plot is indicated using the following abbrevia­ identification problems of non-flowering individuals: tions: BL - Blekinge, BO - BohusHin, HA - Halland, NA ­ Narke, NO - Norway, OG - Ostergotland, OL - bland, SM - Alchemilla: species of the A. vulgaris group were not - Smaland, SO - Sodermanland, UP - Uppland, V A - differentiated and are given as Alchemilla vulgaris agg. Vastergotland, VM - Vastmanland. Aspect and inclina­ - Carex muricata, C. sp icata and C. divulsa were not tion of the sample plot are also included. differentiated and recorded as Carex muricata agg. The different layers (T l, T2, S, F, B) are presented - Crataegus laevigata, C. monogyna and C. calycina separately in the tables. Within each layer, the differential show a great variation and hybridize with each other; species for the sub-units are given first, arranged in blocks, probably most individuals in deciduous forests belong to followed by all other species arranged in decreasing fre­ Crataegus laevigata. Given as Crataegus spp. quency. Low-frequent species are listed at the end of the - Dactylorhiza maculata and D. fuchsiicould not always table. In the text description, species of frequencies be­ be separated and are therefore sometimes referred to as tween 60 and 100 % are referred to as constants, those of Dactylorhiza maculata agg. frequencies between 40 and 60 % as common, of frequen­ - Heracleum sphondylium ssp. sphondylium and ssp. cies between 20 and 40 % as occasional, and of frequen­ sibiricum were not separated. cies up to 20 % as scarce (following the terminology of -Malus sylvestris and M. domestica (the domestic form Rodwell l991). which often runs wild) could not be separated with cer­ tainty and are given as Malus sylvestris. - Ranunculus: species of the R. auricomus group are 3.5 Nomenclature of forest communities treated collectively as Ranunculus auricomus agg., with the exception of Ranunculus cassubicus. In this study, the following concepts and criteria were - Rosa: the species of this genus could often not be applied: identified; particularly the separation of R. canina and a. Forest communities are classified as a whole, and not R. dumalis caused problems. Given as Rosa spp.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 21

- Rubus: species of the R. fr uticosus group were not Nomenclature for cryptogams follows Corley et al. ( 1981) differentiated and are treated collectively as Rubus fr uti­ for mosses and Grolle (1983) for hepatics. When young cosus agg. specimens of Eurhynchium hians and E. praelongum - Stellaria nemo rum is not distinguished in its two subspe­ could not be referred to either of these species, they were cies ssp. nemorum and ssp. glochidisperma. treated as Eurhynchium praelongum agg. - Ta raxacum of the section Ta raxacum is given as It might be added that the terms 'thermophilous', Ta raxacum off icinale agg. 'nitrophilous' and 'acidophilous' are applied to species - Valeriana offi cina/is ssp. offi cina/is and ssp. sambucifo lia which are confined to sites with high average tempera­ are treated as separate species (V. of ficina/is and V. tures and nutrient-rich and acid soils, respectively, with­ sambucifo lia ). out inferring that these species are physiologically adapted - Viola reichenbachiana and V. riviniana could not al­ to these conditions. ways be distinguished and were then recorded as Viola spp.

Acta Phytogeogr. Suec. 80 4 Forest communities

4.1 Introduction: Survey of clusters ofbryophytes is usually high. Typical elements of decidu­ ous forests are quite rare. Instead, species common also in The total number of species recorded in the 367 releves is conifer forests occur in high frequencies. The average 321: 256 vascular plants, 64 bryophytes and only 1 lichen. indicator figures point at very low pH and nitrogen con­ Table 1 shows the results of the community classification tents of the soil, as well as at a fairly open canopy of the based on the TWINSPAN cluster analysis: 18 community tree layer. In many respects, oligotrophic oak forests are types have been distinguished. Four main forest types can deviant from other deciduous forests. They are mainly be identified,for which the number of vascular plants and found in the coastal provinces of the Boreo-nemoral zone. bryophytes, the total number of species and the indicator figures are given in Table 2. Both average values and the Mesotrophic mixed deciduous forests (clusters 5-13, range of values of the different sub-units are listed. n = 197)

Mesotrophic mixed deciduous forests are composed of Oligotrophic oak forests (clusters 1-4, n 33) = different deciduous tree species, often with Tilia cordata Oligotrophic oak forests are invariably dominated by as a characteristic component. However, Quercus robur either Quereus robur or Q. petraea, whereas other decidu­ is usually the dominating species. In general, the number ous hardwood trees are less frequent. The number of of vascular plants is higher than in other deciduous forest vascular plants varies considerably, whereas the number types; the number of bryophytes varies. The indicator

Table 1. Survey of 18 community types, derived from the TWINSPAN analysis of a total of 367 releves. Numbers in the second column refer to the TWINSPAN code. For abbreviations of regions, see Chapter 3.

Cluster Cluster Community No. of Region(s) code releves

Oligotrophic oak forests Quereus robur-Betula pendula comm. 1 000 Viola riviniana sub-comm. 15 OL,SM,BL 2 001 Trientalis europaea sub-comm. 5 OL,BL Quercus petraea-Frangula alnus comm. 3 010 Viola riviniana sub-comm. 3 BO,HA 4 01 1 Trientalis europaea sub-comm. 10 HA

Mesotrophic mixed deciduous forests 5 10000 Quereus robur-Fraxinus excelsior comm. 30 GO Quercus robur-Euonymus europaeus comm. Tilia cordata sub-comm. 6 1000 100 Filipendula ulmaria form 20 OL 7 1000101 Stellaria holostea form 26 OL Oxalis acetosella sub-comm. 8 1000110 Corylus avellanaform 19 OL 9 1000111 Carpinus betulus form 10 OL Que reus robur- Tilia cordata comm. Geranium sylvaticum sub-comm. 10 100100 Geranium robertianum form 17 VA.,UP,NO 11 1001010 Deschampsia flexuosa form 19 UP,SO,SM,V A. ,NO 12 1001011 Lathyrus vernus form 28 UP,SO, VA. ,bG 13 1001 1 Stellaria holostea sub-comm. 28 SM,BL,BO

Eutrophic elm-ash forests Ulmus glabra-Fraxinus excelsior comm. 14 10100 Allium ursinum sub-comm. 22 V A,bG,NO,NA,HA,BO,SM 15 10101 Gagea lutea sub-comm. 43 UP,OG,SO,VA. ,NO,SM,VM 16 1011 Ulmus minor-Fraxinus excelsior comm. 25 OL,SM

Eutrophic alder-ash forests 17 110 Fraxinus excelsior-Prunus padus comm. 28 NO,V A.,SO,BO,UP,NA.,SM O A 18 Ill Fraxinus excelsior-Alnus glutinosa comm. 19 BO,S ,HA,NO,V

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 23

Table 2. Number of vascular plants (V as), number of bryophytes (Bry) and total number of species (Tot), as well as indicator figures for light (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for the four deciduous forest types. The average and the range of values for the different sub-units are given for each type.

Forest type Oligotrophic Mesotrophic Eutrophic Eutrophic oak forests mixed deciduous forests elm-ash forests alder-ash forests

Clusters 1-4 5-13 14-16 17-18 Cluster code 000-011 I 0000-10011 10100-101 1 110-1 11

Vas 29.4 (17.7-37.0) 32.9 (24.6-44.4) 25.9 (20.4-32.6) 25.5 (17.7-3 1.1) Bry 6.6 (5.4-9.0) 3.7 (1.1-8.9) 2.7 (1.3-4.6) 6.2 (4.7-8.2) Tot 36. 1 (23.1 -43.6) 36.6 (26.9-5 1.3) 28.6 (22.9-34.5) 31.7 (22.4-39.3)

L 5.5 (5.2-5.7) 5.1 (4.7-5.4) 4.9 (4.6-5.1) 4.8 (4.4-5.2) T 4.6 (4.4-4.9) 5.1 (4.9-5.3) 5.1 (4.8-5.4) 4.6 (4.5-4.8) c 4.0 (3.9-4.2) 3.9 (3.7-4.0) 3.9 (3.7-4.0) 4.1 (4.0-4.3) M 4.9 (4.7-5.1) 5.1 (4.9-5.4) 5.4 (5.2-5.5) 6.2 (5.7-6.9) R 4.3 (3.3-4.9) 6.2 (5.3-6.6) 6.7 (6.4-7.0) 6.2 (6.1 -6.4) N 3.6 (2.9-4.0) 5.3 (4.7-5.6) 6.2 (5.9-6.5) 6.0 (5.6-6.2)

figures for pH and nitrogen are much higher than those for Table 1 is given in Table 3, showing the presence degrees the oligotrophic oak forests; the light figures are usually of differential and the most frequent species. In the fol­ lower. Compared with the eutrophic forests, nitrogen lowing, each of the four main forest types will be de­ figures are lower throughout. Mesotrophic mixed decidu­ scribed with respect to their: (1) species composition and ous forests represent the most common deciduous forests structure, (2) differentiation (communities and lower sub­ in the Boreo-nemoral zone of Scandinavia, particularly in units), (3) affinities (similarity to other forest types and to southeastern Sweden. communities described in the literature; synonyms), (4) geographic distribution, (5) environment, (6) dynamics.

Eutrophic elm-ash forests (clusters 14-16, n 90) = Eutrophic elm-ash forests include the most fertile forest 4.2 Oligotrophic oakfore sts stands in Scandinavia. However, the total number of species is, on average, the lowest of all forest types. The General characteristics tree layer is characterized by Ulmus glabra ( U. minor), Acer platanoides and Fraxinus excelsior, whereas Tilia Species composition and structure cordata and also Quercus spp. occur less frequently. The In all stands, Quercus spp. dominate in the canopy (Table indicator figures point at high pH-values and nitrogen 4. Betula spp., often functioning as pioneers, as well as the contents of the soil. Elm-ash forests are less common than conifers Pieea abies and Pinus sylvestris can be relatively the mesotrophic forests. abundant, in contrast to all other deciduous forest types. As to the shrub layer, the stands are characterized by Juniperus communis and Frangula alnus. Besides, the Eutrophic alder-ash forests (clusters 17-18, n 4 7) = liana Lonicera periclymenum may be frequent in the Eutrophic alder-ash forests are dominated by Fraxinus understorey, especially in the southwestern part of the excelsior, Ulmus glabra and Alnus spp. Acer platanoides study area. Among the constant and common species in is frequent in the lower canopy, while Tilia cordata and the field layer, we observe many acidophilous species, Que reus spp. are fairly scarce. Whereas the average number some of them more typical for conifer forests, e.g. Luzula of vascular plant species is comparatively low, bryo­ pilosa and Va ccinium myrtillus. A characteristic feature is phytes are usually frequent. The moisture figures are the high frequency of some light-demanding grasses, such much higher than in other forest types, and the nitrogen as Agrostis capillaris, Anthoxanthum odoratum and figures indicate nutrient-rich soils similar to those of the Festuca ovina. Acidophilous species are also frequently eutrophic elm-ash forests. Eutrophic alder-ash forests rep­ found in the bottom layer, e.g. Pleurozium schreberi, resent a less common forest type within the Boreo-nemoral Dicranum scoparium and Hylocomium splendens. zone. However, floristically and ecologically related for­ The oligotrophic oak forests form fairly open, low­ ests are widespread and common in the Boreal zone as growing stands; the upper tree layer usually has a cover of well. only 50-60 % and rarely exceeds 20 m height. The cover A synoptic table of the community types described in of the shrub layer varies between 2 and 50 %, that of the field layer between 50 and 80 %. Bryophytes are usually

Acta Phytogeogr. Suec. 80 24 M. Diekmann

Table 3. Synoptic table of the main community types. The figures indicate presence degrees (%) for each cluster. Non-dif ferential species with low presence degrees are omitted. For the explanation of the strata Tl, T2, S, F and B, see p. 18.

Cluster 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Number of releves 15 5 3 10 30 20 26 19 10 17 19 28 28 22 43 25 21 26 T1 Quercus robur 0 0 13 46 72 19 0 86 100 96 100 100 100 40 41 94 78 96 Betula pendu la 13 20 19 5 0 0 15 14 10 0 13 8 4 0 60 80 33 40 Pinus sylvestris 13 0 0 0 0 0 0 5 0 0 3 9 0 0 0 0 60 40 Betula pubescens 0 0 13 0 0 0 0 5 0 0 0 0 2 4 0 0 40 20 Quercus petraea 26 0 0 0 0 0 0 5 10 7 3 4 0 0 0 0 100 100 Fraxinus excelsior 0 0 0 0 0 93 95 30 10 30 64 10 3 90 79 88 90 69 Ulmus glabra 0 0 0 0 0 0 6 25 21 10 70 10 10 77 88 28 85 42 Acer platanoides 0 0 0 0 0 0 40 23 10 35 36 42 21 36 30 40 9 11 Tilia cordata 6 0 0 0 0 0 10 18 9 4 0 0 70 69 35 47 78 39 Carpinus betulus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 90 Ulmus minor 0 0 0 0 20 0 0 0 10 0 0 0 0 0 0 0 0 48 Alnus glutinosa 0 0 0 0 0 0 0 0 0 0 0 0 0 0 11 4 33 61 Alnus incana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 9 11

Populus tremula 33 10 20 5 10 28 11 Picea abies 33 0 0 23 0 3 2

T2 Quercus robur 0 0 0 0 4 4 12 0 3 46 100 43 20 34 21 47 39 14 Quercus petraea 13 0 lOO 0 0 0 0 0 5 10 0 3 0 0 0 0 0 40 Ulmus glabra 0 0 0 0 6 85 11 47 10 64 15 14 3 86 97 60 85 53 Fraxinus excelsior 0 0 0 0 0 96 35 30 5 41 26 7 3 59 53 32 52 76 Acer platanoides 20 0 0 0 0 16 75 76 10 82 73 85 28 63 67 52 66 26 Tilia cordata 13 0 0 0 0 5 0 18 16 12 0 0 75 84 17 68 71 50 Carpinus betulus 0 0 0 0 0 0 3 5 0 5 0 3 0 0 4 0 0 40 Ulmus minor 0 0 0 0 23 0 0 0 10 0 0 0 0 0 0 0 0 52 Prunus padus 0 0 0 0 0 0 0 0 0 0 5 3 0 9 9 0 33 19 Alnus glutinosa 0 0 0 0 0 0 0 0 0 0 0 0 0 4 2 8 19 38 Alnus incana 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 14 11

Sorbus aucuparia 6 20 66 30 26 42 47 0 47 31 46 60 27 13 47 19 Ma lus sylvestris 20 0 0 0 30 38 52 0 0 0 0 32 4 2 0 7 Picea abies 13 0 0 10 0 3 0 10 11 21 28 0 9 0 9 0

.s_Quercus robur 0 20 26 10 0 5 0 11 31 7 14 0 0 0 0 0 93 100 Juniperus communis 10 0 0 21 40 0 5 3 0 0 0 0 0 0 100 100 100 60 Frangula alnus 3 0 11 10 10 0 5 0 0 0 0 0 0 0 40 60 100 90 Sorbus intermedia 0 3 0 3 10 10 0 5 0 7 0 0 0 0 0 20 20 20 Fagus sylvatica 0 0 0 0 0 0 0 5 0 14 9 4 0 0 0 20 33 10 Quercus petraea 20 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 100 90 Picea abies 20 0 33 10 0 0 5 20 23 21 28 7 9 0 0 4 3 70 Fraxinus excelsior 0 0 0 26 80 70 30 36 60 70 78 32 14 86 79 56 90 88 Ulmus glabra 0 0 0 13 10 85 7 42 60 76 36 10 7 86 86 44 76 57 Ribes alpinum 20 0 0 40 3 55 19 84 50 52 26 78 17 36 48 16 14 3 Corylus avellana 0 33 0 lOO 0 86 80 100 100 82 94 100 85 86 83 72 71 30 Acer platanoides 0 0 10 0 40 26 65 76 5 64 78 78 42 59 44 48 47 19 Malus sylvestris 20 0 0 0 0 0 60 70 20 26 47 17 5 25 4 4 12 4 Crataegus spp . 6 0 0 0 5 26 0 42 4 20 0 0 90 100 80 100 90 88 Cornus sanguinea 0 0 0 0 0 0 3 0 0 0 0 0 53 70 15 47 10 20 Euonymus europaeus 0 0 0 0 0 0 0 0 0 0 0 0 0 20 11 31 10 16 Tilia cordata 0 0 0 0 0 20 31 18 12 4 0 20 80 84 23 68 82 57 Carpinus betulus 0 0 0 0 0 0 3 10 0 5 0 7 0 0 4 0 0 70 Prunus padus 0 0 0 0 6 0 0 0 0 0 0 58 21 28 40 41 61 57 Ulmus minor 0 0 0 0 23 5 0 5 0 0 0 0 0 0 0 0 0 48

Sorbus aucuparia 53 20 66 40 80 55 50 42 20 76 63 53 53 40 60 12 47 19 Lonicera xy losteum 13 0 0 0 10 60 50 84 0 70 15 82 3 45 46 20 47 0 Viburnum opulus 6 0 0 0 43 20 3 15 10 11 5 0 7 0 6 16 23 7 Populus tremula 6 0 33 20 16 5 15 0 0 5 5 25 3 9 0 0 0 3 Rosa spp . 0 0 0 3 5 0 31 20 5 5 0 10 0 2 4 0 0 33

.EHy pericum maculatum 0 0 0 0 5 0 0 10 5 10 0 7 4 0 0 0 0 73 Poa angustifolia 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 60 60 Festuca ovina 0 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 53 100 Hypericum perforatum 0 0 0 0 3 5 0 0 0 0 0 0 0 0 0 0 40 40 Agrostis capillaris 86 60 0 0 11 0 0 0 5 0 7 0 0 0 0 0 33 20 Anthoxanthum odoratum 0 0 0 0 0 0 0 0 3 0 0 0 0 0 80 100 33 10 Juniperus communis 66 3 20 3 15 40 0 5 0 0 0 2 0 0 0 73 100 50 Frangula alnus 0 0 0 0 0 0 5 0 0 0 0 0 0 0 20 40 100 90 Vaccinium vi tis-idaea 6 66 0 0 0 0 0 0 5 0 0 0 0 0 0 0 20 60 Rumex acetosa 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 40 60 33 Hieracium umbellatum 0 0 0 3 0 0 0 5 0 0 0 0 0 0 0 33 4U 33 Potent illa erecta 0 3 0 0 0 0 0 5 0 0 0 0 0 0 0 33 20 33 Picea abies 6 60 0 3 0 0 0 30 0 15 3 3 0 0 0 0 0 70 Trientalis europaea 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 40 33 60 Calluna vu lgaris 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 40 10 Quercus petraea 20 0 0 0 0 0 0 0 5 3 3 0 0 0 0 3 100 100 Lonicera periclymenum 6 0 0 0 0 0 0 0 5 0 3 0 0 0 0 0 66 20 Molinia caerulea 6 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 33 90 Lathyrus montanus 3 0 11 0 0 0 0 0 0 0 100 80 100 10 5 78 57 53 Deschampsia flexuosa 3 0 0 10 0 4 0 0 0 0 93 100 100 100 17 78 3 39 Luzula pi losa 6 10 3 5 20 9 9 0 4 0 80 100 100 70 11 31 25 21 Polypodium vulgare 0 0 0 0 0 0 0 0 0 0 40 20 33 30 17 21 3 3 Vaccinium myrt illus 0 0 0 5 10 0 0 0 0 0 33 60 100 100 17 21 7 3 Solidago virgaurea 3 0 3 0 0 0 2 0 0 0 13 40 66 10 35 31 35 7 Hierac ium vulgatum 0 0 0 3 0 20 0 0 0 0 0 26 100 10 17 57 14 7 Ca1amagrostis arundinacea 0 0 0 0 0 0 0 0 6 0 0 0 20 33 20 11 73 78 Polygonatum odoratum 0 0 0 0 3 0 0 0 2 0 0 0 40 66 11 26 42 21 Me lampyrum pratense 0 0 0 0 0 100 100 100 100 23 55 50 31 60 5 15 7 25 Veronica chamaedrys 0 9 16 0 33 7 100 40 33 20 35 30 10 60 17 57 42 50

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 25

Table 3, cont.

Rubus saxatilis 20 20 33 0 96 60 61 21 10 23 36 75 28 9 18 16 9 3 Campanula persicifolia 66 40 66 0 0 0 11 5 20 0 47 14 25 0 0 0 0 0 Ste1laria holostea 46 40 0 0 0 10 92 21 50 0 0 0 75 4 0 0 0 0 Fragaria vesca 93 0 66 0 26 75 38 68 70 41 47 39 7 0 13 4 14 3 Dac tylis glomerata 53 0 33 0 23 15 38 26 90 29 84 17 67 4 13 8 4 3 Viola riviniana 93 20 100 0 76 60 76 73 80 70 89 89 82 22 62 24 28 3 Corylus avellana 80 20 66 10 66 50 65 84 40 52 73 60 64 45 32 40 28 3 Poa nemoralis 73 40 100 0 40 100 100 73 90 70 89 78 78 50 76 36 19 7 Hepatica nobilis 53 0 66 0 86 100 100 100 100 94 89 96 60 81 81 72 42 3 Fraxinus excelsior 33 0 33 0 93 100 73 63 90 64 68 32 28 95 90 88 85 73 Carex sylvatica 0 0 0 0 73 90 26 42 100 0 0 0 0 22 4 8 9 7 Brachypodium sylvat icum 0 0 0 0 36 60 23 36 90 5 0 0 0 0 2 12 0 0 Cornus sanguinea 0 0 0 0 33 60 11 52 10 0 0 0 0 0 0 4 0 0 Hedera helix 0 0 0 0 6 50 11 78 40 0 0 0 3 0 0 8 0 0 Bromus benekenii 0 0 0 0 6 35 11 10 20 0 0 3 3 0 4 8 0 0 Epipactis helleborine 0 0 0 0 30 15 0 5 40 0 0 0 0 0 0 0 0 0 Dactylorhiza fuchsii 0 0 0 0 10 30 0 5 30 0 0 0 0 0 0 4 0 0 Viola reichenbachiana 0 0 0 0 0 50 26 57 100 0 0 0 3 0 0 20 0 0 Hordelymus europaeus 0 0 0 0 0 0 0 21 20 0 0 0 0 0 0 8 0 0 Rubus caesius 0 0 0 0 46 35 23 47 50 5 5 0 0 9 2 60 9 3 Anemone ranunculoides 0 0 0 0 30 100 100 52 100 0 0 0 10 0 6 88 4 0 Orchis mascula 0 0 0 0 56 30 19 0 10 0 0 0 0 0 0 20 0 0 Euonymus europaeus 0 0 0 0 0 55 30 42 70 0 0 0 0 0 0 32 0 0 Melica uni f lora 13 0 0 0 36 65 61 89 80 0 5 3 64 4 4 44 0 0 Lathyrus vernus 6 0 0 0 20 20 50 10 60 23 10 85 17 9 13 8 19 3 Laserpitium latifolium 0 0 0 0 3 30 30 0 0 5 15 32 28 0 2 0 0 0 Lathyrus niger 13 0 0 0 3 20 53 0 20 0 36 28 39 0 2 0 0 0 Melampyrum nemorosum 0 0 0 0 0 20 3 0 0 0 15 14 3 0 0 0 0 0 Filipendula ulmaria 0 0 0 0 70 60 3 10 0 11 0 0 0 4 6 68 76 96 Geum rivale 0 0 0 0 73 95 34 47 30 11 5 0 3 9 11 72 52 65 Carpinus betulus 6 0 0 0 0 0 3 0 70 0 5 0 7 0 0 0 0 0 Geranium robert i anum 0 0 0 0 0 5 0 15 0 35 5 3 0 4 11 8 0 3 Epilobium montanum 0 0 0 0 0 0 0 0 0 23 0 0 0 0 6 0 19 0 Glechoma hederacea 0 0 0 0 0 0 0 5 0 17 0 0 0 0 4 0 14 15 Urtica dioica 0 0 0 0 0 0 0 0 0 29 5 3 0 4 23 8 9 76 Dryopteris filix-mas 26 0 0 0 3 10 0 36 0 76 68 35 25 40 44 8 71 23 Prunus padus 0 0 33 0 3 0 0 0 0 70 42 39 3 45 51 0 66 53 Actaea spicata 0 0 0 0 0 0 0 10 0 70 10 57 0 31 65 8 71 11 Stachys sylvatica 0 0 0 0 6 5 0 10 0 11 0 0 3 50 46 12 61 50 Campanula latifolia 0 0 0 0 0 0 0 0 0 11 0 0 3 27 25 44 61 42 Ribes uva-crispa 0 0 0 0 0 0 0 0 0 17 5 0 3 13 37 12 4 0 Allium ursinum 0 0 0 0 10 0 0 0 0 0 0 0 0 54 0 0 9 3 Al liaria pet iolata 0 0 0 0 0 0 0 5 0 0 10 0 3 4 4 36 0 0 Corydalis bulbosa 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 20 0 0 Athyrium filix-femina 13 0 0 0 0 0 0 5 0 11 5 0 0 0 9 4 66 88 Equisetum pratense 0 0 0 0 0 0 3 0 0 0 0 0 0 0 2 0 47 50 Crepis paludosa 0 0 0 0 0 0 0 0 0 11 0 0 0 0 2 0 28 19 Va leriana sambuc ifolia 0 0 0 0 0 0 0 0 0 11 0 0 3 0 4 0 19 42 Stellaria nemorum 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 19 84 Chrysosplenium alternifolium 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 76 Ranunculus repens 0 0 0 0 3 0 0 0 0 5 10 3 0 0 9 4 0 65 Matteuccia struthiopteris 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 14 57 Caltha palustris 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0 46 Cardamine amara 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 42 Impatiens noli-tangere 0 0 0 0 0 0 0 0 0 11 0 0 0 0 0 0 4 34 Poa trivialis 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 4 34 Ange lica sy lvestris 0 0 0 0 3 20 0 0 0 0 0 0 0 0 2 0 9 30 Solanum dulcamara 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 23 Carex remota 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 15 Anemone nemorosa 93 100 100 50 100 100 100 100 100 82 94 100 92 95 93 100 90 65 Sorbus aucuparia 73 80 66 80 70 50 57 78 60 76 73 89 64 40 62 16 47 19 Viburnum opu lus 13 20 33 0 70 45 23 57 10 35 21 35 7 9 39 12 47 7 Acer platanoides 40 0 0 0 10 85 88 31 70 88 89 89 64 81 81 84 76 26 Geum urbanum 13 0 0 0 30 40 61 57 70 64 42 50 39 59 95 100 76 34 Geranium sylvaticum 6 0 0 0 86 55 61 68 70 47 63 71 3 59 62 28 52 19 Taraxacum officinale agg . 6 0 0 0 80 25 3 36 10 41 36 21 3 18 65 68 33 11 Ranunculus auricomus agg . 6 0 0 0 76 95 84 89 100 47 57 42 82 45 74 96 38 42 Paris quadrifolia 5 6 0 0 0 46 75 15 30 70 26 75 3 36 60 68 80 61 Deschampsia cespitosa 0 0 33 0 73 75 42 21 30 17 10 10 46 9 27 12 33 46 Elymus caninus 0 0 0 0 50 65 23 15 20 29 10 3 3 45 23 28 4 15

.l2Dic ranum polysetum 33 80 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Rhodobryum roseum 66 0 33 0 0 0 0 0 10 0 5 0 0 0 0 0 0 0 Pleuroz ium schreberi 86 80 66 80 0 0 0 0 0 0 21 0 0 0 0 0 0 0 Hylocomium splendens 53 60 33 20 0 0 0 0 0 0 15 0 0 0 0 0 0 0 Dicranum scoparium 33 60 66 100 0 0 0 5 0 0 5 0 0 0 0 0 0 0 Polytrichum formosum 26 40 100 40 0 0 0 0 10 0 0 0 0 0 0 0 0 0 Hypnum cupressiforme 20 20 66 90 0 0 15 15 0 0 5 3 3 4 0 0 4 0 Leucobryum glaucum 0 40 0 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Rhytidiadelphus loreus 0 0 66 20 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Brachythecium velutinum 20 40 0 0 6 20 57 42 60 0 0 0 3 4 4 4 4 0 Plagiomnium affine 93 20 66 10 0 45 34 42 60 11 21 7 10 13 6 0 42 50 Scleropodium purum 6 0 0 0 16 30 3 47 30 0 0 0 0 9 2 4 4 0 Fissidens cristatus 0 0 0 0 0 30 23 26 30 0 0 3 0 0 6 0 9 0 Plagiomnium undulatum 0 0 0 0 66 95 53 42 80 11 21 3 14 36 37 36 80 84 Eurhynchium angustirete 0 0 0 0 0 10 11 31 30 0 0 0 0 50 4 0 61 30 Eurhynchium striatum 0 0 0 0 30 25 15 5 10 0 0 3 0 0 4 24 4 3 Thuidium tamariscinum 0 0 33 0 26 15 19 42 30 5 0 3 0 0 0 0 14 38 Brachythecium rivulare 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 76 Rhizomnium punctatum 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 23 Brachythecium rutabulum 100 40 33 10 26 65 50 47 90 11 31 10 25 45 41 44 90 15 Rhytidiadelphus triquetrus 20 20 33 10 46 60 23 26 80 11 42 28 7 27 9 0 42 7 Eurhynchium hians 20 0 66 20 56 100 65 68 90 23 21 17 7 95 81 56 95 76 Cirriphy llum piliferum 13 0 0 0 6 20 30 47 80 17 36 10 21 40 44 20 85 61 Fissidens taxifolius 0 0 0 0 26 35 7 15 40 5 0 3 0 13 2 16 4 3 Atrichum undulatum 0 0 33 20 3 10 3 0 20 5 0 0 10 0 4 0 23 26 Plagiochila porelloides 0 0 33 0 3 25 7 5 50 0 0 7 0 13 4 0 42 38 Eurhynchium praelongum 0 0 0 0 3 25 3 5 20 0 10 0 0 0 0 8 0 26

Acta Phytogeogr. Suec. 80 26 M. Diekmann

Table 4. Releve table of oligotrophic oak forests. (See Table 3.)

Running Number 1234567891 111 11111 122222 222223333 0 123 45678 901234 5 67890123 Cluster 3 2 la 1b Cluster code 011 010 001 0000 0001 Region HBB BOOOO HHHHHHHHHH ssssss B00600000 AOO AAAAAAAAAA LLLLL MMMMMM LLLLLLLLL Cover Tl (XlO) 6667657755 766 46466 566456 456666655 % T2 (X10) - 1 - 1 --1 - 1 - 222 11222 121-21 --1111112 % S (XlO) 1115211311 322 33442 343542 53 5554353 % F (X10) 7775756667 677 66565 777875 677 877 87 8 % B 1212112111 312 124-3 125211 212522321 0 0 00 0 0 0 5 N Tot 2113213222 434 23323 444443 445534454 - 3686784600 681 83704 008808 470061447 N Vas 2112112211 333 22222 333433 344433343 - 0125847264 540 47605 858112 713424657 N - Bry 3569947446 141 451-9 251796 767647891 1 1 0 0 1

'll Quercus petraea 6666656655 666 446.6. Pinus syl vestris 3 ...33 ...3 ..234 ..2 ....2. Betula pubescens ..3 ....3 .. . 22 .. Quercus robur 55565 55.5. 6 556566655

Betula pendula ....33 .3.4 4 .. 44.22 432422234 Populus tremula ....3 ...... 4

'1'2. Quercus petraea . 3.3 ....43 444 ..3.4. Quercus robur 44444 ..3444444

Sorbus aucuparia 3 43 .. . 44 . ..2 . .4 .. .. Picea abies •••...3 ••...... 3... 3 Betula pendula 3 .3 ...... 2. Betula pubescens ••• ••....3 .2 ... .3.. .. Pinus sylvestris • • • • . • • •• ...22 ..2 ...... Acer platanoides 3.4 ...... 2 ... .. Malus sylvestris ... 2 ..22 .

.s. Quercus petraea 322444344 . 442 ..4. 54 Picea abies 2223.2.2.2 3 .. . 2. 23 . Quercus robur . . . 32 ..... 42454 44332. 441444444 Corylus avellana 4 4424 .. 445524244

Malus sylvestris .• ...1. 22 ...2 . 211 .211. Acer platanoides ...2 ...... 243 .42 . .. 1 ..... Ribes alpinum ....1 ..133 . 333 Fraxinus excelsior 322 ...1 .....

Tilia cordata .345••• ..

Juniperus communis 2 ...43242 . 422 45444 233334 424354444 Frangula alnus 3234 . 22433 232 211. . 4221. .13 . Sorbus aucuparia . . . 4 ..232 . . 44 3 .... 23 .2.3 .213 .3 ... Betula pubescens ..4 .2 ..3 .. 3 .. .3 .....2. Sorbus intermedia ... . . 2.2 .. 2 .... .2.... 3 ...... 1. Fagus sylvatica . . . . 2 2 .2.2.2 Rosa spp . ....• .• 2 ...... 33 .. 11 Betula pendula . ..2. 2. 2 .. ...1. Populus tremula ....3. 2 ... ..3 ..3 ......

EQuercus petraea 4434443443 444 ..3. 43 Molinia caerulea 4.24323444 ..1 ....1 ...... 1. Lonicera periclymenum 4. 4 ... . 54 5 ...... Picea abies 2232•••• .1. .12 1.1.1 . ...1. Trientalis europaea 4 ..3 ..4442 3 .. 41... Dryopteris carthusiana 1. .4 3.2 .

Betula pubescens . . . 23.• ...... 2 .. Vaccinium uliginosum ....31. 1.. Calluna vulgaris ....2 ...... 32 .. Vaccinium myrt illus 5545454545 644 5.26. 35.434 Vaccinium vitis-idaea . . 1242 . 3.3 32 . 3 .. .. 1 ..... Viola riviniana 233 1 .. .. 3.2332 312233222 Fragaria vesca 21. 31.223 113433432 Pea nemoralis 122 .1. .1 414313 .1. .1423 . Corylus avellana ...... 2 .. 1.1 22.2.. 224413222 . . . . 1 Hepat ica nobilis 12 . 413444 3 ...... 1. Melica nutans 24 . 4.3434 4 ...... 2 Dactylis glomerata ..1 31 .... 1.22 .1.11 Polygonatum odoratum 11. ..3. 2. .12 ....21 Veronica officinalis 121 22 ...21. . Mercurialis perennis 22. .. 2.12 ...... 3 Fraxinus excelsior 1. 221. ..22 ... .. Prunus avium . 3. 1.1..• .. 1 ...... Trifolium medium ...1. .1.21.2.2. 3 Galium boreale ...4 .. .. 122 .323 Vicia sepium 3.12 .1 ...2 ....4 Acer platanoides 234 .21 ...1 ..... Quercus robur 34444 3412 .. 344444444 Veronica chamaedrys .. 3 ..3.1 423433 123423234

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 27

Ta ble 4, cont. Festuca ovina ...3 ...... 34323 . . 12 .. 331.2.33 . Poa angustifolia 32 .1. 431241422 Campanu 1a persicifolia 12 . .1. .1 3.1.13 .1.1.2112 Hypericum maculatum 2.122 . .23313 .43 Ribes alpinum ...12 .13331332 Stellaria ho lostea 4 .. . 344 . 4234 4. Hypericum perforatum .11 .. 121. .1.1.

Melampyrum pratense 4442444424 344 35434 424343 454555554 Deschampsia flexuosa 5564534555 345 45444 44 .443 543243444 Anemone nemorosa 5 ..4 ..43 5. 455 34124 .56444 544444444 Luzula pilosa .2241 .31.3 214 44422 13.322 13 ..32334 Sorbus aucuparia 3232 .. 2222 . 33 .1111 22 . 21 . .2211.111 Lathyrus montanus 1 ...... 323 .3213 424244 432331232 Juniperus communis . 1. .3312 .. 22 . 22322 1.11. . .11132223 Convallaria maj alis 3 ..4 ..4. 3. .4. . 441 . . 5555 . .45551455 Agrostis capillaris ...2 ...2 .. .. 2 32 ..2 4. 4. 33 421244332 Anthoxanthum odoratum ...2 ...... 2 34422 ..13 1. 342344434 Frangula alnus 3122 . 21132 211 2 21. .1. ... .1. . Maianthemum bifolium . 1.3.. 3 ..1 3 .. 2.3.. . 2 .... .222 .2.2 . Polypodium vulgare ....1. 2 .1. . . 4 2 .... 322 .22 3 ...... Oxalis acetosella ...... 42 .. 434 4 . 2 .... 144 ....3 . Rumex acetosa ..1 .1.•..1.1 . 212 .. 111 Rosa spp . ....1 1 ...... 111111 .1 Hieracium vulgatum ...2 ...... 111 3 ...... 2. 11 Hieracium umbe llatum ..1 .11.. 2 .. . .. 1 .1. 11. . . Solidago virgaurea ... 1 ...... 42 .11 .. ..11. Sorbus intermedia ...... 1 ...... 11 . 1. ..1.1. . 1.. Potent illa erecta .. 1 .1...... 3 .. .2.... 111 Pteridium aqui l inum 1.3 ..2 .2. 1 ..... 3 ...... Calamagros tis arundinacea 1 .....2 ...• ... 2 .. . 4.4. 3 Carex digitata 1 ...... 24...... 3 .1 ...... Rubus saxatilis . ..1...... 122 . 3 Carex pallescens . 21•• . .11. . ..1 ...... Rubus idaeus 2 ...... 2...... 1 5 .. Pinus syl vestris .....1 ... . .11 .. .1 ...... Viburnum opulus 2 ...... 1 ...1 .. ..2 ...... Hieracium murorum . 1. 4 .1. .1 Mi l ium effusum ..1 ...... 222 Avenula pubescens ....1 .... 21 ..2 Galium verum 1. 2 ..1 1 ...... Dryopteris filix-mas .2.2.. 4 ...... 1 Cardamine bulbifera ...4 ...... 211 Festuca rubra ...3 .. 4 .....11 . Galium aparine ..2214 ... Carex montana ...3 2 ... 2

Galeopsis tetrahit ...... ••4 .1. .• .1. Carex pilulifera ...... 1 .. .1.. 1 .... Melampyrum sylvaticum .. 2 ...4 .. .2...... Anthriscus sylvestris ...2 .. ..13 ..... Malus sylvestris .....1 .2 ..1 .... Rubus fruticosus agg . 2 ..2 ...3 . Crataegus spp . . . . 1.1. 1

Geranium sanguineum . ...1.1.1 .

� Rhytidiadelphus loreus ....2 ..1 .. 1.3 Leucobryum glaucum 22 ... 2 . .1. . Plagiomnium affine ...••..1 ..... 2.1 ....2 223232 .12411213 . Rhodobryum roseum .. 1 . . 2.41 3113.22.3 Dicranum polysetum 144.4 .. 1 ... .12.2 ...1 Brachythecium velutinum .. 1.1 .....121 .

Pleurozium schreberi . 3223 . 1321 2.4 245 .. 4 . 33242 34122 .224 Dicranum scoparium 1322412221 3.2 2 .1. 2 .1. .1. 2 ....1. .1 Brachythecium rutabu1um ..1 ...... 4 .11. . 232321 113313 314 Hypnum cupressiforme 2223421.22 22 . . .1...... 1. 2 ...... 1. Hylocomium splendens ....2 ....2 ..4 3 .1.1 .22 .31 ..12 ..1. 2 Polytrichum formosurn ...3.2 3.1. 322 .. 1.1 ..1 ...... 1. 11 Eurhynchium hians . 33 ...... 2 .1 ..222 . Rhytidiadelphus triquetrus ...... 1. .. ..3 ....1 . ..2 ...... 12 Brachythecium populeum ..1.2.2 ..1 .....1 Lophocolea bidentata ...1 ..2 ... 2 .. ..1 .. Atrichum undulatum 1 ..2 ...... 2.

Additional species (occurring in one or two re1eves ): 2X: (T2 ) Tilia cordata 20:3, 21:3 (S} Lonicera xylosteum 27 :2, 28 :1, Prunus avium 12 :2, 25 :2, Ulmus glabra 19 :2, 20 :2 Achillea (F) millefolium 24:1, 28:1, Athyrium filix- femina 20:1, 22:2, Campanula rotundi folia 16:1, 28:1, Geurn urbanum 27 :2, 28:1, Lathyrus niger 21:3, 23 :2, Luzula campestris 26:2, 29:1, Melica uniflora 21:2, 23 :3, Polygonaturn multiflorurn 27:1, 28:2, Populus tremula 7:3, 27 :2, Vincetoxicum hirundinaria 19 :2, 25:3 (B) Aulacomnium androgynum 15 :1, 16 :1, Brachythecium reflexum 7:4, 23 :1, Brachythecium salebrosum 26 :2, 30 :1, Cirriphyllum piliferum 21:2, 22 :1, Isopterygium elegans 5:3, 12 :1, Rhytidiadelphus squarrosus 27 :1, 28:1.

1x: (T1} Betula sp . 9:3, Picea abies 11 :3, Tilia cordata 22 :4 (T2 } Populus tremula 27:2, Sorbus intermedia 19 :3 (S} Crataegus sp . 23 :2, Prunus spinosa 21 :2, Prunus sp . 7:2, Viburnum opulus 27 :1 (F) Arrhenatherum elat ius 28 :1, Carpinus betulus 20 :1, Deschampsia cespitosa 12 :1, Fagus sylvatica 14:1, Filipendula vulgaris 24:1, Galium odoratum 14:2, Geranium sy1vaticum Gymnocarpium dryopteris 19:3, Holcus 22 :3, mollis 1:3, Lathyrus pratensis 30:1, Lathyrus vernus 24 :2, Luzula luzuloides 8:1, Moehringia trinervia 7:1, My celis muralis 21 :1, Paris quadrifolia 27 :1, Phleum pratense 25:1, Platanthera bifolia 16:1, Primula veris 23:2, Prunus padus 11:1, Prunus spinosa 21:1, Pulmonaria officinalis 22 :2, Ranunculus acris 22 :1, Ranunculus auricomus agg . 22 :4, Rhamnus catharticus 28:1, Sedum telephium 21:1, Serratula tinctoria 32:1, Silene dioica 7:2, Stellaria graminea 25 :1, Taraxacum officinale agg . 22 :1, Thelypteris phegopteris 4:3, Tilia cordata 21:1 Cladonia fimbriata 15:1, Homalothecium sp . 31:1, Isothecium alopecuroides 12 :1, Lepidozia reptans 4:2, Lophocolea heterophylla (B) 4:2, Plagiochila porelloides 11 :2, Plagiothecium denticulatum 11 :3, Plagiothecium nemorale 31:1, Plagiothecium undu latum 13 :2, Polytrichum juniperinum 16:1, Scleropodium purum 32:1, Thuidium tamariscinum 11:2.

Acta Phytogeogr. Suec. 80 28 M. Diekmann

3 N !Q X

2

4 4' 1

0

1 s ,. . 5

- 1

AXIS 1 -2 T------.------,,------.------�

-2 -1 0 1 2

Fig. 8. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 33 releves of oligotrophic oak forests. Numbers of different sub-units plotted onto the sample plot positions which are slightly adjusted in order to avoid overlapping figures. 1-2. Quercus petraea­ Frangula alnus community (1. Trientalis europaea sub-community; 2. Viola riviniana sub-community); 3-5. Quercus robur-Betula pendula community (3. Trientalis europaea sub-community; 4. Viola riviniana sub-community, Acer platanoides form; 5. Viola riviniana sub-community, Ribes alpinum form). not very abundant and only occasionally reach up to 40 % community (releves 14-18) and a Viola riviniana sub­ cover. community (releves 19-33). The latter is divided into a Acer platanoides form (releves 19-24 ), based upon stands Differentiation from SmcUand, and a Ribes alpinum form (releves 25-33), Two communities are distinguished, occurring in differ­ including stands from bland (one stand from Blekinge). ent geographic regions: The differentiation described above is well expressed in 1. Quercus petraea-Frangula alnus community (releves the ordination diagram (Fig. 8). 1-13), in releves from westernSweden, usually character­ ized by Quercus petraea, with Frangula alnus as a con­ Affinities stant species in the shrub layer and with a few differential There are only few studies of oligotrophic oak forests species in the field layer which require fairly high soil from the Boreo-nemoral zone, e.g. by Malmstrom (1937) moisture (e.g. Molinia caerulea and Vaccinium uligino­ from Halland, Ttixen (1951), Ivarsson (1962) from sum). Two sub-communities are distinguished: a species­ BohusHin, Berglund (1963) from Blekinge and Wallin poor Trientalis europaea sub-community (releves 1-10) ( 1973) from Vastergotland. Olsson ( 197 4, 197 5) described with the basic species assemblage and an often fairly high similar communities from S Sweden, mainly with respect abundance of Picea abies, Pinus sylvestris and Vaccinium to the Nemoral zone, and Rtihling & Tyler (1986) com­ spp., and a comparatively species-rich Viola riviniana pared the oak forest vegetation of Skfme with that of sub-community (releves 11-13) with a larger number of Smiiland. A. Larsson (1974) studied the forest 'Ottenby differential species, many of them also found in lund' on southernmost bland, an open oligotrophic birch­ mesotrophic and eutrophic forests. oak forest, structurally resembling a wooded meadow or 2. Que reus robur-Betula pendula community (releves 14- wooded pasture. Studies of this forest type from the 33), in releves from eastern Sweden, characterized in the Nemoral zone and the Western broad-leaved and pine canopy by either Que reus robur or Q. petraea and Betula forest region in Norway were made by A. Bj�rnstad pendula and with some continental elements in the field (1971), Skogen (1971), Bakkevig (1974), Blom (1982) layer (e.g. Campanula persicifolia ). In principle, the same and 0vstedal (1985). differentiation is found as for the Quercus robur-Fran­ The releves in the literature from SW Sweden and gula alnus community, i.e. a Trientalis europaea sub- Norway reveal a high affinity to the Quercus petraea-

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 29

Frangula alnus community. Particularly the Norwegian caused by the fact that these two site factors are interre­ stands show a high frequency of species with an oceanic/ lated (the same may be true for some other species, e.g. sub-oceanic distribution, e.g. Holcus mollis, Hypericum Molinia caerulea), because a higher precipitation causes a pulchrum, Luzula sylvatica and Rhytidiadelphus loreus. higher rate of leaching of the acid soil (Weimarck 1947). In Sweden, the forests belonging to this community mainly This stands in contrast to the situation in central Europe, consist of Quercus petraea or the hybrid between Q. robur where Q. robur is considered as having a larger amplitude and petraea. However, the stands situated in the Western with respect to low nutrient-levels and the moisture re­ broad-leaved and pine forest region are exclusively formed gime (Ellenberg 1986; Oberdorfer 1992). by Quercus robur, which in Norway shows a more oce­ anic distribution than elsewhere in Europe and extends Geographic distribution further to the north than Q. petraea. The releves in the The forests are located in Halland and BohusHin (stands of literature from SE Sweden can be assigned to the Quercus the Quereus petraea-Frangula alnuscommunity), as well robur-Betula pendula community, based on the occur­ as in Blekinge, Smruand and bland (stands of the Que reus rence of a number of continental species, e.g. Vincetoxicum robur-Betula pendula community). All stands are located hirundinaria and Campanula persicifolia. Whereas the in the vicinity of either the west coast or the Baltic Sea, i.e. stands on bland only contain Que reus robur, both Q. robur in relatively oceanic regions. Occasionally, they grow and Q. petraea, as well as their hybrid, can occur in the close to the sea, are influenced by saltspray and have a mainland stands. character of very low-growing and wind-bent forests, so­ Most of the descriptions of oligotrophic oak forests called 'krattskogar' (Sjors 1967). The literature reveals mentioned above reveal a similar division into two sub­ that the oligotrophic oak forests in Scandinavia are clearly units, as distinguished here at the level of sub-communities: concentrated in coastal regions, being particularly abun­ - a species-poor sub-unit on very nutrient-poor sites, dant in SW Sweden and S and SW Norway. This regional usually differentiated by high frequencies of Pinus distribution is well in accordance with that of correspond­ sylvestris, Trientalis europaea and Va ccinium spp. Often ing oak fo rests in central Europe, which are confined to used synonyms are: the oceanic and sub-oceanic parts and which, further to 'Hedekskog' (e.g. Lindquist 1934; Sjors 1967; Wallin the east, become replaced first by mixed oak-pine forests 1973; Bergendorff et al. 1979); 'Ekskog av ortfattig typ' and eventually by pure pine forests. The distribution (Anon. 1984); Quercus petraea-Deschampsia flexuosa pattern of oligotrophic oak forests in Scandinavia is prob­ community (Berglund 1963; Nylander 1975); Melampyro­ ably determined by the distributions of Quercus spp., Quercetum roboris (Olsson 1974, 1975) and Populo­ Picea abies/Pinus sylvestris and Fagus sylvatica and by Quercetum (Tiixen 1951; A. Bj�rnstad 1971; Kielland­ the competitive relationships between these species on Lund 1971). oligotrophic sites. SW Sweden and S and W Norway - a species-rich sub-unit on less nutrient-poor sites, partly lie outside the natural distribution areas of Norway characterized by a large number of more demanding spe­ spruce and beech, allowing oak to dominate these sites. cies, such as Viola riviniana, Poa nemoralis, Melica Picea is limited by its dependence on sufficiently cold nutans and Dactylis glomerata. Main synonyms are: winters (Dahl 1992; Holten 1993). Outside its natural area 'Angsekskog' (Sjors 1967; Bergendorff et al. 1979); 'Ek­ in SW Norway, Picea grows well when planted, but fails skog av ortrik typ' (Anon. 1984) and Melico-Quercetum to regenerate (Dahl 1992). Where Quercus and Picea (A. Bj�rnstad 1971; Kielland-Lund 1971). occur together, the latter' s competitive ability becomes However, many of the synonyms mentioned above increasingly weakened in comparatively oceanic areas. (and by Bergendorff et al. 1979) are ambiguous, since Fagus is limited by both a short growing season and by they partly indicate oligotrophic oak forests, partly low winter temperatures (and late frosts), but is able to mesotrophic mixed deciduous forests described in Chap­ grow on similarly dry and nutrient-poor soils as Quercus ter 4.3. (Leuschner et al. 1993 ). In fact, both species form stands As to the ecology of the two oak species, Quercus on oligotrophic sites in the transition between the Nemoral petraea has its ecological optimum in S Norway in the and Boreo-nemoral zone in Sweden (Sjors 1967). Populo-Quercetum, whereas Q. robur occurs more fre­ The coastal districts in Scandinavia, with their rela­ quently in the Melico-Quercetum and, thus, prefers the tively oceanic climate, are thus particularly suitable for better sites (A. Bj�mstad 1971). In Sweden, Q. petraea is oligotrophic oak forests. On the other hand, hardly any also confined to acid, nutrient-poor soils, especially on stands can be found in the northeasternparts of the Boreo­ hill-tops, escarpments and upper slopes (Weimarck 1947; nemoral zone with their comparatively continental cli­ Sjors 1967; Olsson 1975; Bergendorffet al. 1979), whereas mate and low precipitation. In SE Norway with its fairly Q. robur can be found in the whole range from oligotroph­ dry and continental climate, the oak forest communities ic to eutrophic sites. The preference of Q. petraea for both are replaced by locally restricted vicarious spruce forest relatively oceanic (wet) and nutrient-poor sites is partly communities (A. Bj�rnstad 1971).

Acta Phytogeogr. Suec. 80 30 M. Diekmann

Fig. 9. Oligotrophic oak forest, Quercus robur-Betula pendula community, in the canopy com­ pletely dominated by Quercus robur. Juniperus communis is fre­ quent in the shrub layer. A thick oak litter layer is visible on the ground. - Stora Vickleby, Oland, May 1993. Photo Folke Hellstrom.

Environment pressed in a somewhat thinner litter layer due to higher Oligotrophic oak forests were encountered both on level humification and weaker podsolization (see also A. ground and slopes of varying aspect and steepness, with­ Bj0rnstad 1971). This could be seen in the soil profiles of out any clear differentiation in different sub-units. How­ three stands in an oligotrophic oak forest on bland: all ever, they are most often found on the top or upper parts of stands grew on oligotrophic brown earth rankers; the one small, morainic hills with rocky and coarse-grained soils of the Trientalis europaea sub-community was extremely (cf. Berglund 1963; Wallin 1973). For southern Norway, shallow and podsolic, whereas the two belonging to the A. Bj0rnstad (1971) showed that stands belonging to the Viola riviniana sub-community were much deeper and Populo-Quercetum are confined to north- or west-ex­ less podsolized. posed slopes, whereas stands of the Melico-Quercetum The indicator figures in Table 5 point at much higher prefer more or less south-exposed, often steep slopes. pH and nitrogen contents of the soil in stands belonging to Oligotrophic oak forests grow on either coarse sandy soils the Viola riviniana sub-communities. This trend, judged on, e.g. aeolian and glaciofluvial sediments or shallow from the floristic composition alone, is confirmed by stony soils on Precambrian bedrock, often gneisses and measurements from stands on bland (Diekmann, un­ granites. On bland, oligotrophic oak forests are found on publ.), and by the investigations of A. Bj0rnstad (197 1), gravelly soils over Cambrian shale. The soils can be Wallin (1973) and Riihling & Tyler (1986), who found classified as coarse-textured podsols, rankers and oligo­ higher values for pH, total N (A. Bj0rnstad) and base trophic brown earths, as well as transitional forms be­ saturation, as well as lower organic matter contents, for tween them (e.g. Ivarsson 1962; Berglund 1963; Wallin the species-rich sub-communities than for the species­ 1973; Olsson 1974, 1975). The humus form is usually raw poor sub-communities. The moisture figures are slightly humus or mor, covered by a thick, closed oak litter layer. higher for the two sub-communities of the Quercus Stands belonging to the Viola riviniana sub-communities petraea-Frangula alnus community, whereas the light occur on comparatively better soils, morphologically ex- figuresare lower.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoralScandinavia 31

Table 5. Oligotrophic oak forests. Number of vascular plants (Vas), number of bryophytes (Bry) and total number of species (Tot), as well as indicator figuresfor light (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for different sub­ units. Average values are given. The sub-units correspond to: 1-2. Quercus petraea-Frangulaalnus community (1. Trientalis europaea sub-community; 2. Viola riviniana sub-community); 3-5. Quercus robur-Betula pendula community (3. Trientalis europaea sub­ community; 4. Viola riviniana sub-community, Acer platanoides form; 5. Viola riviniana sub-community, Ribes alpinum form).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 17.7 5.4 23.1 5.3 4.4 4.2 5.1 3.3 2.9 2 33.0 9.0 42.0 5.2 4.5 3.9 5.0 4.2 3.7 3 24.2 6.0 30.2 5.6 4.6 4.2 4.9 4.0 3.3 4 37.0 6.6 43.6 5.7 4.9 4.0 4.8 4.9 4.0 5 35.2 6.2 41.4 5.4 4.9 3.9 4.7 4.8 3.9

The correlation coefficients between releve scores on capillaris) have been favoured at the expense of Vaccinium Calluna vulgaris the first two CA axes and explanatory variables are given chamaephytes such as spp. and in Table 6. Two main complex-gradients appear to be of (Sjors 1967; Bergendorff et al. 1979). The deforested major importance for the floristicvariation of oligotroph­ areas were later oftenplanted with conifers, on sandy soils ic oak forests: a moisture complex-gradient, connected mainly with Pinus sylvestris (Olsson 1975). In other ar­ with longitudinal location [the releve scores on axis 1 eas, secondary succession led to the recovery of a more (eigenvalue 0.413) show a strong negative correlation natural vegetation. In Halland, BohusHin and Blekinge, with longitude and strong positive correlations with yearly young oligotrophic oak forests of different stages are precipitation, precipitation during the growing season, de commonly seen on rocky or sandy ground. According to Martonne's index and M-figures], and a complex-gradi­ Wallin (1973), oak is also expanding its range in Vaster­ ent in nutrient-status (strong negative correlations for R­ gotland and can locally invade conifer plantations. Many and N-figures with the releve scores on axis 1). As previ­ stands will eventually develop a closer canopy, favouring ously mentioned, these complex-gradients are interre­ true forest species at the expense of more light-demand­ lated. Axis 2 has a much lower eigenvalue (0.285), and the ing species of open habitats. As to the succession of trees, releve scores on this axis are negatively correlated with Quercus spp. show a rich regeneration and will probably the three temperature variables and L-figures. There is hold their dominant position also in mature stands, while poor correspondence between the climatic variables lon­ gitude/continentality index and C-figures, as well as be­ tween the temperature variables and T -figures. Table 6. Oligotrophic oak forests. Spearman rank correlation Dynamics (rs) between sample plot scores on the first two eA-ordination The present areal extension of the oligotrophic oak for­ axes and 15 environmental variables. The number of observa­ tions is 33. Significancelevels p < 0.05, p < 0.01 and p < 0.001, ests includes only a small fraction of the former one (Sjors the latter in bold face; n.s. - not significant. For abbreviations of 1967; Bergendorff et al. 1979), due to human impact in environmental variables, see Chapter 3. form of cutting, burningand grazing, and later, afforesta­ tion with conifers. The two provinces Halland and Variable Axis 1 Axis2 BohusHin, where these forests once were very common, rs p rs p first suffered from widespread deforestation, but later L -0.529 0.01 -0.440 0.05 have been extensively planted with pine or spruce T -0.506 0.01 0.202 n.s. (Malmstrom 1939; Fries 1958). In westernSweden, oak c 0. 159 n.s. -0.391 0.05 forests were often replaced by Calluna heathland M 0.452 0.01 -0.183 n.s. R -0.849 0.258 n.s. (Malmstrom 1937; Fries 1958; Wallin 1973). They were 0.001 N -0.650 0.001 0.373 0.05 only exceptionally treated as wooded meadows, espe­ TEM YEAR 0.068 n.s. -0.461 0.01 cially since the poor soils did not allow a continuously TEM FEB 0.125 n.s. -0.406 0.05 high hay production. TEM JULY -0.425 0.05 -0.526 0.01 PRE YEAR 0.777 0.354 0.05 Nearly all of the oligotrophic oak forests have been 0.001 PRE VEG 0.792 0.001 0.345 n.s. intensely grazed, but, in most areas, grazing has very MAR IND 0.806 0.001 0.351 0.05 much decreased or ceased during this century (e.g. Wallin CON IND -0.656 0.001 0.122 n.s. 1973; Anon. 1984). However, its impact is still visible in LAT 0.298 n.s. 0.228 n.s. LON -0.789 -0.280 n.s. many stands. Grasses (e.g. Festuca ovina and Agrostis 0.001

Acta Phytogeogr. Suec. 80 32 M. Diekmann

other deciduous tree species, such as Acer platanoides Table 7. Mesotrophic mixed deciduous forests. Speannan rank and Tilia cordata, may slightly increase in frequency on correlation (r5) between sample plot scores on the first two CA- the richest sites (see Chapter 5). ordination axes and 17 environmental variables. The number of observations is 197. For further explanation, see Table 6.

Variable Axis 1 Axis 2

4.3 Mesotrophic mixed deciduous rs p rs p forests L -0.059 n.s. -0.144 0.05 T 0.212 0.01 -0.178 0.05 General characteristics c -0.266 0.001 0. 116 n.s. M 0.458 0.001 0.740 0.001 Species composition and structure R -0.688 0.001 0.474 0.001 Five tree species are of major importance; they occur in N -0.442 0.001 0.589 0.001 varying compositional mixtures and abundance: Acer TEM YEAR -0.590 0.001 0.060 n.s. TEM FEB platanoides, Tilia cordata, Fraxinus excelsior, Ulmus -0.588 0.001 0.164 0.05 TEM JULY -0.265 -0.010 n.s. glabra and Quercus robur. The latter often dominates in 0.001 PRE YEAR 0.392 0.001 0.107 n.s. the upper tree layer, but is much less abundant in the other MAR IND 0.507 0.001 0.024 n.s. layers. Where present, Tilia differentiates from the eutro­ CON IND 0.478 0.001 -0.154 0.05 phic elm-ash forests. Populus tremula, Betula pendula, LAT 0.437 0.001 0.074 n.s. Sorbus aucuparia and Malus sylvestris are frequent 'trivial' LON -0.204 0.01 0. 169 0.05 INCL 0.691 trees, the two latter only occurring in the understorey. 0.001 -0.391 0.001 ASP 0.105 n.s. -0.255 Among the shrubs, Corylus avellana is particularly 0.001 HEAT IND 0.180 0.05 -0.329 0.001 abundant; it usually forms a 5-8m, locally up to lOm, high sub-canopy between tree layer and lower shrub layer. Other species reaching high presence degrees are Cratae­ gus spp., Loniceraxylosteum and Ribes alpinum, as well as young individuals of the tree species mentioned above. very abundant and reach up to 90 % cover, but may In contrast to the oligotrophic oak forests, the field exceptionally be almost absent. Also the cover percent­ layer is characterized by a strong floristic seasonality. ages of the field and bottom layers vary considerably. During spring, the ground is densely covered by geophytes such as Anemone nemo rosa, Ranunculus auricomus agg. Differentiation and R. ficaria. These wither in early summer and leave Three communities are distinguished, comprising releves space for a large number of herbs, grasses and tree sap­ from geographically separated regions: lings. Besides forest species (e.g. Hepatica nobilis, Elymus 1. Quercus robur-Fraxinus excelsior community (Got­ caninus, Paris quadrifo lia and Oxalis acetosella), ele­ land), characterized in the canopy by Quercus robur and ments of gaps, forest fringes and meadows are frequent, Fraxinus excelsior. The field layer is relatively species­ e.g. Geranium sylvaticum, Geum urbanum and Vicia poor, without any species differentiating from the two sepium. Among the occasional or scarce species in Table other mesotrophic forest communities (Table 8, 30 releves ). 3, some are quite rare within Scandinavia, e.g. Bromus 2. Quercus robur-Euonymus europaeus community benekenii and Hordelymus europaeus. Several species (bland), with a varying species composition in the canopy differentiate the mesotrophic forests from the eutrophic as described above for the mesotrophic forests in general. elm-ash forests, most of them with comparatively low In the shrub layer, Euonymus europaeus and Cornus demands of nitrogen and an ability to grow where pH is sanguinea (also on Gotland) differentiate from the Quereus moderately low. Among these are some elements of forest robur- Tilia cordata community. A large number of dif­ fringes (e.g. Laserpitium latifolium, Campanula persici­ ferential species are found in the field and bottom layers, fo lia andMelampyrum nemorosum), indicating fairly high of which the majority also occurs on Gotland. These often light availability in the interior of the stands. Others are show a southeastern and/or coastal distribution within acidophilous species (e.g. Hieracium vulgatum and Scandinavia, e.g. Carex sylvatica, Anemone ranunculoides, Vaccinium myrtillus) and 'true' forest species such as Hedera helix and Viola reichenbachiana. Among the Lathyrus vernus and L. niger. Besides, the mesotrophic bryophytes, Scleropodium purum and Brachythecium forests are differentiated by the absence or scarcity of velutinum can be named (Table 10, 75 releves). some species occurring frequently in elm-ash forests. 3. Quercus robur-Tilia cordata community (Mainland), The mesotrophic mixed deciduous forests (in the fol­ without differential species in tree and shrub layer. In the lowing called mesotrophic forests) form 20-25 m, excep­ field layer, however, some species can frequently be tionally 30 m, tall stands which are often structured into found which are either rare (e.g. Solidago virgaurea and two more or less distinct tree layers. Shrubs are usually Lathyrus montanus) or absent in the bland and Gotland

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 33

4 N � X

2

1

0

-1

-2

-3

AXIS 1 -4

- 3 -2 -1 0 1 2 3

Fig. 10. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 197 releves of mesotrophic mixed deciduous forests. Numbers of different sub-units plotted onto the sample plot positions which are slightly adjusted in order to avoid overlapping figures. 1. Quercus robur-Fraxinus excelsior community; 2-4. Quercus robur-Euonymus europaeus community (2. Tilia cordata sub­ community; 3. Oxalis acetosella sub-community, Corylusavellana form; 4. Oxalis acetosella sub-community, Carpinus betulus form); 5-6. Quercus robur-Tilia cordata community (5. Geranium sylvaticum sub-community; 6. Stellaria holostea sub-community).

stands (e.g. Calamagrostis arundinacea). The mainland 'Ekskog av ortrik typ' (particularly the 'Hassel-variant'), stands also have some differential species in common 'Lindskog' and 'Blandlovskog av ortrik typ' (Anon. 1984 ). with the elm-ash forests on the mainland, e.g. Dryopteris Some of the community types mentioned are more or filix-mas and Actaea sp icata (Table 12, 92 releves). less congruent with certain sub-units of the mesotrophic The differentiation as described above is also obvious forests as described here. The strong geographic variation in the ordination diagram (Fig. 10). The releves of the as expressed in the differentiation of communities has three communities are well separated. The division into also been pointed out by Kielland-Lund (1971). the lower sub-units will be treated later on in this chapter. Geographic distribution Affinities The mesotrophic forest type is the most common and The majority of studies describing deciduous forest veg­ widespread deciduous forest type within the Boreo­ etation in the Boreo-nemoral zone deal with mesotrophic nemoral zone, with its centre of distribution in the eastern forests, as can be expected from the wide distribution and parts of the area, i.e. on the Baltic islands of bland and commonness of these forests. However, the distinctness Gotland, and in the mainland provinces of Uppland, of this type of mixed deciduous forests has not always Sodermanland and Smaland, where most of the forests are been recognized, since many studies placed too much located. Other investigated stands are located in emphasis on communities dominated by only one tree Ostergotland, Vastergotland, Blekinge and Bohuslan, as species. The existing classification schemes and general well as in SE Norway. Besides, mesotrophic forests can vegetation surveys for Scandinavia partly list more than be found in the SW Finnish archipelago (Skult 1956; one community type covered by the group of mesotrophic Hinneri 1972) and in areas adjacent to the Boreo-nemoral forests: zone: in the oceanic parts of W Norway (Aune 1973; Ulmo-Tilietum (Kielland-Lund 1971); Polygonato­ Fremstad 1979; Fottland 1980; Blom 1980), and in the Ulmetum, Pulmonario-Tilietum and Va ccinio-Tilietum southern parts of the Boreal zone in Finland (e.g. Tapio (KlOtzli 1975a); 'Rasmarkernas angslovskog' and 'Ek­ 1953; Koponen 1967; Makirinta 1968). Similar forests hasselskog' (Sjors 1967); 'Angsekskog', 'Avenbokskog', have also been described from the Boreo-nemoral zone in 'Lindskog' and 'Lonnskog' (Bergendorff et al. 1979); Estonia, e.g. by Linkola (1929, 1930), Lippmaa (1935,

Acta Phytogeogr. Suec. 80 34 M. Diekmann

Table 8. Releve table of mesotrophic mixed deciduous forests, Melica uniflora 43432 .5.4 ...2. . ...43 3. Viola mirabilia community. (See Table 3.) 1. .2312 .....3. . 3 ..444 . Quercus robur-Fraxinus excelsior Anemone ranunculoides .43 .. 4 ...4 .. 4. 3 ...45 . 4 Lathyrus vernus 3 .. 4.33 .. 3 ... . 2 ...... Running number 12345678 91111111111222 22222223 Veronica chamaedrys ...1 ...24 ...... 3 ..2 .. 1 0123456789012 34567890 Ulmus minor 31212 ... Cluster Sa Sb Galium aparine .1...... 2 ...... 3131. .. Se Cluster code 1000000 1000001 100001 Anemone nemorosa 76677577 66657565666677 54655555 Cover % Tl (xlO) 54666466 66566556566643 65746666 Rubus saxatilis 32341232 44333224441.22 34334344 % T2 (XlO) 2-121111 22222211111153 13122232 Fraxinus excelsior 44 . 44223 223433334 .4533 43533333 % S (X10) 79886777 75887777656838 75775564 Geranium sylvaticum 4.123445 44343244433 .44 4 ..54444 % F (XlO) 65553364 78774867775435 76658868 Hepatica nobi lis .4.3.154 44445444444 .45 44444434 % B 111111-- 323-141111111- 124-2115 Ranunculus ficaria .5.4 4344 3.36.455544244 4544215. 000 0 0 0 0 050 0 Taraxacum officinale agg . 241212 .2 2221 .1.4.22.21 22412 .23 N - Tot 33332222 44332423333232 33325435 Quercus robur 211111 .. 2211221132 .11. 2.1211 .3 41107441 01908463336041 89572291 Viola riviniana 3334433 ..341 . 3143.1.2. 4332 .223 N - Vas 32222222 33332323223122 33224334 Ranunculus auri comus agg . . 4.44 .24 344414332 .2 .. 1 44352 .33 19952141 43607920992991 35875977 Geum rivale 44343... 4444 .4244.4.4. 253 . 3344 N Bry 322553-- 683-154344415- 547-7324

- Carex sylvatica 3.2.142 ..432 . 33242 ..3. 41414444 Deschampsia cespitosa 222 ..1.1 3332 .2.2212 .12 2.2.2322 ll Crataegus spp . 22 . 313 .3 222 .221 .32.2 .. .2222233 Betula pendula 33 ..34 .. Filipendu1a ulmaria 422 ..21 . 24 .314 .42543 .. 4.2.4444 Ulmus minor 446 .444. Viburnum opulus 3312 .322 22132 ....22221 2 ...111 . Sorbus aucuparia 2112212 . 1.12.2.1212 ..2 12 ..12 . 2 Quercus robur 55654556 44 . 54353554544 55354544 Corylus ave llana 22222122 .1121.112 .23 .. 121 . .... Fraxinus excelsior 4 ..55443 65655546445444 44445455 Orchis mascula 14 .3. ... . 44 .12.3 ..221 . 32 ..4234 Populus tremula .4 ...4.3 ..3 ..3 ...... 3 Malus sylvestris .11.111 . . ..112 ..22 .1. . 121 .2 ..2 Betula pubescens ...... 43 ...... 33 ... All ium oleraceum 342441 . . 3 ..4.2.2 ..2.1 4 23 ....2. Melica nutans 1.31. ... 122 ..3 ..2. 1.2. 12 . 232 .4 .I2. Vicia sepium 2 ...... 2 4333 .. 2411 .... .2.1.222 Betula pendula ..3. 33 .. Rubus caesius 22 ...... 142 . 4 ...4. 4 .. .32.2412 Ulmus minor 4443444 . Listera ovata 23 .22. . . . . 3 ..4 ...3. 3 2. . 21. . 333 Convallaria ma jalis 224 ...4. .3 ...4 ..444 .3. 4 ....4 .4 Fraxinus excelsior 4.343343 44434434344434 34344434 Poa nemoralis .. .. . 1.3 .23.2 .. 12...... 123 .11 Quercus robur ...3 .... 34 . 333 ..43 ...3 .3.33 ..4 Brachypodium sylvaticum 2.3...... 12.2 ..1...... 333 .2 ..3 Malus sylvestris 4 ....4 .. 3 ..3.3 3 .....3...... 33 Cornus sanguinea ...1 .... 1.1.2 ...2 ...... 3.1 222 . Sorbus aucuparia 4.43 .... . 3.33.3.3 .... . Epipactis helleborine 11 . .1.1...... 2.2 .. 1. ...1.1 Acer platanoides 3 ..3 .....54 ...... 4. Geum urbanum ...4 ...... 2 ..1 ..1 .2322 .2 . Populus tremula ...... • 3. 3 .....3 Fragaria vesca 1 .....2. .22 ..2 ...... 1. . 2 3 Dactylis glomerata .. 3 ...... 1..... 11 ...... 2 .2 ....•2 .s. Melampyrum pratense ...... 1 .....1 ...23 ...... 42 .1 Quercus robur 322 ..3 ..4 ...... 23 .. 3 Rosa spp . ..1 ..... 11 ..1 ...... 1 ...... 1. Acer platanoides 2 ...3 ...3 3 44 ...... 32 Allium scorodoprasum ...2 ...... 3.4.2 .....4 ...... 2. Ulmus minor .• 4443444 . Populus tremula .2 .....2 ..12 .2 2 Maianthemum bi folium ...... 1 ...... 2 ....2 .. . 1••••...... 2 .. Corylus ave llana 67666666 65666666445636 54554544 Ranunculus acris .1 ...... Crataegus spp . 44443222 3442442 .4.2433 233 .3344 ! ...... 1 .. Scorzonera humi lis ..2 ...... 1 ...... 2 ..2 Fraxinus excelsior 32332 .34 ..442 . 32444 22 . .4453334 All ium ursinum .....7.5 5 ...... • Sorbus aucuparia 4.442222 3432432 .4.43 .3 222 .22.4 Dactylorhiza fuchsii .1...... 1 ...... 3 Malus sylvestris 2343 .... 2.2.2.4.32232 . 23234344 Carex montana ....25 ...2 Cornus sanguinea . 232... . 3.2332 ..34 .... .422344 . Acer platanoides ...... •.223 Viburnum opulus 3 .222233 2 ..3.2 ...2 .... 2 .....2. Populus tremula .3... 3 .2... 2 ...... 3 Lonicera xy losteum 2.2.2 ... .li Thuidium tamariscinum 122 ...... 1 ... . 1.3. 1. .2 Juniperus communis 2 ....32 .. Fissidens taxifolius .2 ....1. .1.21. ...31. . Picea abies ....••• 2 2 4 ...... 1 Ulmus glabra ..•. ...•....4. .... 22 .. Plagiomnium undulatum 2 . 1212 .. 435 ..224122 .2. 344 .2 ..3 ...•...... Eurhynchium hians 1 ..212 .. . 2 ...113 1.2.3. 344 .321 . E Rhyt idiadelphus triquetrus ...2. 1 .. 543 .252.3.2.2. . 4 ...2. 5 Galium odoratum 4.544 . . . . . 5 ...... Eurhynchium striatum 2 ..13 ...... 1. 2 .4.. 4. 4.4 ..... Paris quadrifolia 2 ...11 .11.31.. 322213 .2 Brachythecium rutabulum 2.2 .... . 11...... 24.2.1. Elymus caninus ...... 2 .22.22 .22 ...2. 343 . 3432 Scleropodium purum ....1 .. . 1 ...... 3 .1. .1 Primula veris 1. .. 1.31. .1. 423 . 4312 .1

Additional species (occurring in one or two releves ):

2X: (T1 ) Ulmus glabra 21:4, 29 :4 (T2 ) Ulmus glabra 21:4, 28:3 (S) Prunus padus 9:2, 22 :4 (F) Acer pseudoplatanus 19 :2, 21:1, Alchemilla vulgari s agg . 9:1, 30:1, Bromus benekenii 9:4, 12 :3, Carex digitata 9:2, 27:1, Filipendula vulgaris 27 :1, 30:2, Hedera helix 21:1, 29:2, Lonicera xy losteum 1:1, 5:1, Luzula pilosa 7:1, 24:1, Platanthera chlorantha 4:1, 27 :1, Ribes alpinum 2:1, 22:3, Stachys sylvatica 2:2, 19 :2, Ulmus glabra 21 :3, 30 :1, Valeriana officinal is 14 :1, 30:3 (B) Brachythecium velut inum 2:1, 17 :1, Cirriphyllum piliferum 9:3, 25:2, Mnium hornum 4:1, 5:3. lx: (T2) Betula pubescens 18:3 (S) Acer pseudoplatanus 29 :2, Betula pubescens 3:2, Frangula alnus 18:2, Ribes alpinum 22:3, Rosa sp . 27 :2, Sorbus intermedia 28:2 (F) Angelica sylvestris 19 :1, Bromus ramosus 27 :3, Campanula trachelium 24 :1, Carex pallescens 28:1, Carex tomentosa 27:1, Deschampsia flexuosa 3:2, Dryopteris filix-mas 30:1, Galium boreale 30:1, Geranium sanguineum 30:1, Juniperus communis 9:1, Laserpitium latifolium 13 :2, Lathyrus montanus 17 :1, Lathyrus niger 26:3, Mycelis muralis 19:1, Neottia nidus-avis 11:4, Oxalis acetosella 11 :4, Picea abies 23 :2, Potentilla erecta 30 :1, Prunus padus 22 :3, Pyrola minor 7:3, Ranunculus repens 11 :1, Rhamnus catharticus 20:1, Sanicula europaea 26:2, Solidago virgaurea 13 :2, Taxus baccata 8:1 (B) Atrichum undulatum 3:1, Ctenidium molluscum 14 :2, Eurhynchium praelongum 23 :1, Homalothecium lutescens 10 :1.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 35

1940), Kalda (1960) and Riihl (1960). Northern outposts, almost all stands on bland and Gotland occupy level floristically impoverished, also occur on warm, steep, ground, whereas most mainland stands are found on slopes. often south-facing slopes north of the 'limes norrlandicus' The releve scores on axis 2 (eigenvalue 0.208) show in Sweden (Du Rietz 1917; Selander 1955; Sjors 1967). strong positive correlations with all three edaphic factors, Within the Nemoral zone, mesotrophic mixed deciduous i.e. the indicator figures for M, R and N, as well as forests hardly occur, because corresponding sites are usu­ negative correlations with the physiographic variables. ally occupied by forests dominated by the highly competi­ The correlationsof the L-figureswith the releve scores on tive Fagus sylvatica. An exception are the areas outside both CA axes are weak, indicating that light is of minor the natural main distribution area of Fagus, like in NW importance for the differentiation of mesotrophic forests. BohusHin and S Norway (A. Bj!('lrnstad 1971). There is a weak correspondence between the climatic The geographic and environmental distribution of the variables and the corresponding T- and C-figures. mesotrophic forests indicate that the communities in ques­ tion are thermophilous, i.e. that they are confined to areas Dynamics or local sites with a favourable temperature regime (cf. Like the oligotrophic forests, the area of the mesotrophic Kielland-Lund 1971, 1981; Fremstad 1979). Consequently, forests has decreased considerably. The decrease has prob­ several of their differential species are thermophilous, e.g. ably been most pronounced in the eastern parts of Swe­ Tilia cordata and Lathyrus niger (Oberdorfer 1983). In den, where the mesotrophic forests generally occurred on the eastern, relatively warm parts of Sweden (bland, level ground which could easily be transformed into Gotland and the easternmainland), stands are found both wooded meadows, arable fields or pastures. Where the on level ground and on slopes of varying aspect and stands mainly occurred on more or less steep slopes, as in inclination. In western Sweden, the forests are more often the western parts ofthe area, the influencewas less strong found on slopes. Finally, the stands in western Norway, and consisted of extensive cutting and grazing. However, with a relatively oceanic climate, are confined to fairly in many regions, the area of mesotrophic forests has steep, preferably south- to west-exposed slopes (Aune increased again since the last century in relation to a more 1973; Fremstad 1979). intensified agriculture, followed by a concentration of arable land to areas with the best soil properties, and due Environment to the cessation of mowing and later also grazing within As can be expected from the large floristic variation of the wooded meadows and pastures, which then became more mesotrophic forests, there is a large variability with re­ closely wooded. The fairly diverse tree species dynamics spect to environmental conditions. This concernsthe topo­ will be dealt with in Chapter 5. graphic location as well as soil type and chemistry, which differ between sub-units. However, among common fea­ Quercus robur-Fraxinus excelsior community (Table 8) tures can be mentioned the following: usually a ranker, rendzina or brown earth as soil type, a fairly good base The Que reus robur-Fraxinus excelsiorcommunityshows saturation and nutrient supply and medium to high pH­ little structural and floristic variability. Almost all stands values (e.g. A. Bj!('lrnstad 1971; Aune 1973; KlOtzli 1975a). are dominated by Que reus robur and Fraxinus excelsior. This is also indicated by the N- and R-figures in Table 2, The shrub layer is very dense and consists mainly of which are much higher throughout than for the oligo­ Corylus avellana. Other frequent woody species are trophic oak forests. Crataegus spp., So rhus aucuparia and Malus sylvestris. Table 7 shows the correlation coefficients between the Both field and bottom layers are quite species-poor com­ releve scores on the first two CA axes and explanatory pared with the stands of the Quercus robur-Euonymus variables. The releve scores on axis 1 (eigenvalue 0.290) europaeus community. This may partly be explained by are positively correlated with both physiographic vari­ the fact that, on Gotland, several deciduous forest species ables (inclination, heat index), climatic variables (yearly have a very restricted distribution (e.g. Mercurialis precipitation, de Martonne's index, continentality index, perennis and Cardamine bulbifera) or are completely latitude) and soil moisture (M-figures). Strong negative lacking, such as Pulmonaria offi cinalis and Calamagrostis correlations are found for R- and N-figures and for the arundinacea (Hulten 1971). The stands are located in variables connected with temperature (annual mean tem­ different parts of the island, with a slight concentration to perature, mean temperatures for February and July), as S Gotland. well as for longitude. It is obvious that axis 1 mainly Two sub-communities are distinguished: a Typical expresses the differentiation into the three geographically sub-community (releves 1-8), without any good differential separated communities. As for the oligotrophic oak for­ species, and a Paris quadrifolia sub-community (releves ests, a complex-gradient in nutrient status and a moisture 9-30) with a group of differential species, some of them complex-gradient can be recognized. The high correlation indicating a high soil-pH, such as Elymus caninus and coefficient for inclination is explained by the fact that Anemone ranunculoides. The latter sub-community can

Acta Phytogeogr. Suec. 80 36 M. Diekmann

Table 9. Que reus robur-Fraxinus excelsior community. Number of vascular plants (Vas), number ofbryophytes (Bry) and total numbers of species (Tot), as well as indicator figures for light (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for different sub-units. Average values are given. The sub-units correspond to: 1. Typical sub-community; 2-3. Paris quadrifolia sub­ community (2. Typical form; 3. Ulmus minor form).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 24.6 2.6 27.3 5.4 5. 1 3.8 5.4 6.2 5.3 2 29.2 3.4 32.6 5.3 5.1 3.9 5.4 6.5 5.4 3 36.0 4.0 40.0 5.3 5.2 3.9 5.3 6.6 5.4

further be divided into a Typical form (releves 9-22) and Tilia cordata sub-community (releves 1-46) an Ulmus minor form (releves 23-30). The indicator fig­ (Mesotrophic mixed deciduous forests with linden) ures for the three sub-units are very similar for all factors The mesotrophic forests of the Tilia cordata sub-commu­ except R which is clearly higher for the stands of the Paris nity are characterized by a mixture of different tree spe­ quadrifolia sub-community (Table 9). cies in the canopy (Fig. 11). Most frequent in the upper Hardly any literature dealing with deciduous forests tree layer are Quercus robur and Tilia cordata, whereas on Gotland is available, although many studies and inven­ Acer platanoides is particularly frequent in the lower tree tories have been made of wooded meadows, e.g. by layer. Tilia and Acer serve as differential species. The Pettersson ( 1946) and recently by Borgegard & S. Persson cover percentage of the shrub layer varies from less than ( 1990). Wooded meadows once were very common and 10 % to 80 %, mainly depending on the cover of Corylus widespread on the island, but have largely been aban­ avellana. Several species in the fairly dense field layer are doned in the course of the last two centuries. However, the confined to the Tilia cordata sub-community, e.g. majority of wooded meadows still existing in Sweden are Campanula trachelium and Lathyrus niger. The cover of nowadays found on Gotland. The human influenceon the bryophytes varies considerably, from 0 to 50 %. In some landscape of the island has been so strong that no prima­ releves, more than 10 different bryophyte species were rily natural or semi-natural forests can be found any encountered. longer. All the stands studied have originated from wooded Two forms can be distinguished: a Filipendula ulmaria meadows and still show signs of the former land use, both form (releves 1-20), differentiated by Ulmus glabra and in structure and floristic composition. This strong human some species indicating high soil moisture (Filipendula impact on the forest vegetation is probably another reason ulmaria, Angelica sylvestris and Festuca gigantea), and a for the relative species poorness of the existing stands. Stellaria holostea form (releves 21-46), differentiated The scarcity of Acer platanoides and particularly Tilia only by Stellaria holostea (including a Milium eff usum cordata in the mesotrophic forests on Gotland is, accord­ sub-unit and a Dactylis glomerata sub-unit). ing to Pettersson (1958), due to human influence. As so Mesotrophic forests of the Tilia cordata sub-commu­ many wooded meadows have been abandoned, different nity have often been described from Gland, but they have successional stages of mesotrophic forests can be seen at not been clearly separated from the forests belonging to many places. the Oxalis acetosella sub-community. Studies were, for instance, made by Sterner(1 926), Tiixen (1951), Sjogren (1964; 'Poa nemoralis-Foderation') Ekstam & Sjogren Quercus robur-Euonymus europaeus community , (1973, Betulo-Quercetum melicetosum), Ekstam (1979, (Table 10) 'Filipendula ulmaria-Geum rivale-Rubus caesius-typ' and The Quercus robur-Euonymus europaeus community on 'Poa nemoralis-Melica uniflora-typ'), Ekstam & Martins­ Gland surpasses in species richness all other deciduous son (1981) and 0. Johansson (1982, 1985, 'Blandadel­ forest communities of the study area. They show a consid­ lOvskog'). The forests are concentrated in three areas: erable variability in structure and composition. In the Mittlandsskogen (the largest deciduous forest area in following, three sub-units will be treated separately in Sweden south of the Scandes), the coastal plain at the some detail: Kalmarsund south ofBorgholm, and northernmostGland. - Tilia cordata sub-community, Mesotrophic forest communities dominate the deciduous - Oxalis acetosella sub-community, Corylus avellana form, forests in all these areas. - Oxalis acetosella sub-community, Ca rpinusbetulus form. The stands are usually found on level ground and occur only exceptionally on slightly sloping ground. The bedrock in the Mittlandsskogen area and in northernmost Gland consists of limestone, in the coastal plain of

Acta Phytogeogr. Suec. 80 Deciduous fo rest veg�tation in Boreo-nemoral Scandinavia 37

Table 10. Releve table of mesotrophic mixed deciduous forests, Quercus robur-Euonymus europaeus community. (See Table 3.)

Running number 12345678911111111112 222222222333 33333334444444 4445555555555666666 6666777777 01234567890 123456789012 34567890rL23456 7890123456789012345 6789012345 Cluster 6 7a 7b 8 9 Cluster code 1000100 10001010 10001011 1000110 1000111 Cover % T1 (X10) 67876767768757886378 787776676765 67767566783476 5344465654465765566 8989927887 % T2 (X10) 24457441312232121131 324351221523 54135645!144324 323341114-1111261-2 1--1-7--12 % S (X10) 55542348472384158865 558448446288 11772345235422 8978668988899985778 3223411212 % F (X10) 78888888999889886789 897999898988 47677889787887 7888687887878778667 7887869876 % B 22111523542452111322 551152111112 -2111---111113 111111111121122151- 2211113113 0000 000000 0 0 0 0 0 0 0 0 5 050 5 0055 005 0 N Tot 44555565655654434544 553444343444 23333333453454 5323333334544454454 5665544434 - 38096513831239565678 047384472346 75116273315326 8671754940003442651 2361238496 - Vas 34454454544544334444 443343342343 22233333333344 4332333333323343344 4554433333 N 82316827363352930955 366765249905 77402273791881 9557141834882551561 4252836284 N - Bry 56781796178987635723 781629233441 -8714---614545 912461311611199199- 8119411111 0 5 1 2 221 1 11 022 2 n Ulmus glabra ..2 ...... 2 .....4.2. 4 ...... 2.3 ...2 .2 ...... 2.

. Tilia cordata 33542 .424 . 65456 .2 ... 45 . 33344 ... . 4542542 .55 ..44 ...•.4 •..•

Fraxinus excelsior 243444565544 .4242254 424552 ...... •.••.•2 •••.•.. 53 ...... • . . 4 ..•.32

Acer platanoides . 2 ..22 ...... 35 3 .25 ..22 ...2 .... 2 2 3. . ..•.....2 Carpinus betulus ..•... .• .• 67767 .6766

Quercus robur 66555532442345335454 526555556666 55665556555555 3455566655565666566 4 ...24 ..2.

Betula pendula 2 .....2 .. . . 2 ....2 ...... 4 ....2. 2 ...2 . •..•.•.••••••••2 ••

· ...... •

Ulmus glabra 42333 ..222 4233 .44243 ....2 .....2. • . • . • • • • 2 .•• 4.24 5.2.5.3 ....5.4...... 3 .

. • Tilia cordata 255355534 .44343 ..2 .. 44 . 45423 .43. 5444553 554 .234 •...2 •.•••••••••••

• Acer platanoides . 444544242 ...433 . 222 22543 .44242 . . 4 ..4544 . 45444 .2 .....2 ...•..•....

Malus sylvestris .•..2 .... .•..•...•...... 2222 . 4 ..3 ..2 ..223 .. .44422 ..2 .... 2 ..222

Fraxinus excelsior . 2 ..2 ...22 . 3 .....22 . 3.222 .2.223 ...... •..2 ..•..•....

Carpinus betulus ••..•2 •.••. .•...... •..4 ..•. ...4. 6 ..34

Sorbus aucuparia . • . • • . • . . • • . 2 •.••••. ..2 ..23 ...... 2.42 .4 22 .2 ..4 ..23 .2 ..34 2 ...34

.•. Quercus robur . 2 ...... 2 ...... 22 . . 22 ..2.3 2 . . . ..2 ..22 42 ...... 23 .. Sorbus intermedia ...... 23 . . . •...... 2...... 3 ..2 ....2

...... • •.. Betula pendula ...... 11.1...... 22 .

Ulmus glabra . 1423 . 22233 .34134333 ...... 1...... 1 ...... 41 . .22.44 ...... 4.2 . 1121. ...22

. Cornus sanguinea . 221231234 ..21. .4.25 22 .3 ...... 1 ..1 111.1.. 3 ...23 .4. ..•.4 .•.•.

Tilia cordata .32322434444332 .11 .. 24132441 .1.. 3353431444 .232 ...... 11 Acer platano ides .2.334311 . ...21 3112 . 2353212 .12 .. .32.4 .24.23341 ...... 1 ...... Corylus avel lana 5535 ..46353 . 6424565 . 546556555566 32562255444424 5655436565566665666 Lonicera xy losteum . . 43142333 ...43 1. .4. 2 . 1443112 .4...... 1. 4 ..2.1 455555554554552 ..4 .

Malus sylvestris ....1.1 .... 1 ....1 ...... 11 ..2 3 •. 4 •• 1. ...2 .312.2 ..1 .....1. 313

•• Rosa spp ...... 1 ...... 21 ...1 ...111 ...... 1.1... . . Juniperus communis .2 ...2 ....1 ...1 ... 211.1. .. Carpinus betulus .....3 4 ..2. . 34132 ... 4.2 .

..•..... • ..•.••••.....

Crataegus spp . 43433442221422154234 32333 .1.4.33 4.34244342442 4 4434443443344343343 44 . 4443433 Ribes alpinum 1.12221 .11. .1. ..11 ...... 4 ....2. 12 .12231 . 23341343 .311 21111 . .... Sorbus aucuparia 1.1121 .221 . 112 2112 ..21 ..3 1 1.1...... •...3 ..42 ...... 1.2.312 ...... 222 12 ......

...•.. Fraxinus excelsior 1142 ..323 2.4 .41.2.22 35.32 ...2.3. . ..•.2 1 53 .32 ...... 3 ..1 23345 ..1. . 1. Euonymus europaeus .....1 .. 13 ...... 1 ....11 ...... •.... 11 .11 ...... 1.2 . ...1 ..... Viburnum opulus 2 ....1 ..11 . . 1 ...... 2 .....2 ..2 ...1 ..... Frangula alnus ...... •...... 21 . . . 1 ...... 24 .1...... Populus tremula ...... 1 ...... 2 ...... 1 . ... . • ..3.

Sorbus intermedia ...... 2...... 1. 2 ...... 1 ......

.

E. Filipendula ulmaria 4 ...3 . 14344 ..31 .1.21 ...... 1 ...... •..•..•..•.44

Dactylorhiza fuchs ii ...... 213.21...... 1 .•....•..•. ..•.•2 .• . 112 ......

Scrophularia nodosa 1 ...... 2 .2 ...... •.2 Festuca gigantea ...2 ....22 ...... 2 .. 1 ...... 1 . 23 ....1 .. Angelica sylvestris .....1111 ...... •...

Viola mirabi lis . 244244313 ..22321 244 44413 ...... 32 4 ••••••••••••••••••

. Mi l ium effusum 1144 .2 ..22 .3424 .13 .. ..21 . 121 .23 ...... 2. 1 2 . . . 3 ......

.....•...... Paris quadrifolia 421. .122123131 . .13.2 ...42 ....22 . • . . • . • • . . . . . • 4 .2.31 . ... . Polygonatum mul tiflorum ...1222113 ...... 12 ....2. 1. . 1 ...... • ...... • . • • .

Melampy rum sylvaticum ...12 ...1. 34421.24 .. . 2 ...... •.•3 ... . Orchi s mascula ...... 1 .2113 ...2 .... . 4.21 ...... 1 ...... 1 ...... 2 Ti lia cordata .21.1.122222311 .12 .. 21.3143 .... . 1241312421 ..34 ...... •..1 Campanula trachel ium .1.11124 . 11112131212 2 ...2 ...... 1.1.1. . ...2 ....1

.

Acta Phytogeogr. Suec. 80 38 M. Diekmann

Table 10, cont. Lathyrus niger . . 1.2... 1 .... 1 ...... 2 .1.121. .. . . 3 3 142 . 2 3 . 2 4 4. .22 ......

Laserpitium latifolium . . . 113 ..1 ....23 ...... 14.212 ...... •...2 .•5.

Lathraea squamaria ...1 ...... 1. 2 ..1 .. • . 3 . ....1 ...... Stellaria holostea .3...... 2 ...... 4452344• • • .•..•.44443 145 . 44454444 .4 ...... 4 .....4. 4.1 .....44444 Dactylis glomerata ...... 23 .2...... 11 .1.31112121. .1... 3.3 ...21 ...... 123 . 133432 Sanicula europaea ...... 1 ...... 11 .2 .. 1. .2 12 .....1 .....2 ..... 344332311 . Oxalis acetosella 2 ...... 1. 44 ...... •..•.22 ...... 4422 .1.4444.5 4344 .22111 Dryopteris filix-mas ...•••••••.•••••••••••1 ...... 2 ...... 2. 2 . 1 .. 42 .. 1.1. Rubus fruticosus agg ...... 1 ...... 1 ...... 121 3 .2 .. 1...... 2 ...... Hordelymus europaeus ..224 ....4 ...... 23

Prunus avium • . . . . 1 2 ..2 23 .....3 .. Carpinus betulus .....3 ...... •.• ..•.•• 242 ..1342 . Galium odoratum .....21 424

Anemone nemorosa 55555554415555454455 445656455656 55664455445554 4365555556455544545 6455456555 Hepatica nobilis 34444443424443354544 445434443333 13444233442244 4424412443232344333 4444522223 Crataegus spp . 322122211 .2212332223 221221213 .22 122323 . 2232212 2211 . 22331323232 .21 2322222243 Poa nemoralis 12332532344344513424 453132433121 24335444441134 1. ...41 44 . 343242331 3443 .23311 Ranunculus auricomus agg . 434113 . 2242322352224 134333 .24.23 22444 . 444412 .4 4153 532244 .4.441214 2122332334 Anemone ranunculoides 45444444424344444444 454443445455 54345345554444 3.4 .1. 3. 42 ..2. 4. 3 . 4 4444453552 Cardamine bulbi fera 132334211 ..3242 . 443 . 43343434 .44 . 344 . 3434344444 .2333 .44432334224 .1 2321 . 44333 Fraxinus excelsior 34434423432424414334 444443422241 .1. ....323 .213 44223 .1.3.2.1. 244 .. 24354 . 2223 Ranunculus ficaria 54 ...3 ...444334 . 1344 3.3354.45.44 555555555454 .5 3346 . 45455 .3.54.5.5 5542 . 55545 Convallaria majalis 45335255442142 ..211 3 424424454544 . 4444 .3344 .. 42 2 .....2443545 . 25564 .21 ...... Mercurialis perennis 2634 ..42 .1.. 22 452444 444771266642 42243453351123 41663 ..4.4 ...... 2.1 .....12411 Melica uni f lora . 121 ..3 ...214 . 244444 . 4211 .2.2 ... 233425 . 43445 .. 44143 5654455524.32 . 4.3.544344 Viola riviniana 4.2.123 ...23223 ...12 31.22232 ..31 23 . 32334 .2.. 22 4 ...42234 .233 .24444 44444 .332 . Acer platanoides 14223433 .2 ..224 43243 2243233313 .. .4223244234444 .11. ..1. .1. .1. .1. .. .11. .44222 Geranium sylvaticum 3232333221. ....4 .... 431.2.14.234 ....33 .22 . 4412 1.213 ..2231 . 1312 .1. .42.2 31 .43 Vicia sepium ..22 .3.12234323114 .4 .1. .11. .1..2 ...2. 3221.24.1 22 .11 3344313212 .. 1. 2 ....33432 Corylus avellana 12 ...12 .121.1. .11. .. 11 .212.13 .. 3 1231. .2.211.2. 212.2.322 . 332332422 3332 ...... Primula veris . . 123222122323343133 34 .12 ..1 .... .1. .12.2 232222 32 .....2.2 2 ..1 ..... 232 .2.1143 Sorbus aucuparia .2122 ..2.1. .111. ..1. 21121 .4 ..13 . ..211 .22 ....11 1121 .11.2112 .2224 .3 131 .1. .1.1 Fragaria vesca .11.422 .1.2311 312 .21 2 ..2 ..221 .32 ..1 ...... 11 . . ...122322242222 .. 1 12444.11. . Carex sylvatica 232 . 332454221323 . 214 ...1. .1. ..12 ...... 1 ..11 ...... 21 1224 .42 2414122421 Geum urbanum ..3 ..3 ....12 ..13 ..33 34 ..3 ..14 ..1 2142 .21 33 .12.1 42242 . 2322 ...11 .... 34 ...23411 Geum rivale 523244456643461 .4445 43 .22 ..22244 .....2 ..21 ...31 4344 1 ..42 ..... Gagea lutea . 221 . .11...... 124342 444 .2 ...423 . 44 .. 3423333 .3. 3 ...... 1 .... . 3434334434 Viola reichenbachiana .3.1.1...... 322233 .2 2 .. 1.22.2 3.3 .. 3 3234 ..334 4 ...4.3.1 . 3333244442 Melampyrum pratense .1.24 . 3.2 ..123 .. 112 . ..• 2.. ..•..•..353 .4 ••.. . 2 ..142 . 44 ..44 ...... 22 ....2323 .. 244.4.32 .. Pulmonaria officinalis .2 ...41 1.1. ....1. .32 312141123311 .1122 . 2.44 ...4 .1..... 3 .1...... 1 ...... 2. 31 .

Ribes alpinum ..1.1 31 .1. ..1. ..11 . . ..111 3 ....1. .•••.•.••.••.2 . 133412334323431 ..1 21211 . .... Brachypodium sylvaticum .3133 . 33234 .1... 1.2. 1. . 3 ....11 ...... 1 ...4 22 ...1.3 ..2.1. .1. .. 2444411.22 Quercus robur .1 ..1 .....12 ....1 ...... 1 ..12 1 2 .1.1...... 1 . 11 ..1411 .11 21.2.111 1121 .12 ..1 Al l ium oleraceum ..2131 ....441332 21.1 3.2 .43.33 .33 4 ...2.1.4 1.1.1 2121.2 ...... 3 Rubus saxatilis 2 ..143 .34 . 34222 ....3 1.14.444 . 443 ..11 .. 1.32 .. 12 ...... 1 ...214 . 2 ...•.... Euonymus europaeus 1. .1222212 .1...... 11 1.1112.1. ..2 22 ...... 111 112 1.1•..22 .1•.••.11 . Deschampsia cespitosa 321 . 322332231 ....121 11.2.1..2 .34 ...1 ...1.2 ...... 1142 1.2.1. ... .

Hedera helix . 342 .34.1... 24 ....22 . 2 ..•..•..•.22 .. 45 . 5321422 .44544 .4. 424 .3.... . Lonicera xy losteum ..2 ..2. 21. ...2 .....1 •.... • • 111 ..••.•.. .11...... 1.2 .... 1 223323333213231 .1 ...... 1 .. . Ulmus glabra ..213 .11.22.22.22222 ....1. .2 .... 3.3. 2 ..44 .1. . 114 ... . 2.12 ....3 Viburnum opulus 22111 .212 ...... 1.. . .1.1...... 1 ...1 ....21 ..1. . 12 .....11 22 ..14143 ....1 .... . Rubus caesius 31 .12 ..31 ...... 1 1. ..32 ...33 . 2 ...1.1 .....12 .4144 22342 .... . Cornus sanguinea 1..1421.23 ..22 ..2.1 4 2 ..12 ...... 2.112 ....13 . 2223 ...... 1 ...... Lathyrus vernus ..1.2 ...1 ....1 ...... 4.1 .. 2 ...1 ...143 . 4331242 ...... 4 .....3 .... . 3 ...11 233

Elymus caninus 13 .244444444 .....24 . 242 .1...... 1 ...... 1 ...1. 4 2 .•2 ..1. 4 .... . Galiurn aparine ...... 11 ...... 124 14 ..1. .13 ... 41 .21 .4312 .2 .. 41.•.••.••..•..• .....2 42 ...... 3 .. . Veronica chamaedrys ...... 134 .22 .2.2 ...... 21 .1. . 2 .....11 ....13 ...... •..•.2 ..... 2 .. ..11 21 .11. Aegopodium podagraria ..15 .42244 ...... 234 .2 ..42 .2. . ..3 ...... 34 5.5. 4 ....1 Carex digitata .3212 .21. .1.1. ...1.. 2 ...... 2...... 2.1.2 3.3. 33211. .... Me lica nutans ..1 ...3 ....2 ...... 1. 2 ..2.1 2 ...... 1 ..3. .1. ..2.1...... 32 .1 1.234 ..... Rosa spp ...... 1 ...1 ...... 1 ...... 1 .. 1 .. 1 ...... 1.1 ... 1 .. . . .11 111111 Anthriscus sylvestris ...1...... 11 . 3434 .2 .2 ...... 111 .....12 1 ...... 3 ...... Listera ovata 1 ...... 12 . 1.1.1. 2 .1...... 11 ...... 1 ...... 1 .1.11. .... Malus sylvestris ..2. 112 ..1 .....1 ...... 1.1 ...... 2 .....1 .... 1 ...... 12 .1 ..1 ...... Hieracium murorum ....2 .....22 ..1 ...... 1 ...... 131 2 .2343 ....2. 3 ...... Maianthernum bifolium .1. ..24323 ..4 ...... 1 ...... 1 ...... 1 ...... 2 .21 ...3 ...... Brornus benekenii ..2.3 221 . .3 ...... 2 .. . 2 ...... 21 ...... 1 ..2 ...... 21 .. Taraxacum officinale agg . . . . 12 ....1 ...... 12 . • • . • • • • . • 1 ...... 3 ....111 . ..1.1. ...2 1 ...... Juniperus communis ....1.1 ..... 1 .....1. .1...... 1 ....1 ...... 1. 11 . .11. ... Adoxa rnoschatellina ...... 22 ...... 1 .... 1. ...21 1. ...1...... 3 .... . Epipactis helleborine . 2 .....1 ...1 ...... 1 2.211 . ....

Ulmus minor . 2 2 2. 23 ...... ••2 .•.••.

..•...... •. Luzula pilosa . • ...... 22 ...... 1 ...... •..2 ...... 1.1 ... . Campanula persicifolia . • .1 ...... 1 .. 1 ...... 1. .12 ...... Melampyrum nernorosum . . . 4.34 ..1 ...... 4. Viola riviniana reichenbachiana ...... 1 ...... 2 ...44 .1. x Ranunculus acris ...... 1...... 1 ...... 1 .... . •.••••.•• ...... 1.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 39

Table 10, cont. Geranium robertianum ...... 1 ...... 1 ...... 1 ..1 .... . Popu lus tremula ...... 3 ...... 1 12 Carex muricata agg . 1 .. 1 ...... •...... 1 ...... 1 .....

. . .•.

.6. Eurhynchium hians 13444445442324241434 44 . 441 .23211 . 2 ..1. ..34111 . 411.2.1 ...31 433314 . 4444 .34314 Plagiomnium undulatum 234423345543422224 .4 23 .11 1.1.222 ....1. ..11 ..23 ....1 ...2 ..3. 2 1413 . 3423 .444 .2 Brachythecium rutabulum ..2 .22122322 .11.13 .. 12 ...3 ..2.1 2 . 31 .1.1.221. .. 2 ...... 2 ..4233231 .. 22331111 .2 Plagiomnium affine ..12 .1. . 2.3334 ...3...... 11 ..2.1 . . 1.... . 12 ..22 ...... 12221 .213 . 3.2 ..333 .1 Brachythecium velutinum ...... 1. ....1. .12 ... 1.21.2 ..1.2. .11.2.... 31121 2 ...1.2 ...1.3 21.1 . .111 . 322.. Rhyt idiadelphus triquetrus ..21 2 .2. 434522 ...4.1 12 ....1. ...1 ...... 1.2 ...... 2. 2.2 2 ..4 . .3312142 .2 Cirriphyllum pi liferum . . ..2 ...... 31 ....1 .. 1 ....1 .....1 . 12 2 .2 1 ..11 ....1 .21221 ... 3432 .242.2 Fissidens cristatus ..121 ...1 ..1 ...... 1. .1 ...1 ...... 11 .....•.....11 ..... 1 ...... 21 3 .1 .. . 211 ...... Thuidium tamariscinum ...1 ...... 21 ...... 2 ...... 1 . 1 ...... 1 ...1...... 1332 . 1114...... 22 .2 Fissidens taxi folius . 1 .221442 ...... 1 ...... 1 ...... 21 ...1 ...... 1312 Plagiochila porelloides . 2 . . 1.1.1 ...1 ...... 2 ...... 1 ...... 1 . .121.1. ..1. Eurhynchium angustirete ..2 .....4...... 1 ...... 3 ...... 11 .3 ....4.1. 1. . ..1. .22 Eurhynchium striatum 34 .1 .....1 ..1...... 2 .1 .....2. 2 ...... •.4...... •.3 Eurhynchium praelongum ...... 23 .4.. 11 ...... 1 ...... 1 .....•. . . ..1. 3 .... Ctenidium molluscum 1 ...1 ..11 ...... 1 . ....1 ...... 2...... 1 Campy lium calcareum . . . . . 11 . 11 ...... 1 ...... 2 ...... 1 . . .1...... Hypnum cupressiforme ...... 1 ...1 ..1 ...... 1 ...... 211 ...... Atrichum undulatum 11 ...... 12 ...... 1 .... Plagiomnium cuspidatum 2 . .2 ...... 1 ..1.1 .. . Mnium stellare ••....•.. . . . 2 ..1.1 .••.• ... 1 .••......

.

Additional species (occurring in one , two or three releves):

3X: (T1) Populus tremula 45 :2, 46 :3, 63 :2, (S) Picea abies 61 :1, 67 :2, 69:2, Quercus robur 9:1, 15:1� 52 :2� Ribes uva­ crispa 50 :1, 51 :1� 54 :1 (F) Hieracium vulgatum 42 :1, 71:11 73 :31 Hypericum hirsutum 9:2, 10 :11 57 :1, Lathyrus montanus 28 :1, 34 :1� 39 :1� Mycelis mural is 45 :1, 47 :3, 57 :1, Picea abies 66:1, 67 :1� 68 :1, Platanthera chlorantha 67 :11 69:1, 70 :1, Stachys sy lvatica 1:2, 51 :3, 65 :1 (B) Climacium dendroides 18:1� 64 :1, 72 :1.

2X : (T2 ) Picea abies 21 :2, 66:2 (S) Prunus avium 52 :1� 59 :1, Ulmus minor 1:4� 64 :2 (F) Actaea spicata 47 :1� 51 :1, Agrostis capi llaris 26:1, 28 :1, Allium scorodoprasum 11 :4, 16 :31 Bromus ramosus 32 :4, 57 :11 Clinopodium vulgare 68 :1� 69 :11 Deschampsia flexuosa 52 :11 58 :2� Dryopt eris carthusiana 1:1� 65:11 Fi1ipendula vulgaris 12 :1� 67 :11 Hypericum maculatum 12 :11 68:1, Hypericum perforatum 30:1, 61 :1� Neottia nidus-avis 12 :1, 70:1� Sorbus intermedia 67 :1, 68:1� Stellaria media 47 :1� 67 :11 Vaccinium my rtillus 59 :2� 66 :1, Viola hirta 47 :4, 50 :1 (B) Polytrichum commune 9:1� 59:1, Thu idium recognitum 10 :1, 70 :1.

1X : (T1 ) Malus sy1vestris 70 :11 Sorbus intermedia 57 :2, Ulmus minor 69 :2 (T2 ) Populus tremula 46:2, Ulmus minor 69 :2 (S) Berberis vulgaris 70 :11 Ribes rubrum 65 :4 (F) Alchemilla vulgaris agg . 3:1, Alliaria pet iolata 47 :1, Athyrium filix­ femina 65:1 Berberis vulgaris 70 :1 Betula pendula 69 :3, Calamagrostis ep�gejos 27:1, Caltha palustris 7:1, Cardamine amara 65 :11I Corydalis bulbosa 47 :31I Coryda lis sp . 48 :21 Daphne mezereum 9:21 Epilobium parviflorum 10:1, Equisetum hyemale 65 :1, Equisetum pratense 30 :2� Galium boreale 12 :1� Glechoma hederacea 48 :1, Gymnocarpium dryopteris 64 :2, Heracl eum sphondylium 20:1� Hieracium umbell atum 46 :1� Lapsana communis 47 :1, Lathyrus pratensis 57 :2, Moehringia trinervia 33 :3, Platanthera sp . 39 :1, Polygonatum odoratum 45 :2, Rhamnus catharticus 63 :1, Ribes rubrum 65:3, Rumex sanguineus 73 :1, Solidago virgaurea 45 :2� Tr ifolium medium 52 :2, Veronica officinal is 11 :11 Vincetoxicum hirundinaria 12 :1 (B) Amblystegium serpens 42 :1, Calliergonella cuspidata 32 :2, Dicranum scoparium 57 :1, Homalothecium lutescens 72 :11 Homalothecium sp . 50 :11 Isopterygium elegans 67 :1, Lophocolea bidentata 67 :1, Plagiomnium rostratum 67 :1, Plagiothecium succul entum 63 :1� Polytrichum formosum 73 :11 Rhodobryum roseum 68:2.

sandstones and shales poor in lime. However, the bedrock somewhat higher M-figure (Table 11). The stands of the is usually covered by morainic material, composed of Stellaria hol�stea form (mainly found in the coastal plain) calcareous tillsor non-calcareoustills containing siliceous show deepeli, less clayey soils with a lower and more boulders andstones. The soil type is usually a mesotrophic, constant grolmd-water level. loamy brown earth with a mull humus layer. pH-values Mesotrophic forests of the Tilia cordata sub-commu­ are usually not lower than 5.5 (Ekstam 1979; Diekmann nity would probably cover large parts of the island, but 1988) and are particularly high in the areas with calcare­ have to a large extent given way to arable land and ous bedrock. The stands of the Filipendula ulmaria form managed forests. However, in the Mittlandsskogen, the (mainly fo und in the Mittlandsskogen area and in north­ sub-community is expanding considerably, due to the ernmost Oland) grow on shallow and often clayey soils, abandonment of the pastures formerly widespread in this which are wet (sometimes even inundated) during spring area (cf. Sjqgren et al. 1974; Ekstam et al. 1984). As but often dry out during summer. As a consequence, the different parts of the area were abandoned at different lower soil strata sometimes are developed as gley-hori­ times, several successional stages from young secondary zons (Diekmann 1988). This explains the occurrence of forest to almost mature forest can be fo und. some moisture-demanding species, also expressed in the

Acta Phytogeogr. Suec. 80 40 M. Diekmann

Oxalis acetosella sub-community, Corylus avellana form the often widely spaced oaks with broad crowns, indicat­ (releves 47-65) (Oak-hazel forests) ing that the trees did not grow up in dense shade in The mesotrophic forests of this form represent the oak­ competition with other trees. Also the high abundance of hazel forests as described in the literature from S Sweden shrubs, particularly of Corylusand Malus, is due to former by, e.g. Selander (1955) and Sjors (1967) (Fig. 12). The land use. In a first stage of secondary succession after tree layer is highly dominated by Quercus robur, the abandonment, shrubs increase in cover, to eventually shrub layer is very dense due to the high abundance of decrease again due to outshading by trees. However, oak­ Corylus avellana. Other frequent shrub species are hazel forests may persist for a long time, since Corylus Crataegus spp. and Lonicera xylosteum.The fieldlayer is creates very shady conditions and hampers the regenera­ comparatively species-poor and has only a few weak, not tion of trees (Lindquist 1938; Anon. 1982; cf. also Chap­ constantly occurring differential species, namely Oxalis ter 5), whereas Quercus is very long-lived. acetosella and Dryopterisfilix-mas, both indirectly fa­ At many places, Corylus forms groves without any voured by the dark interior of the stand caused by the overstorey of trees, especially in the Mittlandsskogen dense shrub layer. The shade-tolerant Hedera helix is very (Ekstam et al. 1984; Stemer 1986). These hazel-groves frequent and often seen climbing up the oak stems. are bound to stony, shallow and often fairly dry soils, Oak-hazel forests from bland have been studied by preferably on calcareous bedrock. Their floristic compo­ many authors, e.g. Sjogren (1964, 'Quercus robur-Betula sition is very similar to that in mesotrophic forests, apart verrucosa-Wald mit Corylusavellana' ), Ekstam & Sjogren from a somewhat higher frequency of more light-de­ (1973, Betulo-Quercetum melicetosum), Ekstam (1979, manding species. They often are remnants of former 'Hedera helix-Oxalis acetosella-typ'), Ekstam & Martins­ wooded meadows where hazel was regularly cut back. son (1981, 'Ek-hasselskog av lundgras-typ') and 0. Structurally similar Corylus coppices without a dense Johansson (1985, 'Hasselrik ekskog'). The majority of overstorey of trees have been described from Norway, the stands are found in the coastal plain at the Kalmarsund. e.g. by Fremstad (Melico-Coryletum, 1979), 0vstedal One of the most extensive stands of this type is Borga (1985) and Rjljsberg & 0vstedal (1987). Hage, a 93 ha large nature reserve close to the town of Borgholm on bland, early described by Du Rietz (1917), Oxalis acetosella sub-community, Carpinus betulus Stemer (1926) and S. Johansson (1955). This forest may form (releves 66-75) (Hombeam forests) give the impression of a 'primeval' forest, but has devel­ The forests are highly dominated by Carpinus betulus, oped from an open forest, formerly used as pasture and creating a stand structure different from other mesotrophic wooded meadow (S. Johansson 1955). forests (Fig. 13). Carpinus forms a close, fairly low canopy The forests grow on mesotrophic brown earths on which locally is overtopped by old individuals of Quereus morainic, often sandy deposits poor in lime. The forest robur and Fraxinus excelsior. Despite their fairly small floor has a mull humus layer and is almost completely size (about 20 m), the oldest individuals have an age of covered by oak litter. pH-values are usually lower than more than 150 years (cf. Danielsson 1917). Carpinus and those for soils of the Tilia cordata sub-community, vary­ Crataegus spp. are the only constants in the poorly devel­ ing between 5.0 and 5.5 (Diekmann 1988). This is also oped shrub layer. Thus, there is a clear structural similar­ indicated by the comparatively low R-figures (Table 11). ity to beech forests. The fieldlayer is fairly species-rich, Oak-hazel forests have been influenced to a high but poor in species differentiating from the other sub­ degree by man (Sjors 1967). Most Swedish stands have units. Bryophytes show a high cover and the number of arisen from wooded pastures or wooded meadows species is large (mean 8.9 species/plot) probably due to (Selander 1955). Evidence of former land use is found in the rapid decay of the hombeam-litter.

Table 11. Quercus robur-Euonymus europaeus community. Number of vascular plants (Vas), number of bryophytes (Bry) and total number of species (Tot), as well as indicator figures for light (L), temperature (T), continentality (C), moisture (�). reaction (R) and nitrogen (N) for different sub-units. Average values are given. The sub-units correspond to: 1-3. Tilia cordata sub-community (1. Filipendula ulmaria form; 2. Stellaria holostea form, Milium effu sum sub-unit; 3. Stellaria holostea form, Dactylis glomerata sub-unit); 4-5. Oxalis acetosella sub-community (4. Corylus avellanafo rm; 5. Carpinus betulus form).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 44.4 6.9 51.3 4.9 5.0 3.9 5.2 6.6 5.5 2 37.4 5.0 42.4 4�9 5.1 3.9 5.1 6.6 5.5 3 33.3 3.9 37.2 4.8 5.2 3.9 5.0 6.5 5.6 4 34.4 5.7 40.2 4.9 5.0 3.9 5.1 6.3 5.6 5 41.5 8.9 50.4 4.7 4.9 3.9 5.1 6.4 5.6

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 41

Fig. 11. Mesotrophic mixed deciduous forest with linden (Quercus robur-Euonymus europaeus community, Tilia cordata sub-community). In the foreground, individuals of the two characteristic tree spe­ cies are visible, Quereus robur (to the left) and Tilia cordata (to the right, regenerating with basal shoots). Conspicuous in the field layer are Anemone spp., Acer platanoides (sap­ lings) and different grasses. - W Eriksore, bland, May 1993. Photo Folke Hellstrom.

Carpinus forests are restricted to a few areas on Gland. indicator figures are similar to those determined for the Releves were made in the two largest stands: Halltorps Tilia cordata sub-community, except for a lower L-value Hage and adjacent areas to the north (Stemer 1926; pointing at low light levels in the interior of the forest Akerlind & Sjogren 1975; Ekstam 1979; Ekstam et al. (Table 11). 1984 ), as well as an area of ea. 1 km2 in theMittlandsskogen The occurrence of more or less pure Carpinus forests (Sjogren et al. 1974; Stemer 1986). These forests on instead of mixed forests with Tilia is probably caused by Gland represent the northernmost Carpinus stands in man's activity. Carpinus is fairly tolerant of repeated Sweden; the poor growth of the hombeam trees may be cutting and grazing (cf. Ellenberg 1986) and is considered caused by the unfavourable climatic conditions for this as a characteristic species of abandoned pastures in S species at its northerndistribution limit. Within the Boreo­ Sweden (Bergendorff et al. 1979). Halltorps Hage has nemoral zone, smaller stands and groups of trees can also been used as wooded meadow and temporarily also as be found in Blekinge, Smaland and Halland, but these pasture (Ekstam et al. 1984). The Gland stands might usually grow on much poorer soils where Carpinus oc­ have developed from small, locally restricted populations curs together with other tree species, particularly Quercus profiting from land use practises disfavouring other tree spp. and Fagus sylvatica. Within the Nemoral zone in S species. The stands are thus compositionally probably not Sweden, Carpinus forests are more frequent. The edaphic stable, but will eventually develop into more mixed de­ conditions resemble those found for the mixed forests of ciduous forests (Anon. 1984). the Tilia cordata sub-community, with brown earths as the most common soil type and pH-values between 5.3 and 5.8. A soil profile studied in Halltorps Hage was identified as (pseudogley) para-brown earth. Also the

Acta Phytogeogr. Suec. 80 42 M. Diekmann

Table 12. Releve table of mesotrophic mixed deciduous forests, Quereus robur-Tilia cordata community. (See Table 3.)

Running number 12345678911111111 112222 2 2 2 2 2 2 3 3 3 3 333 3334444444444555555555566666 666667777777777888 8888888999 01234567 8901234567890123456 7890123456789012345678901234 567890123456789012 3456789012 Cluster 10 11 12 13a 13b Cluster code 100100 1001010 1001011 100110 100111 Region SUUUUUSUSSUSSSSUUVN SSSSSSSSSSSSSSSSBB SSSSSSSBBB VVUUUUUUVVVUNNN usuuuuussuuuuuuuuuussvssosov AAPPPPPPAAAPOOOOA 0PPPPPMPMMP6666PPAO POPPPPP00PPPPPPPPPPOOA66GOGA MMMMMMMMMMMMMMMLO MMMMMMMLLL Aspect --S--N-SEEWWNS--E SSEW-W-SSSNSSSSESES WS-NWW- SS-NNS-WWENSSWENSWS -E SS-NWSSSSWENN-WWSN -SN-W- -NWS E E E EE E WEWWW EE S E E W W S E S S W SWEEE W N E E W W E WW w Incl ination --2--1--2 4 5 1445-4 1581-8-582183881155 31-538-15-188-118852558181-5 21-12331w 43142-5213 -55-5-- 551 (0 ) 5 0 000500 5 0 0 02 0000 0 0 5 55 5 0 0 5 0 03 00 0 0 0 00 0 Cover % T1 (X10) 44764766757677435 634 643 5 55 6 543444455 4554454476543643365547654546 665556655555256675 4655556556 % T2 (X10) 57-333254-3153675 -2-23533167756634 433554-55354246122165135331- 325362634233741134 53111117-- % (X10) 65456442374533434 7724845323243454342 5655547436257628154351457674 133434252543158714 4567624167 s % F (X10) 88674526554334326 7766765865866867831 6867775675765474658776867777 778386776776418837 6362667455 % B 121--1---5-1--5-- 111-5-111112121--1- 1111--1---11----1-2--51-1------15------11---5 1-1-11-1-- 00 0 0 5 N - Tot 34333432252223324 2443343333333443331 343 3 3223 3 42 3 223333 53243243 3 3 222232232322222312 3232233232 50292029804742725 9306948456177710663 351727958083589413 0081152035 558968634268813189 2192661633 N - Vas 33233332242223324 2333343333233433321 3433322334232233334323323333 222232232322212312 2232233232 04992829894642425 8896344112605050683 2206278580625894032089053035 558718634268493182 9122441233 N - Bry 563--2---1-1--3-- 151-6-434457276--8- 1311--1---21----1-8--21-9------2 5------42---7 3-7-22-4--

nUlmus glabra . . 43 .34.653 . 54545 ...... 3 ...3 ...... 3. 4 ...... 4 Fraxinus excelsior 45 ...33 ..34. 43434 . 4 3 ... . 4 •••••••......

.•••...... • Quercus robur . . 665353 ..4 ...... 645655555445455543 . 445554 . 443 . 44 . 345555 ..34555 . 55555565555545665 . 5656666556 Tilia cordata ...... 46 ..56 .53 ...... 44544 ..4555 554445556554 .55 ...445655 .4.4 33344 4.4.3.434 .... Acer platanoides . . 3 ..6 ..4. 3 .34 ... .33 ...4.3 ....3 ...43 34.434 ...4. 334.4. 5 ...... 3 .. 544344 ...... Populus tremula ...... 4. . ...4 ...... 3...... 4.4 .... 4 ...... 4.4 33 .. 4 Betula pendula .. 3 ..33 ...... 3 ...... 33 ...... 3 . 3 ...... 3 ...... 3 ...... Sorbus aucuparia ..4 ..4 ...4 4 ...... 4 ....3 ••••••••••••...3

Picea abies ...... 34 ..3 ....•.. ...3 .. • .•...•...... 3 ..•..••••••••••. . . 3

Quercus petraea ...... 4 ...... 35 ...... 4 5 ...... ••••5

. ••••• • •...... •.•••

UlmusT2. glabra . . . 544355.4.5.565 4 ...3 ....4 ...... 4. 34 .....3 ...... 2 ..

Fraxinus excelsior 55 ...3 ....23 ..33. ....•...... 3 ....543 ..3 ...... 3 ...3 ...... 4 Malus sylvestris 33 ....335 3 ...3. 2 . 4 ......

. Acer platanoides 45.. 34344 . 4334544 ..4 ..34434 . 44455444 344455 . 45444455344 .442 . 4443 . 445444 ....3 ...... 4...... Tilia cordata ••••••4 3 4...... 4544356535535 544443 . 54443 .35 ...354355 .4.. 4. 445465345465 ... 3 ......

Sorbus aucuparia 33 ..43 ...... •••• •••4344 ..4 ..4.3 ...3 ...34 .. . 43443 ...4. 444 ..33 ..3 ..3 .... 3. 3 ...3333 . 43323 .3. 33 .3.33 .. . Quercus robur . . 3 ..33343 ..4 ...43 . 3.34 ...... 3 ..33 ..43 ..343 .. . 3 ...... 3 ..... 5 ....4 .. .

Picea abies ...... 32 ...... 3 ...3 ...... 33 . 3.3 ....4.3.3 3 ...... 3 ..3 ...... Populus tremula . . . . 4 ...... 3 ..3 ...... 34 .3.... . Fagus sylvatica ...... 3 ...... 3 ...... 55 ...... 3 .. Betula pendula 3 3 ...... 3 ...... 43 .4.....

•••••••• .•••••••......

.s.Fraxinus excelsior 444 ..4.3.3.5 34434 2 ..2 . 23223 . 33434322 . 3 .....24 .3 ..23 3.2 ...4 ...... 43 ...... 35

Lonicera xy losteum ...23244 .33424 .23 ...24 ...... 2. 34344444 .4.4352232 ..43434 .23...... ••••••••••4

Ulmus glabra ..33 . 2234. 3444443 .....4 ...3.3 .22 ..42 ...... 2 .2 ...... 4 ...2 ••2

Prunus padus 2 ..44. 3244 ..2 .24. ..2 ..2 ...... 3 .4 .. ..2 ..2 ...2. 224 ..42 ...... •..• Crataegus spp . 2 ...... 3.3 ..242 ... ..22 ..323 ....324 .. . 2 ...2 ..33

Malus sylvestris 32• ...... •..2 ...... 3 ...... 43 .2 .222 ...... 3 .. ..

Prunus avium ••4 ...... ••••••••••• 2 22 ....4 .... .4...... 24

•••••••••••••• ...... ••.•...... ••. Corylus avellana 55 .55.4.265443435 66456554334 . 2354444 4545436535446336655453456564 4342244434443266.4 .566645 .6. Acer platanoides . . 4.23324 .2.2.423 .33 .232344 . 434342 .2 444144 . 3324432443323 ...32 ..4 . 43424 .2.2232 ...3 . .2 ...... Tilia cordata ...... 4 ...33 .3 ...... 34 44454444434 544323224334232 ...444344 .4.4 334544452544453 ...... 2

Sorbus aucuparia 334334 ...3 .222432 ..2. 3 34 ..33243333 .. 3.4233 .2.34. 42 .23 .. 3.2 ••3 ... ..2343 ...2. 2 .32 2.2. .32 ..22 ..3 Ribes alpinum 32323344 ...2 .... . 2.2 .3...... 44 .. 33333443 . 434232342322 ....22 ...... 2 .. 3 . . 3 ...33 ...

Picea abies . . 2 ...2 ..2 .....3. . ...2. 3 ..2 . . . 2.2...... 4 ....2 ....2 ..333 ... . • • 2 ...... 2 .. 3. Quercus robur 2 ....2 ...... 3. 3 ...222 ...... 2. ...2 ...... 2 . 3.••3 •••...... 3 ...4

Populus tremula . . . . . 2 ...... 3 ...... 2.323 ...... 2.2...... • ..... 2 •••... 2 ......

Ribes uva-crispa . . 2. 43 ...... •••••••2 . .• . .22 ......

Viburnum opulus ...... 2 ...... 3 ...... •••..•••...... 11 .. 3 Rosa spp . 2 2 .. .3...... 2 ...3

Fagus sylvatica • • • • • •.••••••••. • ••.•...... 2 ...... 3 ...... 32 ...... 3 ..

..••......

EGeranium robertianum . . . 1..1.1 2.2. 2 ...... 1. 3 ......

Urt ica dioica ....1. .. 31 . ..2. 2. 2 ••••••••••••••• ••..•••••...... 1 ......

Geranium sylvaticum .4.314 .22 .....2.2 333234. • • .3 ...113 . 21 .. 3234212 .3221. . 21323 ..2 ..11 . 3...... 2. Paris quadrifolia 44 .4322422 ...12 . 2 ..2 ..1 ....1 ....32 .. 3433244434 .434314 .2 ..2 ..4. 32 . .1 ...... Calamagrostis arundinacea ...... 3.2 ...... 13 .252.44 .5442344 .. 4544333 ..2332 ..3 ..53444 44335 Carex digitata . . 3 ...3 ...13222 .3 313 .41.2 ...... 31. .3.2 ...21322 .123 ..21 23123112 ...... 1 ...... Lonicera xylosteum ....2. 22.3231.2 .. .1...... 1. 2322 . 222 .2. 3241211 ..22322223 3

Prunus padus 21.1312123 .11.3 .. . 2.2. 2 .....12 3 3 . 3 .. . .1 ..12 .11.122 ..22 1 ...... ••••••••• .•2

Actaea spicata ...341 3244 ..33434 . 3.3 ...... 11. ..4. 1334423343 ...••••31. ...2 ...... ••••••••••• Fragaria vesca . . 4 ..2 ...22 1. .12. .233422 ...... 432 ... 1.132 ..1.1. ....2 ....232 .2 ...... 2 ... 3 . Ulmus glabra . . 211223 ...32 .132 .2.3.1. ..2. 2 211. .21 ...... 1. 2 ...2 ......

Viola mirabilis ....3 ..3 ....4. 1.4 2 ...... 2 ..444413 ..4. 32 ..1. ..22 ..22 . ·

Taraxacum officinale agg. 22 ..2 ....2. 12 .1...... •... 21113 ...... 12 ...... 22 1.1.1...... 2 ...... 1 .....

Myce lis muralis ....232 ..4 ..2. 2 .. . 32333 ..11. ...2 ..21 ...... 1 ..1.1 .. 3 ...... Hieracium vulgatum .....21 ..1 ...... 33 .33.1.21.2212 .... 11 ...... 1 ...... 1 ...... 2 ...... 1 . .

Ranuncu lus acris . . . . 332 ...... 43432 .....1 ..1.1 .. . .11 ..2 ...... 11 . 2 ...... 1 ...... Maianthemum bifolium 43 .3... 1. ...3 ... . . 412 ...22322 .2 ....3 ...... 1 ... ..

rnu ltiflorum 1 ..4 ...3 ....3 ...... 43 .4 ..3 ...... 2 ...... 3 ...... Polygonatum ...... 1 ...... Vaccinium myrt illus . . 2 1 2 .. ..3. 3. 4 ...3 ...... 3 ...... 1 ...... 2 ......

Scrophularia nodosa ...... • . ••••3 ...... •. 1. ...1 ...... 1 ...... 1 .....1 Melica nutans .. 3 .. 22 .. 43.... 14 3.4 ..4133 . 2341 .4331 333233 ...311 ..32223333333324 2 ...22 .1. .2 ...... 1 Tilia cordata ...... 31 ..241 .2...... 232323 2324423 3221.2 ...22 ..22 ...2. 3233.3.3 12242221122323 ...

Polygonatum odoratum . . 3 ..2 ...... 1 ....1 ...... 21 .2 12 .3 .. 2 .. 1.4. 323 .. 2 .. 2 ....2. 21 ..2 ..333 ......

Solidago virgaurea ...... 1.1.1. .12. 2 ...1213 .12 ...... 2122 .....2.1...... 1.1 1. ....1 ...3 .....1 ......

Deschampsia flexuosa . . 2 ..2 ...... 1 ... 313 . 3443321322 .12...... 1 ...... 1 ..1 ...... 1 . 1.1.232332 Campanula pers icifolia . 113 .....123231 ...... 1 ...21 ...1 ...... 11 ...... 2.1.111 . Lathyrus montanus ....1 ...... 323 ..34443443 43.32. 31122 .....3 ....3 ..23433322 .1 .2.21. .1.211 ...... 21222122

Lathyrus niger 1...... 434 ..342 ...... 2444 .3.324 34.14412223 . ..1 ......

Lathyrus vernus . . . 2 41 ...... 3 4 3 .. 233334134 .33.33.2.3343444344 . .2.42 ...... 2 ..4 ..•

Milium effusum . . . 3 •.•...4 ...... 11. ...•••••••42 ...... •...1 ...... 323344.32433334 .2.444 . 4424 .. . .1.1...... 1 ....

Pulmonaria officinalis ...... 14 ...... 3 . 12 ..4 ...... 4 . . .2 ...... 5 . .

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 43

Ta ble 12, cont.

Daphne mezereum . ..232 .....221 ...... Cardamine bulbifera ••••••••••2 ...... 3 ....32 .....24 . 3.33.322 .2434 .44 ..

Stellaria holostea •••••••• 2 ..44334 44342454 .. 4.44444 .4. Melica uniflora . 4 4 542135.323531. 554 ...... 3. 23

Mercurialis perennis • • . • . • • •••••••••• ...... •.••..••• •...•••• 4.6.4465555.5.1146 ..4 ...4 .. . Galium aparine ...... 2 ...... 3...... 12 ..4 ....3. . 42 .1.... . Prunus avium ...... 1...... 12...... 4 ...21 .3.

Rubus fruticosus agg . . .1...... 21 .... 1. ...2 ..43

Anemone nemorosa 66 . 5564543551.2.3 545444755456655655 . 5655555665556565854554555553 657556656656776565 67767676 .. Hepatica nobilis 4344243544342.424 444434 . 43324444441. 444444433424445443434444 .443 424443 ....443 ..433 4 ...4.4 .43 Convallaria maj alis 3 ..44444222 .1.2.4 643443355434234543 . 4.444444 .3444434454442554444 21.3443445253 .443. ..42444 .2. Viola riviniana .142121 .1.221 .2.1 34232443214244323 .. 242322 .2432 ..2332 14122423331 1.123121.311224.32 34434332 .. Acer platanoides 342 . 31332 . 2232332 .44.234334234323321 342231 . 433333243]43 ..2334332 423243 .2.2324 ...2. 132 .1. . 31. Poa nemoralis . 144223 . 4344.3 ..4 442423 . 53445554444. . 3 ..2. 32115 ..134�34544545544 45125423232 . 1 .. 433 .233.1222 . Sorbus aucuparia 213 . 321 ..2132122 . 2.21 ..2211 .21.21112 22322 . 22222223122212 . 121211. .1123 .....12 . 122 .1 1221211 ..2 Vicia sepium ..2133 .....2 ..1. . 233443 ...... 1313 .2. 23 . 4212241323 . 322224434 .3442 33 .. 33222333 ..24 .. ..22 .21.3. Corylus avellana 21 .12 ....1133 ...2 2.22322 .121. . 21.322 . ....2212 .2 ..221 .23 . 312222 .2 . . 1... 1121211 .32.1 .123121 .2 . Quercus robur 2. 32 .11. ...1. .1. . . 12233222 .33223 .11. 22 .. 2.122 ...... 1. .. 222 ...1. .. .12222111111.2.23. 3222113424 Ranunculus auricomus agg. 33.2443.1. .3 ...... 4. 441.3 ...44 3334 .. . 4 ....4. 24 ....44) ,4 ....3. 332 . 424 .4.444444 .33424 44424422 .. Dactylis glomerata .1.3.3... 1.1. ... . 233423231231144 .3 .. .2.. 2.1.1 ...... 3 ...... 132 . 212 ...222 .2 2132342 .33 Geum urbanum 124233 . 122 ....2. 2 . 312 ...... 34422 .. . 2 .1.1.44 .2.124 . 3.22.. 2 ..2 ... . . 2 ...23 .3... 3 .. 1. 221. .2 ..3 . Dryopteris filix-mas ...5 . 412444334445 ..43121 .134 ...43222 ...... 2 ..1. 5 ..3. 322 .3.... 44 ...2.1 ...... 1 ...... 21 33 Fraxinus excelsior 4. 4 ..2. 3. 2244443 . 13 ....3. 34.33434413 .2 .....2415 ...42 .;2 ...2 ...... 422 .1...... 224 .. ..1 .... . Rubus saxatilis 42.3...... 1 .. . ..24.4 ..3. 11. ..2 .. 434444 ...4.4. 432432 .41443 .22 ...... 33 .4221.4 .. . ..1 ...... Veronica chamaedrys ...... 4 ....1. 4 ..1. .2 13 .34.24333 .. .1...... 3 ...... 2 ..41244 .3424 . 1. .321. ..12 .....1 2.42442 .3 . Ribes alpinum . 2.13233 ...... 2.2 .. 1 ...... 33 .. 22223 .43.3232313323 .2.22.21...... 3 ...2 .. 42 ...32 ... Anthriscus sylvestris 13. 123 ...... 3 . 3.4 ...... 3. 3 ... . 3.333 .....1. ...13 ....4 ..43 . 3 ...211 .2.2.. 2 .... .4.1. 4 ..4. Oxalis acetosella . 2.3 ...333 ...4 ... 4 . 4. 222 . 454 .4 .1. . 43 ...3 ..4. 4...... 33 .1.4 5344 .. 55..

Populus tremula . . . . 1 ...... 12 .21 3•••••• ...11 ...... ••••••••••••1 . . 31 .1.333 ...... 1. . 22 .. 222 .3 . . . 2 .....2. 2 ...... 2. 3 .1... Ranunculus ficaria . 124... 2 .1...... 55 ..... 3.2 ...... 444 ..42 . 44314554 .7 5 ..4 ..44 .. Viburnum opulus . 1. ....1.1.1 3 ...3 .1.... 3 ...... 2 1 . 3.3 .1.1. ...2 ..11. . 2 ..3 .....3 ...... ••••••2 ...... 2 ..

Deschampsia cespitosa 22 .2...... 4 ...... 1 ...... 1 ...... 1 ...... 1 ...... 12 ...21 .2. 211 . •••..1 . ..21 1.1 . Laserpitium latifolium . . . 2 ...... 2 ...... 2 ...... 1 .. . . . 2.3. 2 ...... 3244 .4 ..4 .. 12 ...412 .222 ......

Luzula pilosa •...... 1.1...... 1.1. .11. ..2. 1...... 1...... 1 ...... 1111 ..1 ...... 2. . .122 ..31 . Crataegus spp . 22 2 22 ...... 11 .....22 ...... 1 .. . . . 2 ...22 .11 1..24 .. 1 ......

Allium oleraceum . 3214..•.....••••••• ...... •. .••••••• ...... •••• . . 1 ... 3 .. . 4 12 ....24 .312 . 31. .. 42 ...2

Rosa spp . . . . . 11 ...... 211 ...... 2 ..1 .... •. ••••••••••••...... 11 ••••...•••••• ...... 1 ... . . 1 .....1.1 ...... 1.2.••. . Galium odoratum . . 2 ...4 ..5.3 ...4 ...... 1 ...... 3 .....43 ...... 4. 4 .....4 4 ...... 4

Galeopsis tetrahit • ...1. ...12 .121. ..2...... 1 ...... 1...... ••••••••••• ...... 31•• . ..1.2 .2 .. Melampy rum pratense ...... 1 ...... 2 .21...... 3.2...... 2 4 .. . ..2. 2322 .

Primula veris 1 ...... 2 .....21 ...... 1 ...... 3 ..1.1 ...... •. . 1.1.. 3 ....

Hieracium murorum . . 4 ...... 3 ...22 .. •.•.•.•...... 4 ...... 1...... 1 ...... 1.12. 2 ...... Rubus idaeus . . 2 ...... 3 ... ..3 ..2 ...... 3 ...... 2...... 1 ....4 ...4 Elymus caninus . . . . 2 ....2.1.3 ..3 ...... 32 .... 2 . ...2 ......

Polypodium vulgare . . . . . 13 ....1 ...... 2 .1...... 11 ...... ••••...... 3 ...... ••.•••...... 2 ... Galium boreale . . . . . 1 ...... 1.2 .....2 ..1 .. . .1 ...... 2 ..1 ...... 1 ...... Dryopteris carthusiana ...... 1 ..3 ... . . 1 ..1 ...... 1 ...... 1 ...... 1 ...... 3 .. Melampyrum nemorosum ...... 4 ..5 .....3 .. 4 2 ...... 2 4 ••••••••• .....3 ......

Campanula trachelium .1.2 ...... 3 ...... 1 ...... • •.•...... ••2 ...... 3 Carex montana . . 2 ...... 3.2 ...... 2 1 ...... 4 .. . Pteridium aquilinum 2 ..4 ...... 3 ...... 24 ... • 1 ...... •••••••••...... Sanicula europaea • • • • • • 2•. ••••.••...... 1.1 1 ...... 3 ....2.

Gymnocarpium dryopteris •. .• .• .• .• .• .• . 2 ....422 . 4 ...... 2 ..

Alcherni lla vulgaris agg . 23 ...... 1 ...... 11 .... •••••••••••• •••••••..•••••• Gagea lutea . . 23 ...... 2 ...... 3. 2 ...... Ribes uva-crispa ..2. 3 ...... 1...... 1 ... . 1 ...... Hypericum maculatum . . . . . 1 ...... 1 ...... 2 .... . 1 ....2 .. .. Carex muricata agg ...... 1 ..2 ...... 1 ...... 1 ...... 1 .

Trifolium medium .1.. 12 ...... 1 ...... 1 .. ..

Picea abies .....12 .1...... 2 ...... 1 ..

Malus sylvestris ...... 1 ...... •...... 1 ...... ••••••••••••••••••1 ...... 1 ...... 1 .. .

� . . 1 ..2 ...... 4 ..3 ..3 ..4. 133 ..2. 2111 ...... 1 ...... 2 ..2 ..3 ... ..2.1.... . Rhytidiadelphus triquetrus Cirriphyllurn piliferurn 22 ...... 2 .. .2.... 22 .. 32 .11. .. . .1...... 2 .....3 ...... 22 ...... 12 ...1 ..1 ...... Brachytheciurn rutabulurn 3.1...... 2 ...11 21. .2 ...... 1.3 .....2 ••• 22 ...... 1 ...2 ..1. .2 .1. .

Eurhynchium hians 44 ...... 1 ..4 .. . 1 .....2 ...3 ..2 .... .3 ....1 ...... 2 ...2.4 .. . . • • 4 1 ......

Plagiomnium undulatum 24 ...... 1 ..1 ....1 ..2 ... . 2 •. ... • • • 4 • ...... • • • • • • • 2 • .... • • • • 3 .... 2 .....

Plagiomnium affine 1 ....1 ...... 111 .1 ..... ••...... •••••••••••••••...... 1 •••••••••.....3 ...... 1 ....3 ..1 ...... 2 .1...... 2 ..... 1.1...... Rhyt idiadelphus squarrosus Pleurozium schreberi ..1.2 ..... 1 ..2 .....

Additional species (occurring in one , two , three or four releves):

4x: (T2 ) Quercus petraea 17 :3, 26:3, 27 :3, 82 :3 (F) Epilobium montanum 10 :3, 11 :3, 14 :2, 17 :3, Geum rivale 1:3, 2:4, 19 :2, 67 :1, Ranunculus repens 6:2, 19 :2, 20:1, 43 :2, Veronica officinalis 3:2, 27 :1, 30:2, 35 :1, Vincetoxicum hirundinaria 21 :1, 29:2, 30 :2, 66:2 (B) At richum undulatum 2:3, 69 :1, 82 :1, 90:1.

3X: (T2 ) Prunus avium 73 :2, 83 :3, 84 :4 (S) Carpinus betulus 26 :2, 68 :3, 90 :3, Rhamnus catharticus 54 :2, 91 :2, 92 :3, Sorbus intermedia 24 :3, 69 :2, 90 :3 (F) Agrostis capillaris 24:1, 89 :2, 91 :1, A1liaria petiolata 29 :1, 32 :3, 84 :2, Anemone ranunculoides 70 :4, 79:3, 80 :3, Athyrium filix-femina 14 :3, 16:2, 19 :2, Campanula latifolia 2:2, 15 :2, 67 :2, Campanu la rotundifolia 20 :1, 23 :1, 88:1, Carpinus betulus 26 :2, 68 :1, 90 :3, Corydalis intermedia 3:2, 4:3, 8:2, Fagus sylvatica 68 :1, 81 :3, 90 :2, Glechoma hederacea 9:1, 10 :3, 14 :3, Holcus mo llis 84 :1, 85: 3, 89 :4, Lathraea squamaria 38:1, 51 :2, 53 :3, Melampyrum sy1vaticum 61 :4, 85 :3, 87 :2, Moehringia trinervia 30:3, 33 :2, 34 :2, Quercus petraea · 27:1, 64 :1, 82 :1, Stachys sylvat ica 10 :2, 17 :3, 6 7:2, Stellaria media 9:1, 10 :2, 72:1, Valeriana sambuc ifolia 14 :1, 16 :1, 64 :2 (B) Brachythecium popu1eum 2:1, 27 :1, 77 :1, Hylocomium splendens 22 :3, 24 :1, 31:2, Hypnum cupressiforme 26:1, 61:1, 90:1, Isothecium alopecuroides 30:1, 31:1, 32 :2.

2X: (T1 ) Fagus sylvatica 68 :4, 81:5, Pinus sylvestris 3:4, 76 :3 (T2 ) Carpinus betulus 26:4, 90 :6, Prunus padus 20:3, 37 :3 (S) Betula pendula 19 :2, 23 :2, Cotoneaster integerrimus 6:2, 23 :2, Juniperus communis 23 :2, 56 :2 (F) Aegopodium podagraria 2:2, 38:4, Bromus benekenii 38 :3, 67 :1, Carex pallescens 31:1, 55 :1, Carex pilulifera 20 :1, 24:1, Clinopodium vulgare 18:1, 34 :2, Crepis paludosa 1:4, 2:3, Cystopteris fragilis 10:2, 11 :1, Festuca gigantea 38:1, 80 :1, Filipendula ulmaria 1:5, 2:4, Impat iens noli-tangere 9:1, 14 :3, Listera ovata 1:3, 2:4, Lonicera periclymenum 34:4, 92 :5, Prunus spinosa 87 :2, 88 :1, Rhamnus catharticus 30:1, 91 :2, Rubus caesius 1:4, 35 :2, Silene dioica 67 :1, 77 :1, Vicia sylvatica 58 :4, 61 :4, Viola hirta 10 :1, 12 :1 (B) Eurhynchium praelongum 26 :1, 29 :2, Fissidens taxifolius 15 :1, 47 :1, Plagiochi la porelloides 47 :1, 61 :2, Plagiothecium denticulatum 26:1, 90:1, Thuidium tamariscinum 2:1, 61 :1.

1X: (T1) Betula pubescens 1:4 (T2) Sorbus intermedia 80:2 (S) Cornus sanguinea 64 :2, Frangula alnus 24:2, Quercus petraea 27 :3, Sorbus norvegica 36:2 (F) Allium scorodoprasum 92 :1, Alopecurus pratensis 25 :1, Anthoxanthurn odoratum 89 :1, Asplenium trichomanes 35 :2, Avenu la pubescens 25:1, Betula pendu1a 3:1, Bilderdykia dumetorum 26:2, Brachypodium sylvaticum 13 :2, Cardamine impat iens 10 :2, Chelidonium majus 4:2, Festuca rubra 92 :2, Fragaria moschata 14 :2, Frangula alnus 27 :1, Galium verum 88:2, Hedera helix 80 :3, Hieracium umbe llatum 23 :1, Juniperus communis 27:1, Lapsana communis 4:2, Lychnis viscaria 10 :2, Neottia nidus-avis 19 :1, Platanthera chlorantha 55 :2, Platanthera spp . 83 :1, Poa angustifolia 91 :3, Potentilla erecta 27 :1, Scorzonera humilis 62:1. Sedum telephium Sorbus norvegica 36 :1, Thelypteris 35 :1, phegopteris 14 :2, Trollius europaeus 63 :2, Vaccinium vitis-idaea 23 :4, Vicia spp . 88:4, Viola reichenbachiana 80 :3 (B) Brachythecium velutinum 69 :1, Dicranum scoparium 22 :1, Eurhynchium striatum 58 :3, Fissidens cristatus 61:1, Mnium hornum 82 :1, Rhodobryum roseum 27 :1, Thuidium erecturn 55 :1.

Acta Phytogeogr. Suec. 80 44 M. Diekmann

Fig. 12. Oak-hazel forest (Quercus robur-Euonymus europaeus community, Oxa­ lis acetosella sub-commu­ nity, Corylusavellana form). Large old oaks dominate in the canopy. Hazel forms a dense sub-canopy, other shrubs are sparse. In the field layer are visible Anemone spp. and Convallaria majalis. Lilla Vickleby, bland, May -1993. Photo Folke Hellstrom.

Quercus robur-Tilia cordata community (Table 12) layer contains a large number of differential species. Some show a northern distribution pattern such as Gera­ Compared with the mesotrophic forests on bland, the nium sylvaticum and Actaea spicata. The bottom layer is mainland forests are less species-rich, both with respect to poorly developed or totally absent. vascular plants and bryophytes. Two sub-communities Three forms are distinguished: a Geranium robertia­ can be distinguished, comprising releves from two sepa­ num form (releves 1-17) in stands from Vastergotland, rate geographic regions: a Geranium sylvaticum sub-com­ Uppland and Norway, with no constant differential spe­ Deschampsia flexuosa munity (releves from Norway and different parts of Swe­ cies, a form (releves 18-36) in den, except the coastal areas in Blekinge and Smaland), stands from the same regions as the Geranium robertianum and a Stellaria holostea sub-community (releves from form, as well as from Smaland and Sodermanland, differ­ eastern Blekinge and Smaland). entiated by Deschampsia flexuosa and Campanula persicifolia, and a Lathyrus vernus form (releves 37-64), Geranium sylvaticum sub-community (releves 1-64) mainly in stands from Uppland and Sodermanland, differ­ The forests are dominated in the canopy by Que reus robur entiated by Lathyrus vernus and Milium effusum. (less frequent in the lower tree layer), Ti lia cordata and Mesotrophic forests have been described from SE Acer platanoides. Sorbus aucuparia and Picea abies are Norway (St�rmer 1938; Kielland-Lund 1981) and differ­ common in the lower tree layer. Corylus avellana domi­ ent regions of the Swedish mainland, e.g. by Almquist nates in the fairly dense shrub layer, whereas Crataegus (1929) from Uppland, Ryberg (1956, 1971) and Carlsson spp. are much less abundant than in corresponding forests & Clemedson (1989) from Sodermanland, and Wallin on Gland and in the coastal provinces of SE Sweden. ( 1973) from Vastergotland. Also the mesotrophic forests Lonicera xylosteum and Prunus padus are differentiating mentioned by Sjors (1967), Klotzli (1975a) and Anon. towards the Stellaria holostea sub-community. The field (1984) refer mainly to this community and not to

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 45

Fig. 13. Hombeam fo rest (Quercus robur-Euonymus europaeus community, Oxalis acetosella sub-com­ munity, Carpinus betulus form). Hombeam is domi­ nating and forms a dense, but fairly low canopy. - Halltorps Hage, bland, May 1993. Photo Folke Hellstrom.

corresponding communities on Oland and Gotland. As south-facing hillslopes, so-called 'rasmarker' and 'syd­ synonyms for the three forms distinguished here, the vaxtberg', are most striking with their thermophilous and following are mentioned: nitrophilous vegetation in the Boreal zone of Sweden to - Geranium robertianum form: 'Oxalis acetosella-sam­ the north of the 'limes norrlandicus'. Several southern halle, Geum urbanum variant' and 'Calamagrostis deciduous forest species reach their northernmost distri­ arundinacea-samhalle' (Wallin 1973); bution on such sites (cf. Mascher 1977), here often con­ - Deschampsiaflexuosa 'Oxalis a etosella- amh form: c s alle sidered as relicts from the Atlantic time period. As factors , Typisk variant' (Wallin 1973); Vaccinio-Tilietum (and favouring the occurrence of deciduous forest vegetation partly Polygonato-Ulmetum, Klotzli 1975a); on these sites, mention could be made of: - Lathyrus vernus form: 'Anemone hepatica samhalle, (1) instability: The sites are usually rich in boulders and - Typisk variant + Pulmonaria variant' and 'Poa nemoralis­ stones, inclined to move down the slope. This favours samhalle' (Wallin 1973); Pulmonario-Tilietum (Klotzli plants with a powerful root system able to penetrate the 1975a). small areas of fine soil particles between the rocks. Among The stands are often found on shallow or steep, prefer­ trees, particularly Tilia and Ulmus are capable of growing ably (E-) S- (W-) facing slopes, especially in the western at such sites, whereas Picea with its flat root system has provinces. Several releves are located on the plateau-like problems to establish (Klotzli 1975a). Tilia can reproduce hills of Halleberg, Varvsberget and Gerumsberget in by basal shoots creeping between the boulders and is in Vastergotland and at Korpberget in Sodermanland. The this way also protected against grazing (Sjors 1967). special environment and floristicpeculia rity of such steep, (2) increased insolation: A more or less south-facing, sun­ boulder-rich slopes and screes have been much discussed exposed slope receives a higher insolation than a plain in Swedish literature (e.g. Halden 1950; Du Rietz 1954; surface at the same latitude, thus favouring the occurrence Selander 1955; Sj ors 1967; Bergendorff et al. 1979). The of 'southern' species. Moreover, independent of aspect, a

Acta Phytogeogr. Suec. 80 46 M. Diekmann

Table 13. Quercus robur-Tilia cordata community. Number of vascular plants (Vas), number ofbryophytes (Bry) and total number of species (Tot), as well as indicator figures forlight (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for different sub-units. Average values are given. The sub-units correspondto: 1-3. Geraniumsylvaticum sub-community (1. Geranium robertianum form; 2. Deschampsia flexuosa form; 3. Lathyrus vernus form); 4-5. Stellaria holostea sub-community (4. Cardamine bulbifera form; 5. Deschampsiaflexuosa form).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 30.7 1 . 1 31.8 4.7 5.1 3.8 5.2 6.0 5.6 2 32.2 2.8 35.0 5.3 5.0 3.8 4.9 5.5 4.7 3 32.1 1.1 33.3 4.8 5.1 4.0 5.0 6. 1 5.1 4 25.9 1.1 27.0 5.0 5.3 3.7 4.9 6.2 5.1 5 25.5 1.4 26.9 5.3 5.2 3.7 4.9 5.3 4.7

slope shows comparatively higher minimum tempera­ nated by a few species occurring in large populations, tures and runs less risk of frosts (Sjors 1967). such as Poa nemoralis and Convallaria majalis. Tilia is (3) edaphic conditions: Due to the mentioned instability, often host tree for Viscum album, here at its northern new soil surfaces are constantly laid bare. At many places, distribution limit (Wallden 1961). ground water comes to the surface, creating more or less Stands of the three forms differ considerably with stable moisture conditions, causing a permanent transport respect to environmental conditions. The indicator figures and enrichment of minerals and nutrients, and preventing for M and N for the Geranium robertianum form are the leaching of the soil (Sjors 1967). The effect is most comparatively high, indicating a favourable moisture and pronounced on calcareous bedrock. However, this can nutrient status of the soil (Table 13). The stands of the also be observed in some areas of siliceous bedrock which, Deschampsia flexuosaform, on the other hand, show low in small quantities, can contain calcareous or other base­ M-, R- and N-figures, indicating quite dry, acid and rich materials transported to the surface by the water nutrient-poor soils. These are usually shallow and devel­ movement (Halden 1950). oped as rankers or brown earths, often with a high per­ A correlation of the physiographic variables with the centage of organic matter, where the humus form is a mull releve scores on axis 1 of a CA -ordination of the mainland or mor. This is confirmedby soil analyses of correspond­ stands gave the following coefficients: aspect: rs ing forests described by KlOtzli (1975a,b), giving pH­ = 0.173, n.s.; inclination: rs 0.178, n.s.; heat index: rs values around 5.0 and base saturation-values of 60 %. A - 0.207, p < 0.05. The results= show the combined influ­ fairly open canopy is indicated by a high L-figure. The =ence - of aspect and inclination on the forest vegetation. forests of the Lathyrus vernusform take an intermediate On boulder-rich screes, (deciduous) forest vegetation position, with intermediate indicator figures for the three is absent or confined to the upper parts, and the rocks are edaphic factors. The soil type is usually a sandy or loamy densly covered by bryophytes and lichens. Occasionally, brown earth with a mull humus. For corresponding for­ differentcommunities can be found on one and the same ests, Klotzli (1975a) gives pH-values between 5 and 6 and slope, with oligotrophic forest vegetation on the upper high base saturation-values (ea. 80 %). slope, a mesotrophicfo rest in the middle part and eutrophic The mesotrophic forests growing on more or less steep forest at the foot of the slope, the latter favoured by the slopes were less exposed to activities such as cutting and downward transport of water and nutrients (cf. Wallin grazing than those on level ground. Their species compo­ 1973). sition might therefore have been less influenced, particu­ Other sites preferably occupied by mesotrophic for­ larly with respect to woody species. The habitat instability ests are land areas close to large water masses which as described above probably causes a constant change in create a comparatively warm and even temperature cli­ the environment and, subsequently, in the vegetation, mate. Especially stands dominated by Tilia are often very strongly affecting the regeneration processes of trees. confined to lake shores, small islands in inland lakes and islands in the inner archipelago of the Baltic Sea. Particu­ Stellaria holostea sub-community (releves 65-92) larly in Lake Malaren, many islands with pure linden­ The mesotrophic forests of the coastal areas in Smatand forests can be found, often expressed in local names, e.g. and Blekinge (one releve from Bohuslan) are differenti­ Lindholmen and Lindskar (cf. Almquist 1929). Kers ( 1978) ated by the southerly distributed species Stellaria holostea, described some thermophilous forests from LakeMalaren, Melica uniflora and Mercurialis perennis. Besides, they growing on boulder-rich, more or less outwashed, but lack several species which frequently occur in mesotrophic nutrient-rich moraines unsuitable for cultivation. These forests located further to the north and west. isolated stands are floristically poor and usually domi- Two forms are distinguished: a Cardamine bulbifera

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 47

form (releves 65-82, corresponding to the mesotrophic precipitation, de Martonne's index, soil moisture) and mixed deciduous forests with linden on bland), character­ with those connected with temperature (annual mean tem­ ized in the canopy by Tilia and Acer and in the field layer perature, latitude). R- and N-figures are negatively corre­ by, e.g. Melica nutans, Cardamine bulbife raand Lathyrus lated with releve scores on axis 1 (and yearly precipita­ niger, and a Deschampsia flexuosa form (releves 83-92, tion), indicating that the high precipitation may be interre­ corresponding to the oak-hazel forests on bland), with a lated with increasing soil acidity and decreasing fertility. few weak differential species (Deschampsia flexuosa, Like in Table 7, L-figures are only weakly correlated with Campanula persicifolia, Rubus fru ticosus agg.). releve scores on both axes. Here, a good correspondence Studies of these forests, mainly referring to the Carda­ is found between the temperature variables and the T­ mine bulbife ra form, were carried out by Du Rietz (1925) figures. and Ottosson (1965) on the island BUt Jungfrun, by Berg­ The west Norwegian stands (Aune 1973; Fremstad lund (1963, 'Quercus robur-Stellaria holostea-samhalle') 1979, in the upper, right hand part of Fig. 14) differ in Blekinge and by Riihling & Tyler (1986) in eastern considerably from the stands located further to the east. Smaland. Riihling & Tyler compared the oak forest veg­ Of the demanding deciduous trees only Ulmus glabra is etation between Skane and easternSmaland. The floristic present, together with some trivial tree species. Although differences between these regions were considerable, but the shrub and field layers are not species-poor, they lack probably not only due to the geographic separation. Cal­ many deciduous forest species, especially those of an culated indicator figures pointed at lower pH-values and eastern,continental distribution. On the other hand, some higher N-values for the stands in Skane, probably caused arctic-alpine species can be found, such as Sedum rosea by the high levels of acid and nitrogen deposition from the and Alchemilla alpina. The bottom layer is very species­ nearby continent (Riihling& Tyler 1986). rich, with many bryophytes of a more or less pronounced Most stands presented here are located in the vicinity oceanic main distribution, e.g. Plagiothecium undulatum of and to the north of Kalmar. They occupy slightly and Rhytidiadelphus loreus. sloping ground and morainic, often boulder-rich hillocks. The stands in S and SE Norway (A. Bj�rnstad 1971; The soil types are rankers or brown earths. Indicator Kielland-Lund 1981) have a species composition similar figures for R and N suggest lower pH-values and nitrogen to the Swedish ones. Oceanic and sub-oceanic species are contents of the soil in the stands belonging to the much less frequent than in the west Norwegian stands. Deschampsia flexuosaform, compared with those of the The Norwegian forests are characterized by species which Cardamine bulbife ra form (Table 13). If these differences in Sweden are confinedto the eutrophic elm-ash forests, in edaphic conditions are primary, or if they are the such as Campanula latifolia and Stachys sylvatica. On the secondary effects of soil deterioration due to over-exploi­ other hand, pure elm-ash forests have not been described tation (the sites occupied by oak-hazel forests have prob­ and are probably lacking in this area. Thus, in Norway, a ably been exposed to strong human influences), is not division into mesotrophic and eutrophic deciduous forests known. is hardly possible any longer (cf. A. Bj�rnstad 1971; Aune 1973; Fremstad 1979; Kielland-Lund 1981). The mesotrophic forests in the Aland archipelago Regional comparison (Hinneri 1972) resemble those in eastern Sweden. How­ In order to compare the mesotrophic forests in different ever, Acer and especially Fraxinus are the dominating parts of the Boreo-nemoral zone and adjacent areas, a tree species, whereas Quercus robur is rare and Tilia table was compiled, containing releves from the literature completely absent. The high frequency of some orchids as well as the releves presented in Tables 8, 10 and 12. (e.g. Dactylorhiza sambucina and Platanthera chlorantha) Altogether, 362 releves from Norway, Sweden, Finland and species of forest fringes such as Laserpitium latifolium, and Estonia were used. The results of the CA-ordination Melampyrum cristatum and Clinopodium vulgare is re­ of sample plots based on this table are presented in Fig. markable, indicating a fairly open canopy. The stands on 14. The correlation coefficients between releve scores on the Finnish mainland (Tapio 1953; Koponen 1967; axes 1 and 2 and explanatory variables are given in Table Makirinta 1968) differ considerably from those on the 14. Aland islands. They are located outside the continuous In the ordination diagram, the releves from different distribution area of Quercus robur, Ulmus glabra and regions are quite well separated. A TWINSPAN cluster Fraxinus; besides Tilia and Acer, the boreal trees Picea, analysis (Table not shown) gave similar results, i.e. a clear Pinus and Alnus incana reach high frequencies. In the separation of clusters comprising releves from different field layer, many southerly distributed species are lack­ regions. The correlation analysis reveals that the releve ing, whereas acidophilous, northern species such as scores on axis 1 (eigenvalue 0.392) and axis 2 (eigenvalue Vaccinium myrtillus and V. vitis-idaea are more frequent 0.299) are most strongly correlated with the climatic than in the other regions. variables, both with those connected with moisture (yearly The mesotrophic forests of Estonia (Linkola 1929;

Acta Phytogeogr. Suec. 80 48 M. Diekmann

Table 14. Mesotrophic mixed deciduous forests: regional corn- its southemmost parts, such as Lamiastrum galeobdolon. parison. Spearman rank correlation (rs) between sample plot Although the species composition of the mesotrophic scores on the first two CA-ordination axes (Fig. 14) and 11 forests differs widely between regions, they have a 'pool' environmental variables. The number of observations is 362. of common species separating them from other forest For further explanation, see Table 6. types. They can, therefore, be regarded as belonging to a

Variable Axis 1 Axis 2 group of related communities, divided into several geo­

rs p rs p graphically vicarious communities.

L -0.070 n.s. 0. 123 0.05 T -0.544 0.001 0.123 0.05 c 0.485 0.001 -0.175 0.001 4.4 Eutrophic elm-ash forests M 0.263 0.001 -0.242 0.001 R - 0.48 1 0.001 0.116 0.05 N -0.173 0.01 0.279 0.001 General characteristics TEM YEAR -0.577 0.001 0.636 0.001 PRE YEAR 0.568 0.001 -0.382 0.001 Species composition and structure MAR IND 0.676 0.001 -0.528 0.001 Eutrophic elm-ash forests are usually dominated by LAT 0.492 0.001 -0.506 0.001 Fraxinus excelsior and Ulmus spp. Whereas Fraxinus -0.484 LON -0.045 n.s. 0.001 shows particularly high abundance in the upper canopy, Ulmus and Acer platanoides are more frequent in the lower tree layer. Quercus robur has a much lower fre­ quency than in the mesotrophic mixed deciduous forests Kalda 1960) are located in the east European part of the and is locally absent. Tilia cordata occurs only occasion­ Boreo-nemoral zone, outside the distribution area of Fag us ally. The shrub layer is less dense and diverse than in the sylvatica and Carpinus betulus. They show a surprisingly mesotrophic forests and mainly composed of Corylus clear similarity to the stands in SE Sweden, particularly avellana and young tree individuals. The field layer is with respect to the tree layer. Also the field and bottom characterized by high cover percentages reaching often layers are quite similar, except for the occurrence of a more than 90 %, both in spring (high abundance of number of species which do not reach Scandinavia or only geophytes such as Anemone spp. and Ranunculus spp.)

2 N (/) X <(

1

0

-1

-2 1 • I

AXIS 1 -3 y------,------,------.------� - 2 -1 0 1 2

Fig. 14. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 362 releves of mesotrophic mixed deciduous forests. Numbers plotted onto the sample plot positions represent releves from this study and literature releves. The arrow indicates outliers not visible in the diagram. Estonia: 1. Linkola (1929) 29 releves, Kalda (1960) 12 releves; Finland: 2. Tapio (1953) 12 releves; 3. Makirinta (1968) 19 releves; 4. Koponen (1967) 11 releves; 5. Hinneri (1972) 34 releves; Sweden and Norway: 6. 197 releves fromTables 8, 10 and 12; Norway: 7. Kielland-Lund (1981) 30 releves; 8. A. Bj!ISmstad (1971) 20 releves; 9. Aune (1973) 5 releves, Fremstad (1979) 13 releves.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 49

6 N � X 5 <(

4

3

2 2 �. 6 2 1 • 1 s' . 3 � 1! 3 5 �¥ i/ � !� i � � 0 � � ;�H? . · · 5 �� . ;�1 2 5 5 • . s' �! . \ -1 6 • � 3 . ? 4' 4 • 4 4' -2

AXIS 1 -3

-2 -1 0 1 2 3 4

Fig. 15. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 90 releves of eutrophic elm-ash forests. Numbers of different sub-units plotted onto the sample plot positions which are slightly adjusted in order to avoid overlapping figures. 1-4. Ulmus glabra-Fraxinus excelsior community (1. Allium ursinum sub-community, Tilia cordata form; 2. Allium ursinum sub-community, Loniceraxylosteum form; 3. Gagea lutea sub-community, Convallaria majalis form; 4. Gagea lutea sub-community, Typical form); 5-6. Ulmus minor-Fraxinus excelsior community (5. Geraniumsylvaticum sub-community; 6. Typical sub-community).

and summer, but shows a relatively low species number, 2. Ulmus minor-Fraxinus excelsior community (releves since many species are lacking which frequently occur in 66-90), in releves from bland (two releves from eastern mesotrophic forests. Tall growing herbs such asAnthriscus Sweden) and differentiated by Ulmus minor, Anemone sylvestris, Aegopodium podagraria, Campanula latifolia ranunculoides and some moisture-demanding species, and Stachys sylvatica (the two latter differentiate from the e.g. Filipendula ulmaria and Geum rivale. mesotrophic forests) often give a characteristic luxuriant The differentiation into these two communities is also impression. Mercurialis perennis occurs only occasion­ supported by the ordination diagram in Fig. 15 (sub-unit ally, but is often very abundant with cover degrees of up to numbers 1-4 and 5-6, respectively). 90 %. Usually, the bottom layer is well developed, fa­ voured by the fast decay rate of the litter layer and the Affinities mesic to moist soil. The eutrophic elm-ash forests have earlier been recog­ Elm-ash forests represent the most fertile and impres­ nized and described as a separate forest community of the sive deciduous forests within the study area, forming 25 to Boreo-nemoral zone, e.g. as 'alm-askskog' or Ulmo­ 30 m tall stands. The separation into two tree layers is less Fraxinetum(Kielland- Lund 1971; Bnl.kenhielm& IngelOg distinct than in the mesotrophic forests. 1972; Ekstam & Sj ogren 1973; Diekmann 1988). Sj ogren (1964) and Sj ors (1967) mentioned forests with Fraxinus Differentiation excelsior and Ulmus spp., accompanied by a 'Mercurialis­ Two communities are distinguished, occurring in differ­ sarnhiille'in the fieldlayer. Other synonyms are 'Fraxinus­ ent geographic regions: geofytlund' (Almquist 1929), 'Anemone hepatica-sam­ 1. Ulmus glabra-Fraxinus excelsior community (releves hiille, Aegopodium variant' and 'Mercurialis perennis­ 1-65), in releves mainly from the western and northern sarnhiille' (Wallin 1973), Ranunculo (f icariae)-Ulmetum 'Mercurialis pe ennis parts of the Boreo-nemoral zone, and differentiated by (Klotzli 1975a) and r -typ (Ekstam ' Prunus padus, Actaea sp icata and Dryopterisfili x-mas, 1979). All of the mentioned studies emphasize the par-

Acta Phytogeogr. Suec. 80 50 M. Diekmann

Table 15. Releve table of eutrophic elm-ash forests. (See Table 3.)

Running number 12345678 91111111111222 2222222333333333344444444445555555555 666666 66667777777 77788888888889 0123456789012 3456789012345678901234567890123456789 012345 67890123456 78901234567890 Cluster 14a 14b 15a 15b 16a 16b Cluster code 101000 101001 101010 101011 10110 10111 Region HB6606b68N6NVN 8888UVVNUUUUUVUUVSU8U668UUVU668UUNNNV 66666666686 66666666668666 VVVNVVVV uuuuou AAAoAAAA AOGGGGGGMAGAAO 66M6PAAOPPPPPAPPA6P6PGGOPPMPGGMPPOOOA PPPPGP LLLLLLLLLML LLLLLLLLLLMLLL Aspect WWWW8WWWWN--NW 88WE-NEW8N-EWW-8EEE8-W88W- -WEWNW88W8E W88888 ------N-WWN ------W-E--- WWWWN-WN N W W W 8 W N 8 E 8 W W88W EW8 W W N W W W E WW w w w w Inclination 55413-11 1251115115--14 2 521-22 3 53-534-14151-13 51--51222413 8 5 555211 ------1-134 ------4-4--- (0 ) 5 00 05 005 50 05 0 00 505 0 000 0 0 5 005 00 505 0 5 Cover T 1 (x10) 76747776 67 556345654564 5565576736673746745776577376745565777 657575 87677757645 86766767664888 % T2 (X10) 44445545 32354573338713 3453 6343 64 516-62246 5 66642143444555452 663444 12-111-1756 15374433775323 % 8 (X10) 23232123 13465546474216 3545232434344664155423354744555434231 221485 15644877163 44213222124421 % F (X10) 87959999 99889899968695 77777883762 55675276385665756776874249 988485 98999998867 98787777882894 % B 21111213 1151411541165- 1133532-1111111115--51-2 111115525-511 -51-11 -51-1511--1 311-1251-5-236 % 00 0 00 50 0000 00 0 00 05 0 0 00 0 00 0 00 5 000 00 0 000 N - Tot 23211122 22332232223311 43444442 3222 3 3 3 322322333223 43 33322223 132333 32243345325 24423422331222 98329765 10518905660947 0924653839572 146696373169503966351386 710322 19622644162 31198099644832 N - Vas 23391122 11222222222311 3 3 3 3 43 3 2 3 2 2 2 2 2 3 3 2 2 3 2 2 3 3 3 222 3 3 3 3222223 1213 3 3 32243334324 13323322321211 753 8603 78872340009027 9859274818467923576340118388320821152 788310 17521298167 98992486694599 N - 23231162 4274666566192- 1175489-2111522312--33-5122564653-234 -32-12 -21-1456--5 432-6613-5-343 Bry nUlmus minor 23324522 .2.767

Fraxinus excelsior . 3434443 44545555355564 455455554544 ..53433 . 6334 ...5545 444443 545 ... 26565556656 6556335355 ..4. Ulmus glabra 656 . 5655 5645534 .63.3.4 5.4543544444 .636654645 .644 . 354355 .656 555565 3 ...... 224323 .. . Quercus robur . . . . 3.34 ...... 433 ..34 .35. 4 ...543454 .46 ...3 ...... 4. 4. 4 .34.5542242 24 .2.332435 ...

Acer platanoides 4434 . 344 ..4 .....4 ...... 4.4 55 ...3 ..3 ...... 44 ...5. 43 . . 34 ... 6 ...... 2 ... .34.5545 .43 ... Tilia cordata 34 ..54 ...... 44 .4.3 ...... 4

Popu lus tremula ...... 3 ...... 34 ...... 23 ...... 3 ...... 22••••••••••••• ...... Betula pendula .4 ...... 33 ...... 3. 3.4 ...... 2 ....3 .. .

Alnus glutinosa ...... 3.6 . .. 5 . ....43 ...... 2 ......

.

:uUlmus minor ...... 2 21434444 .2.543

Ulmus glabra 55545555 34455563 .456.3 4444545335535 . 34445556654354453453553 564555 .2 ...3 ..455 342 .2244554 .. . Acer platanoides 4344 .244 ..444 .44 .. 52 .4 44 .44 . 44444 ...4324434 ....3 ..445434434 43 ..2. 32 .....25 .. . 44 . 5444455 .. . Fraxinus excelsior ...... 3. 33433444 .44.44 . 43 . 3. 3. 53 . 35.44.44.424 .3 ..4.4.4 35 .. . .43 ... .3.423 ..244 ...... 4 .... . Tilia cordata . 3 ..4 ..3 ...... 3 ... 34433 ..4 ...... 3 ...... 2 ...... 2 ..2 ...... 8orbus aucuparia ...... 3 ..3.43.4 3 . ..5 ..4 ..3 3 ...... 3 ..2

Prunus padus ...... 3.3 ...... ••••••...2 3 ....3 •••• .....3

Quercus robur ...... •••••••• •. . . 3 ...... •••• . ...4 ...... 2 ...... 2 ...... 2 .....

...... • • • • • • • • • •

.aPrunus padus .24 ...... 24 .23223 ...... 3. 3. 2524 ...... 44 ..43 . . . 444 . 443 ..3 3 8orbus aucuparia ...... 2 . 24444 ..222 ..2 . 3422 .2232 . 342222 .22 ..42 ..3. . 23 . ...23222 .2.... 1. . 2.

Ribes alpinum ...... 2 ..3332333 .2 .. . 323 .2.2. 3. 2 ...32 ..22 .2 ..222223 3...... 23 .. .1. ....1.3 4 Ribes uva-crispa 2 ...... 2 ...... 2 ...... 222 ....2 ..2 .....2.2 .22.3.. 3 ..22. ...2 ...... 3 Lonicera xylosteum ..4443234224 .. . 3443423 ..22 .2442.4.. 2 .....4. 44 . 2...... 3 .. . ..1.. 52 ..2 . .1...... Cornus sanguinea . . . 22 .4.3.1 . Euonymus europaeus ...1 ..11 .. . 2

Ulmus minor ...1 ...... 3 334• ••645 •••••••••..3 ...545

Corylus avellana 44244 .44 44544446464 .46 554543233544 . 544 . 5444 ..43 643 . 55434232 34 .5. 5 25644656 .44 4533434 ....2 .. Ulmus glabra 23442323 2244455 .4444 .3 3 ...4443424344444 .3424453 .44444342443 .42343 .....1 ..254 . 4 ..2. 33345 ... Fraxinus excelsior 2343.234 24344333433 .23 . 234342233 . 4434 ...43334 45444433334232 ..2. 3. . . 342131242 1...... 2.2 12 . Acer platanoides 22 .32.23 ..34 .33 .433 ..2 ...... 3 ..24 ..3 .4332 .22 ....344 .33 .34 3 ...2 . . 2 .....11. . . 442 ..132232 .. Crataegus spp ...... 2 ...... 2.2 ..3.2 2 ...... 2 ...... 323 23245543344 53343 . 2332 .34 . Tilia cordata 242 .3323 ...... 3 .. . .4 ..3 ...... 2 ..2 ...2 ...2 ....2 ...... 2 .... 2 11 . Malus sylvestris ...... 3 ...... 22 ...... 1 ..1 ...... •.•••...... 1 .. Viburnum opulus . . 2 ...... 2 ...... 2 ...... 1.1 ...... 11 ...... Prunus avium ...... 3 ..2 ....4. 2 ..43 ...... 1 ......

.

.EAll ium ursinum 346. 5576 ...66 .77 ..6.5. Prunus padus 113 .1. .. . . 13 .. 22 .2 .1. . ..3. 3 223 .2 ...1. 112 .31.243 ..2 .....2232 2.1. Dryopteris filix-mas ..2 13 .1.24122 ....2. 33344443 .32 .....34.2.2 ....3 ...44 .4 ...... 2 . ...1.1 ......

Actaea spicata ...... ••• 2 1. .. 3 ...11. 112 34.4332 ..433 . 342442 .23 ..442 4.1.43.11. ..11 ...... 1 ..3. Geranium sylvaticum .1. ..1.1 2.24414312 .22. 34 ..324 .4.235 .23 ..4 . 1 . 121 . 123332 ....2 4 32..... 2234 ...... 1 .. .. 31.3. Viola riviniana 24 ...... 1 ...... 2 .33241 .12 ...2 ..1 . 122 . 2 . 33. 4334322332 . . ..12 . 2 ..2.1 4 ...2 ...... 1 ... 8tachys sylvatica 3.4.3432 ...342 ..3 ..44 . . ..4345 .3 ...... 2 ..1. 3 ...... 3. 43 .34 ..4 244424 3 ..21 Melica nutans . 1 ....1. ..112 .2 ...33 .1 ..32.1 2. 3.12 ..2 .1.2 ..2 ...... 24 .1. ...2 ...2 .. 1• ..•••••1.1 ..... Ribes alpinum ..3222132 .1. .. .12 ..1.. 32 .1. . 2 ...23 .3.. 3. 3 .121. ..11 .. 2 3 Lonicera xylosteum ..33 .2. 22214 .. ..32 ...... 11 ..2 .1.2 . . 1. ...12 . 2 ..2 ...... 2 . ...1 .. •••••••...... 2 ...••1 Aegopodium podagraria ....3 ...... 44435 ..2 ...... 4 ...4 .....3 ..3. 625 .....444 .....2 7664 .4 . 2 ....142 .. . ..1 ...... Convallaria ma jalis ...2 ...... 4 .... 35.54241 .4 ....33.5 ..2 ...4. 2 ....2. 2 ..3 ...... 31 .44 2.1...... 1. Mycelis muralis ...... 2 21.41 ...... 1231 ...2 .3.22222 ...... 3 .. ..1...... Campanula trachelium ...2 ...... 3 .....2 ...13 ...... 1. ... 11111 .1 2 ...... 121 .1 .11.1. ..1. . Oxalis acetosella .. 4233 ...... 3. 4 ...... 2 ..3 ...1 ...... 1 ...... Ranunculus acris . 3.3...... 2.3 ...... 1. 31.2 ...... 12 ..1. .. Crataegus spp. 12 ...... 13 .2. 13 ....2 ...2 ...... 1. 2 .....1. . 2. .2.21. 23112222123 122111 .11 1.111 Gagea lutea 3 ...4 ...... 4 ..3 ....4 ...4 ...33 ... . . 334.3 .4.34314 ..3 .3 ..443314 ...3 Primula veris . 1 ....1.3 ...13 .1 ...... 1 ....1.1 ..... 14 . . . 31. 2 ....1. 41. . . 33 ...... 1 .... Milium effusum 31 ..3 .....1 ...... 23 ..2 ... . . 2 .. 3. 3 ...... 2 .. .. 3 ...... Anemone ranuncu loides . 4 . 44... 34.323453 .4 553 5455453 .545 Geum rivale .1...... 3 ...... • . . . . 33.2 ...... •••••.•.••••••..••••2 ...... 2 ...... •••••• . .2332214 ... 45544555 .3 .14 .

Filipendula ulmaria • • . . . 4 .. 2 34 ...... 234424423. 4 ..422 ..13 .14 . Rubus caesius ....1.1. • • • • • • • ...... • •••••••••...... 4 . 2443343 ..2 3142 .2 .....22 . Melica uni flora 4 5. . . . • . • • • • • • • • • • • 2 • • • • . .••••••••••• 6 ....34 ...4 .1. .433313 ... .

Alliaria petiolata •• ••••••••••. 42•....••••••••••• ...... •••••••....•.••••••...... 2. 3 ...... 2444114 ..3 ..... Euonymus europaeus ...... 2 . 22 .1121.1. ... Viola reichenbachiana . . . 3 ...42 . 3 ...... 1 ..... Corydalis bulbosa ...... 1 ...... 14. 44 ... . Orchis mascula ...... 3 . 112 ....1 .... . Ulmus minor ...... 1 . 1223443 .1.31 2

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 51

Ta ble 15, cont.

Anemone nemorosa 454 . 5444 65554544455541 4 5 6555 51. 6 556 5653 55644 565 65555465 ..3 3 344445 64446535356 45465555547432 Fraxinus excelsior 33343424 34443422333334 3444433444233144324433454434444444242 . 2 ..3. ..14 1244 .44 22314344232212 Acer platanoides 22322323 . . 4322 . 23242 .2 42224 . 3342344 .3313323 . 33 . 31.34422 .443 323 .2 . 4211. .13331 144 . 4344431.11 Geum urbanum 232 ..2.2 1224.1.12.2 ... 12433 . 42422343323.4412344344344424323 433444 41444413322 42124333342423 Ulmus glabra 22432322 223 .233 .2423 .3 22224 .3. 233322242 . 242443313423323 ..23 .1134 . 1 .... 1. .223 12 ....3.4 42 ... Ranunculus auricomus agg. 43 ...1. . 442 ...... 223 .. 43234 . 1 .. 434434 ..34243444 434121 .....4 244 .22 44 . 32343443 44444444444453 Ranunculus ficaria ..3 .... . 5613 .32.2 ..... 54 .35 .....4 .4422 ..33 ..4 ...44 .4424 ... . .66746 4 .45444 . 54 . 65445655665577 Paris quadrifolia 32 ..1. .. 3 ....1. .. 333 .. . 443321 ..43 ..244 ..3. 3242 .4.31. . 441123 .2 .... .4442123 .3. . 442444322 ... .

Poa nemoralis . 12 ..... 22 .33331. ..3.1 4233 5. 323124 .23212 . 1. 23321123233 ....2 . 1.333 4 ....4 .. 1 .. .13.11.1.1. . . . Hepatica nobi lis 32 .3.1 .. 444424224242 .4 444444423344 .4. 4344122 . 335 .434442 ..34 .4.423 4.454442244 . 43443412 .... Taraxacum offic inale agg . . .1...... 222 .. 212 ...21 . ....4. 1121..2323 124 .221 . 1313 32 ..21 21.31 .122 . 3 31.11 1. . 2 ..221 .

Anthriscus sylvestris 1.2 .... . 13 .1.12 ...223 . 1.1. .2 ..3 ..2221 ...... 1. . 1323 .4 ...... 31 . 333 2 ..1443 .1. . . 2. 3111244 ..

Vicia sepium .1 ...... 22 ..1. ...12 . . 3. 34 ...31222 .3 ..13 . 232 . 22233322 .22.3 .1.344 2 ...24 .1. . .11.1 ......

Sorbus aucuparia .1. ..11. . . . . 232 ....1. ..2 ...2 ..12 ... 21321 . 12 . 11222221121 . 21221 ...1 .. .1..... 2 ... 1 ..1 ...... Allium oleraceum . . . . 234 ...3 ... . 4. 3 ...... 233 ..4 ..22 .3..... 42 ..42 ...... 4 . . 2 ..3422343 444433433444 .4

Polygonatum mu ltiflorum 42 .. 422 . . 413 ..33 .233 .. . 232 .23 ...... 22 ...... 14 . 3.35 . 442. . 3 ....12 ... . 54 .. 11.. 2 .... Mercurialis perennis 65657655 776 ..6 ..7 ...75 4 45 .....6 467676646 .4 7 ..6 ....5 ..664

Corylus avellana ...... 11 .11212 ...32 .12 ..• ••••••••••••213 ...... 1.1. ••••••••••••••. ...1. ...1. . 22 .2. 2 ..11 ...... 1 11 ...... 132 .1121 .1. Viola mirabi lis 44 .....2 ...... 2 .2 2 ....243 ..3 ...53 ..4. 4 ....4 ...... 413432 .3 ... . 4423323 ......

Campanula latifolia 12 .....3 . 4 ....24 ..2 .. . 22 .2 ..43 ...... 23 ....3 .....4 .....34...... 12 .4.. 1.2.444 4.4.2 .. Elymus caninus 334 ..122 .3.2234 .... 3 ...... 2 ...3 ...2 .....2.2 ...... 4.444 2 ..52 ..1. .1 .2.. 2 ......

Listera ovata 32 ..1.1...... 312 .. .2.2 ...... 1 ....3 ...... 3 ...... 31. ...4.4 1 . 3242222 ...2 ... Galium aparine .1 ...... 2 •••.••••••• ....1 ...4 ...... 1 .....31 ...... 2 .1433 . ..344 ..1. . 131442123 ...

Pulmonaria officinal is •. .• 42 .1.... 3 ... ..3. 41 ...... 1 ....3 ...... 1444 .....4 . .... 5 2.1214 .3 ...... 21 ..1 .....

Viburnum opulus . 1. .1. .. ..31.1. 2 ..1222 . 2 .1.12 ....2 ..21 • ...11 ...... 1 . ..3 ..1. Ribes uva-crispa 2 ...... 1 .. 1 ...... 22 ....22 .21 ....21 ..11 .1.. 2 ..222 ...1 ...... 1 ...... 1 Maianthemum bi folium 3 ...... 1.4 .. . 322231...... 2 ...... 13 ..4 ...3. ..3 ....4 ... . 33 ......

Deschampsia cespitosa 2 ...... 2 .. .2413 .3 .....3. 1...... 22 1...... 1. ... 1 ...... 1 ...... 1 ..1 ...... Rubus saxat ilis 13 .....2 ...... 3 ...22222 ....31 ...... 1.... 2. 4. . .1 . .. .

Tilia cordata 12 ..212 ...... 2.22 ...... 2 .. ...1 ....1 ...... 2 ...... 1 Quercus robur .1...... 11 ...... 2 ...... 21 ...... 1 ..1 ...1 . . .1 ....1 ..1 ... ..1 .. Urtica dioica ...... 1. . ...3 ...2 ....1 ...... 3 ..11 ..2 ...... 214 ...1 ...... 1 ..

Populus tremula . . . 2 ...2. 2 . 2 ....2 ...... 1 ..1.1 ...... 1 ...... 1 . 22 ......

Cardamine bulbifera ...... 3 ...... 24.3 ...... 4 ...... 4 ...... 4.2 . 3 ...... 43 ...... 4 .... .

Carex digitata . . . 2 ...... 2.2 ....1.2.3 ..32 ...... 3 ....1 ...... 1 .

Lathyrus vernus 11 ...... 2 ...21 ...23 ...1 ...... 4 ...... 1.

Carex sylvatica 23 ....3 2.1...... 1 ....2. ...2 •.•4

Veronica chamaedrys .1...... •...... 3 ..2.2 ...2 ...... 2 ...... 1 .. ...2 .••

Geranium robertianum ..4...... 1 ...1 ...... 4 ...... 32 ...... 1 ...... 3 ....

Allium scorodoprasum ...... 5 4 ...... 4 ..... 3. .2 .....434 .. .

Dactylis glomerata 1 ...... •...... 1 ...... 3 ...... •... 1.1 ...... 1 .. .• ....•...... 1 ...... 1 ......

Athyrium filix- femina . . . . 13 ...... 2. 2 ...... 3.

Rosa spp . 1 ...... 11 ...... 2.2 1 ...... 1 .. 1

Prunus avium ...... 1 ...... 1 ..1 ....3 ....22 ...... 1 .

Dryopteris carthusiana ...... 1 .. ..2 .....1 ...... 2 ...23 ...... 1 ...

Fragaria vesca . . 3 .....21 . 221 ...... 1...... Lu zula pilosa . 1 ...... 1 ...... 1.1 ...1 ...... 2 ...... Fagus sylvatica ...... 2 .2 ...... 2 ...... 2 ...... 1 ...... 3.

Moehringia trinervia ...... 1 ...... 2 . .. .3.2 ...... 1 ..1 .. Lapsana communis ...... ••.•...... •...... •••••1 ...... •••••••••••..1 . ....2. ...1 ...... 1.1 Hieracium murorum 1 2 2 . .1 ...... 3

Ranunculus repens ....• 4 ••••••••.•••••••••••• ...... 2 . ..2 . ..2 ...... •...... •.. . 2

Carex montana ...... 2 ..... 2.1. ... . 1. ....1 ...... • •••••• .•. Viola hirta 2 ...... 1 ...1 .. 1 .....1 ...3

Carex muricata agg ...... •...... 1 ...... 1 ...1.1 ...... 1

.aEurhynchium angustirete . 2.2 ..33 ...4532 144 .4 .. ..1 ...... 4 ..... Rhyt idiadelphus triquetrus . 4 ....3 ... 2.31 .....4 .. ..41 .4 ...... 2 ...... Eurhynchium striatum ...5 4 ••••••••••...... ••• ...... 2 4 ...24 ..... 45 . Cl imacium dendroides .•.•..••.•••• 2 ...... 31.

Eurhynchium hians 44341443 4433444453254 . 1144344 . 1224232422 ..41 .32 11444444 .444 . 44 2 .3 ...344 .. 4 .44.4444.2.4.5

Brachythecium rutabulum 1...... 123 .4.2333 .3 .. ..23222 .....1122 ....2 ..3 .1.31.11...... 11.•• . . .1.1.4 ...2 3142 .2..... 22 .

Plagiomnium undulatum 4.3.1 2424 ..4 .. ..43431 ...... 2 ....1. .2 ...223122 ...1 ....2. . .1. ..244 ..1 1. ..12 .1. .. . .

Cirriphy llum piliferum 1.2.33324 ..42 . ..3.1 4 4.1. ..2 ...... 1.3 ..1412232 . 131 ...... 1.1 ... . 1 ..3 ....1 ... . . 1 .... Fissidens taxifolius ...... 2 ... 1 ...... 3 ...... 1 ...... 2 ... 4 ....1 ...... 1.

Plagiomnium affine ...1. ... 1 ..3 ...... •11 ...... 1 Plagiochila porelloides ...... 1 ...... 1 ....3 ...... 21 ......

Addi tional species (occurring in one , two , three or four releves ):

4X: (T2) Alnus glutinosa 9:4, 50:3, 69 :1, 73 :3 (S) Fagus sylvatica 9:3, 10:3, 44 :3, 64 :3, Sambucus nigra 72 :2, 87 :4, 88 :1, 89 :2 (F) Brachypodium sylvaticum 30:2, 69 :3, 73 :3, 74 :1, Bromus benekenii 23 :2, 52 :3, 73 :1, 76 :3, Dactylorhiza maculata 28: 1, 78 :2, 79 :3, 80 :2, Daphne mezereum 19 :1, 20:1, 25:1, 55:1, Lilium martagon 43 :4, 60 :2, 61 :4, 62 :4, Platanthera chlorantha 69 :1, 78 :2, 79:1, 85 :1, Polygonatum verticillatum 28 :3, 29 :2, 56:3, 58:3, Scrophularia nodosa 9:1, 35:1, 39 :1, 71 :1 (B) Brachythecium ve1utinum 4:2, 51:1, 53 :1, 72 :1, Scleropodium purum 7:2, 18 :1, 53 :1, 86 :1.

3x: (T2) Fagus sylvatica 9:4, 10 :4, 62 :4, Malus sylvestris 63 :3, 73 :2, 81 :2 (F) Acer pseudoplatanus 19 :3, 51:2, 76 :2, Calamagrostis arundinacea 31:1, 32 :1, 37 :1, Epi1obium montanum 30 :1, 31:2, 39:1, Festuca gigantea 51:2, 64 :2, 69 :2, Galeopsis tetrahit 3:1, 23 :1, 41 :1, Malus sylvestris 15:1, 36:1, 76:1, Rubus idaeus 50 :3, 56 :2, 57 :3, Silene dioica 26 :1, 36 :3, 58 :1, Stellaria media 65 :2, 73 :2, 74 :2 (B) Fissidens cristatus 46:1, 51 :1, 52 :1, Isothecium alopecuroides 12 :1, 14 :2, 25:3.

2x: (T1 ) Betula pubescens 50 :3, 68 :2, Fagus sylvat ica 62 :3, 64 :4, Picea abi es 22 :4, 57 :3, Pinus sylvestris 4:4, 22:4, Prunus padus 26 :3, 28 :3 (T2) Picea abies 11 :3, 22 :3 (S) Acer pseudop1atanus 19 :2, 51 :2, Picea abies 7:2, 13 :3, Populus tremu1a 12 :3, 18:1, Rosa spp . 65 :2, 66 :1 (F) Arctium tomentosum 3:2, 84:1, Aquilegia vulgaris 19 :2, 58:1, Corydalis intermedia 37 :1, 54 :4, Equisetum arvense 5:1, 45 :1, Fragaria moschata 47 :2, 61 :3, Galium odoratum 23 :5, 27 :1, Glechoma hederacea 30 :1, 51 :1, Gymnocarpium dryopteris 30:2, 35:2, Hedera he lix 76:4, 85 :1, Heracleum sphondy lium 72 :1, 79:1, Hordelymus europaeus 74 :2, 76:4, Humulus 1upulus 39 :1, 56 :2, Lathraea squamaria 41 :4, 73 :2, Lysimachia vulgaris 45 :2, 69 :1, Sambucus nigra 87 :3, 90:1, Sanicula europaea 40 :2, 66:2, Stellaria nemorum 26 :4, 28:4, Valeriana sambuc ifolia 25:1, 30 :3, Vicia sylvatica 2:2, 7:2 (B) Atrichum undulatum 28 :4, 29:4, Eurhynchium praelongum 73 :1, 89 :1.

1x : (T1 ) Quercus petraea 10 :4, Sorbus aucuparia 28 :4 (T2) Acer pseudoplatanus 19 :2, Betula pendula 51:2, Carpinus betulus 74:4, Populus tremula 17 :3, Prunus avium 54 :4 (S) Alnus incana 17 :3, Carpinus betulus 74 :2, Sambucus racemosa 30:2, Taxus baccata 51 :3 (F) Abies sp . 32 :1, Alnus incana 57 :2, Angelica sylvestris 46 :1, Cardamine impatiens 20:1, Chelidonium majus 87 :1, Cornus sanguinea 76:1, Crepis paludosa 28:3, Dactylorhiza fuchsii 76:1, Equisetum pratense 28 :3, Gagea minima 90 :1, Iris pseudacorus 46 :1, Juniperus communis 36 :1, Laserpitium lat ifolium 23 :1, Lathyrus niger 23 :2, Melampyrum sylvaticum 11 :3, My osot is sylvatica 27 :2, Neottia nidus-avis 37 :1, Poa trivial is 69 :1, Polygonatum odoratum 27 :1, Rubus fruticosus agg . 70:2, Sc illa sp . 88 :1, Solidago virgaurea 25:2, Taxus baccata 51:4, Veronica hederifolia 86 :2 (B) Anomodon attenuatus 3:1, Brachythecium populeum 29:1, Brachythecium reflexum 17:1, Campylium calcareum 73 :1, Encalypta streptocarpa 29 :1, Homa lia trichomano ides 86 : 1, Homalothecium lutescens 50:1, Hypnum cupressiforme 11 : 1, Neckera complanata 90 :1, Plagiomnium cuspidatum 29 :2, Thuidium philibertii 20:1.

Acta Phytogeogr. Suec. 80 52 M. Diekmann

ticularly high frequencies of Ulmus and Fraxinus at steep, north-exposed slopes on calcareous bedrock in the eutrophic sites. In other publications, elm forests and ash vicinity of the Baltic Sea and is characterized by, e.g. forests are described as separate communities, e.g. by Lunaria rediviva, Gymnocarpium robertianum andActaea Bergendorff et al. (1979), Anon. (1982) and Anon. (1984). spicata. Corresponding communities - rich in Ulmus However, the elm forests described in these publications glabra, Fraxinus excelsior, Tilia spp. and Acer spp., with are more or less congruent with the eutrophic elm-ash Fagus less abundant - occur also in central Europe on forests as described here, whereas ash forests denote the climatically and edaphically favourable sites (Ellenberg alder-ash forests on wet ground dealt with in Chapter 4.5. 1986). It is often emphasized that Ulmus glabra prefers less wet Both Ulmus glabra and Fraxinus excelsior occur in sites than those tolerated by Fraxinus (e.g. Bergendorff et the Boreal zone, the latter only in the southernmost fringe. al. 1979). This has been shown for eutrophic forests with However, in the Boreal zone in Sweden, as well as in the elm and ash in Skane (Brunet 1991), as well as for Boreo-nemoral zone of SW Finland, elm-ash forests are corresponding communities in central Europe (Ellenberg absent. In SE Norway, both Ulmus glabra and Fraxinus 1986; Oberdorfer 1992). However, this does not seem to excelsior occur frequently in forest communities on rich be the case in Boreo-nemoral elm-ash forests: For the sites (Kielland-Lund 1981); some releves in Table 15 releves of Table 15, there is no significant correlation were made in the Oslo region. However, as has already between the cover degree of Fraxinus and the M-figures. been pointed out in Chapter 4.3, eutrophic elm-ash forests For the Ulmus glabra-Fraxinus excelsior community on and mesotrophic mixed deciduous forests cannot be clearly the mainland, no significant correlation was found for separated any longer in western Scandinavia, especially Ulmus glabra and the M-figures either. However, in the not in the more oceanic parts. Eutrophic forests in W Ulmus minor-Fraxinus excelsior community on bland Norway (Aune 1973; Fremstad 1979) are composed of with comparatively wetter soils, Ulmus glabra is partly Ulmus glabra as only one of the more demanding decidu­ replaced by U. minor. ous tree species, besides some trivial species such as Sorbus aucuparia, Prunus padus and Betula spp. With Geographic distribution respect to vascular plants, the deciduous forests in Nor­ Elm-ash forests can be found in different parts of the way become increasingly species-poor towards the west Boreo-nemoral zone. They are concentrated to four areas, and north; the successive drop out of deciduous tree as reflected by the location of the releves: bland, the species in W Norway is shown by Moen (1987). How­ plateau-like hills in Vastergotland, the fertile plains east ever, some fiord areas in W Norway are surprisingly rich of Lake Vattern in bstergotland and the surrounding of in species, of which many show an extended distribution Lake Malaren in Uppland/Sodermanland. Elm-ash forests area far to the north. are also found within the Nemoral zone in Skane (e.g. Lindquist 1938; Sjors 1967; Brunet 1991), characterized Environment by a few southerly distributed species, e.g. Thalictrum Elm-ash forests occur both on level ground and on slopes aquilegifo lium, Lunaria rediviva and Lamiastrum galeob­ of varying aspect and steepness, preferably in the lower dolon. In eutrophic elm-ash forests of this area, Fraxinus parts. The geographic distribution patternreveals that the and particularly Ulmus can successfully compete with forests are mostly bound to areas with calcareous bedrock Fagus (cf. Malmer et al. 1978; Brunet 1991). A corre­ and/or rich glacial deposits; the stands often grow on sponding forest community, the Tilio-Fraxinetum, has boulder-rich morainic tills. The soils are usually loamy or been described from Estonia by Rtihl (1960). It occurs on clayey and very nutrient-rich (cf. the high nitrogen figures

Table 16. Eutrophic elm-ash forests. Number of vascular plants (Vas), number of bryophytes (Bry) and total number of species (Tot), as well as indicator values for light (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for different sub­ units. Average figures are given. The sub-units correspond to: 1-4. Ulmus glabra-Fraxinus excelsior community ( 1. Allium ursinum sub­ community, Tilia cordata form; 2. Allium ursinum sub-community, Lonicera xylosteum form; 3. Gagea lutea sub-community, Convallaria majalis form; 4. Gagea lutea sub-community, Typical form); 5-6. Ulmus minor-Fraxinus excelsior community (5. Geranium sylvaticum sub-community; 6. Typical sub-community).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 20.4 2.5 22.9 4.6 5.0 3.9 5.3 6.8 6.3 2 21.8 4.6 26.4 4.7 4.8 4.0 5.2 6.5 6. 1 3 29.2 2.9 32. 1 4.9 5.0 3.9 5.3 6.4 5.9 4 25.0 1.3 26.3 5.0 5.2 3.7 5.5 6.7 6.5 5 32.6 1.9 34.5 4.9 5.3 3.8 5.4 6.9 6. 1 6 26.4 2.7 29.1 5.1 5.4 3.9 5.4 7.0 6.4

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 53

Table 17. Eutrophic elm-ash forests. Spearman rank correlation with releve scores on both axes and are of greater impor­ (r) between sample plot scores on the first two CA-ordination tance for the vegetational differentiation than in oligo­ axes and 17 environmental variables. The number of observa- trophic and mesotrophic forests. The correspondence be­ tions is 90. For further explanation, see Table 6. tween annual mean temperature/latitude and T-figures

Variable Axis I Axis 2 and between continentality index and C-figures is good r for axis 1, but weak for axis 2. , p r,. p

0.438 -0.286 0.01 L 0.001 Dynamics T 0.725 0.001 -0.243 0.05 c -0.359 0.001 0.463 0.001 Elm-ash forests used to occupy large areas of fertile soils M 0.23 1 0.05 -0.047 n.s. within the Boreo-nemoral zone (and also within the R 0.721 0.132 0.001 n.s. N 0.483 -0.034 Nemoral zone in S Sweden). However, as these soils were 0.001 n.s. TEM YEAR 0.594 0.001 0.360 0.001 needed by the farmers, nearly all forests were transformed TEM FEB 0.473 0.001 0.305 0.01 into arable land, to a higher degree than deciduous forests TEM JULY 0.293 0.01 0.003 n.s. PRE YEAR -0.410 -0.174 belonging to other communities or conifer forests. Par­ 0.001 n.s. MAR IND -0.536 0.001 -0.227 0.05 ticularly lowland stands and stands on level ground were CON IND -0.249 0.05 -0.344 0.01 exposed to deforestation. Only those forests growing on LAT -0.523 -0.355 0.001 0.001 boulder-rich tills or steep slopes, e.g. at the plateau-like LON 0.334 0.01 -0.46 1 0.001 INCL -0.500 -0.088 mountains in Vastergotland, escaped cultivation. The tree 0.001 n.s. ASP -0.233 0.05 -0.151 n.s. species dynamics will be treated in Chapter 5. HEAT IND -0.316 0.01 -0.068 n.s.

Ulmus glabra-Fraxinus excelsior community

Two sub-communities are distinguished: in Table 16), the pH-values usually vary between 6.0 and -Allium ursinum sub-community (releves 1-22), differen­ 7.3 (Klotzli 1975a; Ekstam 1979; Diekmann 1988). pH­ tiated by Allium ursinum and Eurhynchium angustirete, in values on Gland may today, however, come, at least releves mainly from Vastergotland and Gstergotland. The locally, down to 5.5, due to a progressive acidification sub-community is divided into two forms, a Tilia cordata which has caused an infiltration of low-pH and wide­ form (releves 1-8), characterized by the trees Tilia cordata, range bryophytes on all types of substrates (Sjogren in Acer platanoides and Quercus robur, and a Lonicera press). Klotzli (1975a) gives base saturation values of 80- xylosteumform (releves 9-22), differentiated by the shrubs 90 %. The soil type is usually a eutrophic, humus-rich Lonicera xylosteum and Ribes alpinum; brown earth which, at lower depths, may show character­ - Gagea lutea sub-community (releves 23-65), differenti­ istic gley horizons (cf. KIOtzli 1975a). The humus form is ated by Gagea lutea, Crataegus spp. and Primula veris, in a granular mull. releves mainly fromUppland, Sodermanland, Gstergotland It is evident from the correlation coefficients in Table and Vastergotland. Two forms are separated, a Convallaria 17 that the differentiation of the eutrophic elm-ash forests majalis form (releves 23-59), characterized by Convallaria is connected with both climatic and edaphic factors. The majalis and Campanula trachelium, and a Typical form releve scores on axis 1 ( eigenvalue 0.273) are positively (releves 60-65), without differential species, but with correlated with the temperature variables, with longitude, comparatively high frequencies of some nitrogen-demand­ as well as with L-, M-, R- and N-figures. They are nega­ ing species characteristic for eutrophic forest fringes, tively correlated with some climatic variables (yearly such as Galium aparine, Urtica dioica and Geranium precipitation, de Martonne's index, continentality index) robertianum. and the physiographic variables. Here, the climatic vari­ The differentiation into forms is only supported by ables connected with moisture do not correspond to the comparatively few species. This is also obvious from Fig. M-figures. Axis 1 expresses the differentiation into the 15, where the sample plots of the different forms are not mainland community and Gland community. The latter clearly separated from each other except for the Typical are thus characterized by a warmer, drier and less conti­ form of the Gagea lutea sub-community. nental climate and by more fertile soils with a higher pH. Inclination and aspect, as well as their combined effect, Allium ursinum sub-community are of underlying importance regarding the mainland stands The forests, particularly those of the Tilia cordata form, which are most often found on slopes. Axis 2 has a fairly are fairly species-poor, mainly because of a low species low eigenvalue (0. 172), and the releve scores on this axis number in the field layer. This may be caused by the show generally weaker correlations with environmental closed canopy (as indicated by the low L-figures, see factors. Here, high coefficients are mostly found for cli­ Table 16) and by the frequent dominance of Allium matic variables. L-figures have significant correlations ursinum, which often covers more than 50 % of the

Acta Phytogeogr. Suec. 80 54 M. Diekmann

ground. Allium is confined to nutrient- and base-rich, tion, but are concentrated to three regions: the surround­ moist but not too wet soils with a high pH (Oberdorfer ings of Lake Malaren (Uppland, Sodermanland and Vast­ 1983; Ellenberg et al. 199 1), and often indicates moving manland), the easternside of Lake Vatternin bstergotland ground water (Sjors 1967; Oberdorfer 1983). Due to its (particularly Omberg) and the plateau-like hills in Vaster­ high fertility and its ability for vegetative regeneration, it gotland. A few stands are located in Smaland and SE usually grows in dense patches, hindering the growth of Norway. Mainland elm-ash forests have not been studied other species. Ernst ( 1979) showed that allelopathic sub­ in detail. However, there are some descriptions of single stances do not have an effect on other herbs in the field stands (e.g. Sterner 1926; Ryberg 1971) or studies from layer, but decaying bulbs and roots can hamper the seed restricted areas, e.g. by Almquist (1929) and Wallin (1973). germination of other species (Ellenberg 1986). However, The forests occupy level ground or, more often, slopes of Mercurialis perennis is often a codominant together with up to 40 % inclination. The aspect varies, but a northern Allium, covering more than 50 % of the ground. The exposure is only exceptionally found. The indicator fig­ bottom layer, on the other hand, shows higher cover ures suggest that the stands of the Typical form have values and a higher species richness, on average, than in moister and more base- and nutrient-rich soils than those the stands of the Gagea lutea sub-community. Eurhyn­ of the Convallaria majalis form (Table 16). chium angustirete, a differential species of the Allium ursinum sub-community, is common. Ulmus minor-Fraxinus excelsior community Allium ursinum is a sub-oceanic species and particu­ larly common on the Norwegian west coast (Sjors 1967). The forests of the Ulmus minor-Fraxinus excelsior com­ Correspondingly, stands of the Allium ursinum sub-com­ munity are differentiated by two groups of species: one munity are mainly distributed in the western parts of the group comprising species of a southern distribution (e.g. Boreo-nemoral zone. Most forests are located at the pla­ Anemone ranunculoides, Viola reichenbachiana and teau-like hill of Kinnekulle in Vastergotland (Tilia cordata Corydalis bulbosa), the second group comprising indica­ form) and east of Lake Vattern in southern bstergotland tor species for high soil moisture (Filipendula ulmaria (Lonicera xylosteum form). Single stands are located in and Geum rivale). Orchids such as Orchis mascula, Listera Halland, Bohuslan, Narke and Smaland, as well as in SE ovata and Dactylorhiza spp. occur frequently in the fairly Norway (cf. Ulmo-Tilietum stands with Allium in Kielland­ species-rich field-layer. Lund 1981). The 'Allium ursinum-Eschenmischwald Stands belonging to the Geranium sylvaticum sub­ ' mentioned by Stszjrmer (1938) from the island Haszjya community (releves 66-76) show a fairly dense shrub located in the Oslofiord, however, can be classified as layer, composed of several shrub species and young trees. alder-ash forest dealt with in Chapter 4.5. They have a number of differential species which also The stands are generally located on slopes, preferably occur in the Ulmus glabra-Fraxinus excelsior community of (north-) western aspect. This preference for slopes on the mainland, e.g. Geranium sylvaticum, Viola riviniana exposed to the west may have an ecological background, and Aegopodium podagraria. The stands of the Typical but may also simply be caused by the fact that slopes sub-community (rei eves 77-90) are species-poorer than suitable for the Ulmus glabra-Fraxinus excelsiorcommu­ those of the Geranium sylvaticum sub-community, but nity mainly have a western exposure, as for example on characterized by Ulmus minor constantly occurring in all the eastern side of Lake Vattern. The indicator figures do strata. Besides, the two mosses Eurhynchium striatum not clearly differ from those of the Gagea sub-commu­ and Climacium dendroides are found in some stands. nity, except for lower L-figures (Table 16). The indicator On bland, the stands of the Geranium sylvaticum sub­ figures for M, R and N of the Tilia cordata form are only community are restricted to some small areas in southern slightly higher than those of the Lonicera xylosteum form, and middle bland, especially at the limestone escarpment suggesting that edaphic differences between these two south of Borgholm (cf. Ekstam & Sjogren 1973; Ekstam forms are not very marked. 1979; Diekmann 1988). Stands of the Typical sub-commu­ nity are confined to two forest areas, Vasterstads lund (Fig. Gagea lutea sub-community 16) and Stora Dalby lund (Fig. 17), both located in the Compared with the previous form, the forests show a southernpart of the island. These stands were earlier men­ higher species richness, but lower cover degrees in the tioned as two of Sweden's most luxuriant (elm-ash) forests field layer. This may be due to the fact that Allium ursinum by Sterner(19 26, 1955) and Selander (1955). Vasterstads and Mercurialis perennis are absent or rare, respectively, lund is particularly species-poor, but has a somewhat favouring the occurrence of a number of less competitive higher abundance of moisture-demanding species. species. Among the differential species, some indicate a The occurrence of moisture-demanding species is also fairly high light availability in the field layer, such as reflected in the soil types. An investigated soil profile Primula veris and Ranunculus acris. from a stand of the Geranium sylvaticum sub-community Forests of this sub-community show a wide distribu- corresponded to a eutrophic brown earth. Stora Dalby

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 55

Fig. 16. Eutrophic elm-ash forest, Ulmus minor-Fraxi­ nus excelsior community. Both upper and lower tree layer are dominated by Ulmus minor. Abundant spe­ cies in the field layer are Ranunculus ficaria, R. auri­ comus and Mercurialis perennis. - Vasterstads lund, bland, May 1993. Photo Folke Hellstrom.

lund had a brown earth gley, whereas a soil from glutinosa, sometimes also A. incana. (Although U. glabra Vasterstads lund was classified as gley. All three soils is a constantly occurring species, the forests will be called were fairly humus-rich and very much influencedby the alder-ash forests, since Fraxinus occurs more frequently, activity of earthworms. The high ground water table of the in particular also in corresponding forests in central Eu­ soils in these forests was also mentioned by Klotzli (1975a). rope. In addition, the term 'alder-ash forests' will also be However, the moisture figures are lower than for the used for Alnus glutinosa and A. incana forests without Typical form of the Allium ursinum sub-community and Fraxinus, as long as the general species composition only slightly higher than for the other mainland forms allows them to be assigned to this forest type.) Quercus (Table 16). The R-figures are the highest among all distin­ robur occurs only rarely, while Tilia cordata is com­ guished sub-units, suggesting very high pH-values. pletely absent. The stands are 20-25 m, exceptionally up Although the elm-ash forests on bland are among the to 30 m, tall and, thus, comparatively somewhat lower most mature and 'primeval' -looking deciduous forests in than the eutrophic elm-ash forests. Sorbus aucuparia and Sweden, they have been influenced,e.g. by grazing (Stemer Prunus padus are common in the lower canopy. The shrub 1926, 1955), and are still undergoing secondary succes­ layer is also usually less diverse (and more open) than in sion (cf. Chapter 5). mesotrophic or eutrophic elm-ash forests, with cover per­ centages varying between 5 and 50 %. Corylus avellana is the only common shrub species, but never exceeds 4.5 Eutrophic alder-ash forests 50 % cover. Besides, young specimens of Fraxinus, Ulmus and Prunus are abundant. The field layer usually covers about 70-80 % of the General characteristics ground and gives a luxuriant impression, due to many tall­ Species composition and structure growing, nitrophilous species. Particularly fernsare very The tree layer of the eutrophic alder-ash forests is mainly frequent and occur with several species, e.g. Athyrium composed of Fraxinus excelsior, Ulmus glabra and Alnus filix-femina, Matteuccia struthiopteris, Dryopteris spp.

Acta Phytogeogr. Suec. 80 56 M. Diekmann

Table 18. Releve table of eutrophic alder-ash forests. (See Table 3.)

Running number 123456789111 111111122 2222222 233333333334444 4444 012 345678901 2345678 901234567890123 4567 Cluster 17a 17b 17c 18a 18b Cluster code 11000 11001 1101 1110 1111 Region SUNSSUBBSOVU BBBBBBBHHHSVVSS SNNN NNVVVVVNN NNNNNNN MPAOOPOOOGAP OOAAAMOO 0000000 OOOOOOOAAAOMM 0000 Aspect W-NSW-SS-WEW SENNENNNN SN-NNWN WN-NENNSSS- -S-- -NNS E W N W WE N W WE EW NEE E W E Inclinat ion 8-385-13-184 433232233 82-1333 53-5311188--5-- -433 {0 ) 50 5 0055550 5 5005 0050 Cover % T1 (X10) 766555766675 667667856 3676566 567677677775665 7376 % T2 {X10 ) 336656454444 333533453 5234643 313312312112123 1614 % s {X10 ) 422134241433 542143134 2111325 221121222242132 4222 % F (X10) 797757447976 569988776 7787786 897777588878789 7677 % B 164211723112 425364455 -256125 634535322521543 5412 00000000000 000000000 000000 000000000 00 00 000 N - Tot 213224223231 234454333 1112223 233333433432242 4324 584868686608 969010799 5987512 851604357219185 2048 N - Vas 212224123221 224333233 1112212 222222322322142 3223 139203911146 581163812 5430066 051820179775923 8119 N - Bry 455665775562 488917987 -557556 811881188544262 4939 5 00 42 n Acer platanoides ...... 3. 3 ... .4 ..3. 4

•.... Fraxinus excelsior .6355545.453 645554456 4544565 5 ..444344 ..4 .. . .443 Ulmus glabra 646444644534 ..54 . 6634 . 465 ..4 ...44 .33 .....64 .4 .

Alnus glutinosa ...... 43 5445 . 5 . .. 4665566666656 .4 6.4. . � .. Alnus incana ...... 3 .. 3 ...... 3. . . 56

. N Acer platanoides .4.55444 .44. 3.45 .4.43 443 .433 . . 3 . Quercus robur . 3 4 ..4. . . 4 Alnus incana .....••...... 3 .. 3 ...... •...4 .544

Ulmus glabra 446 . 45555445 .4434454 . . . 45554 4 ..4. 353 .4... 43 ..3 . Fraxinus excelsior . 4.333 .4.32. 44 .....44 544.4.4 . 45444 . 343343 .. 3443 Sorbus aucuparia .....3 ...33 . 333.4.444 ...3 ...... 3 ..42 .. . . 3. Alnus glutinosa . . . 4 ..3.4 3 .. • 3.3 ..4 ..4.3 333 . 33.. Prunus padus 44 .443 ..3 ... 4 3 3 ..... 4 33 .3

.•..•.• . •......

.s. Lonicera xylosteum 4.2.3 ....22 . ..23332 .. Corylus avellana 4 ..24444344 . 344.4 ..33 ...34 45 . 3 ..3 ...... 3. . . 3

Acer platanoides . 4322 .24.3.2 ...... 34 ....2. 2 2 3 . ..2• Alnus glutinosa .••...... •••••••2 ..3 .. 2 ••• .2 Ribes rubrum 3 3 .. 3 ... Alnus incana 4 ...•••.. .•. .323

..•..... Fraxinus excelsior 432 . 33432432 4443 . 3244 4432434 344344244423 ... 2444 Ulmus glabra 444 . 4434 .4.4 . 4 . 252443 2 ..3433 ...22234 . 342 .4. ..2. Prunus padus 44 .444 . 4 .. 4. 5232 ...35 4 4 .....434254244 5424 Sorbus aucuparia .....3 ....2. 232 . 32243 •...... 322 .3 . ..3. Viburnum opulus . . 1 ...... 32 .3 .3 ...... 2 ...... 3 ... Ribes alpinum 3.2 ...... 2 3 .

• • • .•..• . .••...... •..

.E. Hepat ica nobi lis 4.4...... 1 .. ..44443 .1 ...... 1. Lonicera xylosteum 1 ...1 ...... 122211 . 2

Viola riviniana ..1. . 4 ..2 .1. ....21 ...... 1 .. . ••• Maianthemum bifolium ..1.. 2 .....•••2 ...... 334 Ribes alpinum 3.2.2 ....1.. ...2 Polygonatum verticillatum 4.2423••••• ... Pulmonaria officinalis ..42 1.2 .. Lathyrus vernus . . 22 .11. . 2

Epilobium montanum ..11 .1.2. •... .• Dryopteris filix-mas 2 ..131 .21.21 .. 3445343 ....343 ...22 ...... 2. Actaea spicata 2 ..132 ..11 .1 .12233144 ....1 ...... 2. 1 ... Veronica chamaedrys ..1 ...... 1.112 ..21 . ...1.1 Corylus avellana 1. ...2.2. 21. 2 ....1 ......

Poa nemoralis 3 ..2 ...... ••..•••.1.1. . ...1.1 . Galium odoratum ••••• ..25 . 4 ...... 444 Dryopteris dilatata .2.2222 Stellaria nemorum ...... 5 ...... 432 .555564 .45544 . 44445445 .344 Urtica dioica 2 ...... 1 1222 ... . 4. 324111234335 332 . Chrysosplenium alterni folium ..• . 224 ..2 .334 ..422244434 3433 Matteuccia struthiopteris . . . . 4.4 . 3 4554 ..2 ...5 ..3.3 34 ..54 25 .4 Ranunculus repens ...... • .44443 . 44422243 4221 Caltha palustris . 32 ...34444 ..34 4.22 Cardamine amara . 313 ..4443 .2 ..5 .42. Poa trivialis 1 ...... 32 . 32435 .2.2. Impatiens no li-tangere 2 ...... 4 ..34 4.45 331 . Solanum dulcamara ....•... .4 .....313 ...2. 2

Gal ium palustre .1.... 1122 ...... • Carex remota ...... 423 ...... 2. Equisetum sylvat icum 2 .3 .. . . . 4.32 Equisetum arvense 2 ••• ••• ...... 1 ...... 4.22

.•••• ••••.. .

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 57

Table 18, cont. Filipendula ulmaria ..1. .122412 . 554313221 1223 .24 554454456455544 4455 Fraxinus excelsior 334 . 2443 .444 34333 .344 2 ..2324 .2322223241. ... 3232 Athyrium filix- femina 2 ..1.3 24 ..12 .332.4343 .3.4424 .55443444444443 5444 Anemone nemorosa 545664554365 ..5455 511 ...1 ... 544544644453453 5 ... Oxalis acetosella 1. ..35224 .2. 124324454 .2.4424 .32233222 .21.2. 313 . Paris quadrifolia . 344443 ...44 434433233 .1322 .2 ...2 ..23 322 .. 42 3.22 Prunus padus 222134 .3 ..3. 412 .42 ..3 2 ...... 3 .....423243 .32 4323 Geum rivale . . . . . 12421. . 44433 ...2 ...... 1 44444644443 .2 .. 4343 Stachys sylvatica 3243 .43 .3 ..2 3 ..4443 .. 3.24.33 ...11 2 ...... 12 .232 Ranunculus ficaria 4 ...55776 ..2 1.4 ..3 ... 766767566656455 3 ... Ulmus glabra 4.31 24 . 21212 .33.3.222 2 ...21...... 132 .22 ...... 2. Geum urbanum 223423 .24 ..2 ..3343233 ...1 ..2 2 ....41. .43 ..2. 3 ... Campanula latifolia . . . 22 .444 ... .32244324 4344444 2 .. 2 ...... 11 Acer platanoides . 231332 ..121 133 ..31 43 ..1. 122 ...... 1 ...... 2. . .2 . Equisetum pratense . . . 34 ..3 .... .243 . 4444 ....334 ....4. 3 ...3244 . 3442 Ranunculus auricomus agg . . . 2 ..4444 .1. ..2.2 .... 4444444 ...4.1 42 Deschampsia cespitosa .. 11 .1. .2 ... . 22 .2 .... 244 . 13233 .1. .2. 1.1. Geranium sylvaticum . . 2 ..3233 ... . 244311.. 4 .. 1 ...... 1 .... .11 . Anthriscus sylvestris 2.23 ...23 ... ..2 .....2 3.1.231. ..1. .2. 1 ... Sorbus aucuparia 2.1. .3 ...21. .2 ..21.2 2 ....1 ...... 12 ... 1. 1 ... Valeriana sambucifolia 4344 ...... 342 .. 3.33 .41. . . 122 Viburnum opulus ..······1. . 2 ······...142 33 .3.3 2 ...... 3.2. Crepis paludosa ...... 31 . . 321. .3 .. ...1 ..2 ...... 3. . .42 Gagea lutea ...443444 ...... 4 ..4 ...... 23 Taraxacum officinale agg . . . . 1.2 ..1 ... ..2. 2 ..11 ...... 1.2 ...1 . Angelica sylvestris ...... 32 ...... 1. 241 . .3. 3.21 Aegopodium podagraria . 7.65 ...5 ...... 5 .. 6 ...... 5 ..... 1 .. 3 ..... Polygonatum multiflorum ..2 ...... 4 ....2.1 .. ..1 ...3 ...... 11 ...... Rubus idaeus ...... 3 ... . 3 ...2 ..4 4 ...3 .....2 ..1. Glechoma hederacea ..1 ..2 ...... 3 ...... 233 .. 3 ...... Dryopteris carthusiana ...1.1 .... 2. 2 ...... 1...... 1 ... .2.. Elymus caninus ...2 ...... 321. .2...... My celis mu ralis .....3 ...... 33 ...... 1 ...... 1 ... Milium effusum . . . . . 2 ...... 2 ...... 32 ...... 1. Lysimachia vulgaris ..... 1 ..1 ...... 1 ...2 2 ... Thelypteris phegopteris ...... 21 ...... 3 ...... 2.3 . Equisetum hyemale ...... 4 ...... 54 ...... 2 . Carex sylvatica ...... 2 .. . 2 ...... 3 ...... 2 . Fragaria ve sca ...... 1 ....22 ...... 1 . Silene dioica ...... 2 ..2 . .1 ...... 2 ......

.6. Brachythecium rutabulum 453433333 .21 23133331. .22.32. Eurhynchium angustirete .4.23.5.. 33 . 43 .4.4445 .432234 ...... 3 . . 2 .. Thuidium tamariscinum ...... 4 ..22 ...... 4143333241 ..... Brachythecium rivulare ...... 2 ... 344524444344445 4444 Rhizomnium punctatum ...... 3 ..3.2 ...... 223 . Mni horn ...... 3 ...... 23 ...322 ...... urn urn

Eurhynchium praelongum agg . 445444444443 .44454444 .355444 55545554432 ..2. .24. Plagiomnium undu latum 4.1324444 .3. 543 . 24441 .134 .24 444444444442 .54 343 . Cirriphyllum piliferum 34443 . 4442 .. 444444344 .445344 433 .44.2 ..4 ..3. .4.1 Plagiomnium affine . 2.2 ...... 112 . 2222 ...3 ... 2332 .331 . .. 21. . .433 Plagiochila porelloides ...... 2 .... .23333243 ....2.3 .233233 ...... 21. Atrichum undulatum ....1 ...... 224 .3 .. .. .1.. 1 2.2.333 ...... Rhytidiadelphus triquetrus ...... 21 .3 .. . .541 .444 ....1 ...... 2. Cal liergonella cuspidata ...... 1 ...... 1 ..2 ..2 .. .

Additional species (occurring in one , two or three releves ):

3X: (T1) Populus tremula 15:3, 17 :4, 44 :3 (F) Allium ursinum 10 :7, 11 :2, 22:4, Alnus incana 13 :3, 45:3, 46:4, Festuca gigantea 1:1, 13 :2, 38:3, Gymnocarpium dryopteris 20:4, 21:3, 28:2, Humulus lupulus 13 :2, 14 :1, 22 :2, Lathraea squamaria 8:2, 18 :3, 19 :2, Pha1aris arundinacea 30:2, 36:4, 37 :4, Primula veris 6:1, 15 :1, 17 :1, Ribes rubrum 32 :1, 37 :2, 44 :3, Rubus caesius 6:3, 18 :1, 40:2, Rubus saxat ilis 11 :1, 17 :3, 46:2, Vicia sepium 3:2, 9:3, 20:2 (B) Climacium dendroides 34:2, 37 :2, 46 :2, Conocephalum conicum 35:1, 45 :3, 46 :1, Pellia epiphylla 32 :1, 34:1, 44 :1.

2x: (T2) Malus sy lvestris 33 :3, 37 :3, Picea abies 14 :3, 21:3 (S) Picea abies 11 :2, 25 :2 {F) Alnus glutinosa 40 :2, 41 :2, Carex digitata 10 :2, 28:1, Convallaria maj alis 6:2, 17 :2, Dactylis glomerata 6:1, 42 :1, Lapsana communis 6:1, 9:1, Listera ovata 3:1, 22 :1, Me lica nutans 6:1, 10 :1, Platanthera chlorantha 6:2, 13 :1, Poa remota 14 :2, 25 :2, Populus tremula 14 :2, 17 :3, Rumex obtusifolius 42 :3, 43 :1, Scirpus sylvat icus 38:4, 43 :3, Tussilago farfara 42 :1, 44 :2, Veronica beccabunga 40:3, 42 :1, Viola mirabi 1is 3:1, 17 :3 (B) Eurhynchium striatum 11 :3, 36 :2, Fissidens cristatus 6:2, 17 :2, Fissidens taxi folius 8:2, 35:2, Lophocolea heterophylla 17 :1, 38 :1, Plagiomnium cuspidatum 6:2, 14 :1, Thuidium erectum 17 :1, 34 :3.

1X : (T1) Acer pseudoplatanus 17 :2, Betula pendula 2:3, Picea abies 11 :3, Sorbus aucuparia 42 :3 (T2) Betula pubescens 46:4, Populus tremula 42 :3, Prunus avium 22 :4, Quercus robur 29:3 (S) Betula pubescens 42 :2, Ma lus sylvestris 11:2, Populus tremula 42 :3, Prunus avium 12 :4, Ribes uva-crispa 44 :2, Ti lia cordata 6:3 (F) Aconitum septentrionale 21:4, Anemone ranunculoides 2:3, Aquilegia vulgaris 3:1, Arctium tomentosum 9:2, Brachypodium sylvaticum 5:2, Campanula trachelium 6:1, Cardamine impatiens 2:1, Carex montana 17 : 1, Carex pallescens 17 : 1, Cirsium palustre 38 :3, Crataegus sp . 17 : 1, Dactylorhiza sp . 14 :1, Equisetum fluviatile 43 :3, Gal ium aparine 38:4, Geranium roberti anum 38 :3, Glyceria fluitans 44 :2, Juncus effusus 38 :2, Lilium martagon 5:1, Luzula luzuloides 17 :1, Luzula pi1osa 15:1, Lycopus europaeus 38 :2, Lysimachia nummularia 6:4, Moehr ingia trinervia 1:1, Mo linia caerulea 4:2, Myosotis scorpioides 44 :3, Neottia nidus-avi s 19:1, Prunus avium 12 :2, Pteridium aqui linum 22 :2, Quercus petraea 31:1, Quercus robur 6:1, Ranunculus cassubicus 2:2, Ribes uva-crispa 3:1, Scrophularia nodosa 42 :1, Tilia cordata 6:2 (B) Brachythecium velutinum 17 :1, Bryum pseudotriquetrum 1:3, Campy lium chrysophyllum 17 :1, Cephalozia leucantha 12 :1, Dicranum maj us 5:2, Hypnum cupressi forme 17 : 2, Leiocolea sp . 1: 2, Lophoco1ea bidentata 3 4: 1, Rhytidiadelphus squarrosus 17 :1, Scleropodium purum 3:1.

Acta Phytogeogr. Suec. 80 58 M. Diekmann

Fig. 17. Eutrophic elm-ash for­ est, Ulmus minor-Fraxinus excelsiorcommunity, fieldlayer in spring aspect. Visible are Anemone spp., Hepatica nobilis, Paris quadrifolia, Geum spp., Convallaria majalis, Acer plata­ noides (saplings), Allium olera­ ceum, Ranunculusfie aria, Poly­ gonatum multiflorum, etc. - Stora Dalby lund, bland, May 1993. Photo Folke Hellstrom.

and Thelypteris phegopteris. Also Equisetum spp. are europaeus, Lysimachia spp., Calliergon cordifolium, locally abundant. Several species in the field layer differ­ Calliergonella cuspidata, etc. On the other hand, many entiate from all other deciduous forest communities. These deciduous hardwood forest species are lacking. are throughout indicators of wet soil conditions, e.g. Equisetum pratense and Crepis paludosa. As in elm-ash Differentiation forests, Campanula latifolia and Stachys sylvatica reach Two communities are distinguished: high frequencies. On the other hand, several species, 1. Fraxinus excelsior-Prunus padus community (releves which are constant or common in forests on compara­ 1-28), differentiated by Acerplatanoides, Corylusavellana tively drier soils, are much less frequent in alder-ash and some field layer species, and with Prunus padus forests, such as Melica nutans, Vicia sepium and Con­ constantly occurring in the lower tree or shrub layer. vallaria majalis. 2. Fraxinus excelsior-Alnus glutinosa community (releves The bottom layer reaches higher cover values than in 29-47), characterized by Alnus glutinosa and several field all other deciduous forest communities (up to 70 %) and is layer species, e.g. Chrysosplenium altern ifolium, fairly species-rich. Most important are the genera Matteuccia struthiopteris, Ranunculus repens and Caltha Brachythecium, Eurhynchium and Plagiomnium. palustris. Alder-ash forests have a species composition which The differentiation of these communities is clearly widely differs from that of alder swamp forests, the latter supported by the ordination diagram in Fig. 18 (sub-unit being found on very wet, periodically inundated soils in numbers 1-3 and 4-5, respectively). depressions and along lake and sea shores. These are characterized by a large number of species in shrub, field Affinities and bottom layers, such as Salix cinerea, S. pentandra, There are only a few studies of alder-ash forests from the Carex elongata, Calamagrostis canescens, Lycopus Boreo-nemoral zone, but as these forests have a very wide

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 59

5

4

3

2

N � 1 X 3

2 0 1 .

�! 1 1 -

-2

-3

-2 -1 0 1 2

AXIS 1

Fig. 18. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 47 releves of eutrophic alder-ash forests. Numbers of different sub-units plotted onto the sample plot positions which are slightly adjusted in order to avoid overlapping figures. 1-3. Fraxinus excelsior-Prunus padus community ( 1. Hepatica nobilis sub-community, Typical form; 2. Hepatica nobilis sub-community, Polygonatum verticillatum form; 3. Dryopteris dilatata sub-community); 4-5. Fraxinus excelsior-Alnus glutinosa community (4. Poa trivialis sub­ community; 5. Equisetum sylvaticum sub-community).

distribution, several investigations exist from adj acent Alnus incana in western Norway were carried out by areas, particularly in the Boreal zone. Here, Fraxinus Odland (1981, 1991, 1992) and Odland et al. (1990). The excelsior is usually absent, and the forests are mainly richness in ferns is also emphasized in studies of Alnus composed of grey alder (Alnus incana). The studies often forests from the Boreal zone in Finland by Tapio (1953), deal with only one of the communities mentioned above. Koponen (1967) and Makirinta ( 1968, 1982). Havas ( 1967) A monograph ofEuropeanAlnus incana forests including treated the ecology of Alnus incana forests around the Fennoscandia was presented by Schwabe (1985). As syno­ Gulf of Bothnia in Finland. Only a few studies exist from nyms for (and as related communities to) alder-ash forests the Boreal zone in Sweden, e.g. by Runemark (1950) and or one of their sub-units from the Boreo-nemoral zone can Sjors (1960). Finally, alder-ash forests were also de­ be mentioned: scribed from the N emoral zone in Skane by Brunet ( 1991). 'Ask-al-lund' (Selander 1955), 'graalskog' (Selander 1955; Sj ors 1967; Anon. 1984), 'askskog' (Sjors 1967; Geographic distribution Bergendorff et al. 1979; Anon. 1984), Alno (incanae) - Within the Boreo-nemoral zone, alder-ash forests are Prunetum, Alno (incanae)-Fraxinetum and Equiseto­ mainly found in the westernparts. The majority of stands Fraxinetum (Kielland-Lund 1971, 1981), 'Stellaria are located in the provinces of Bohuslan, Vastergotland nemorum-samhalle and Geranium robertianum-Urtica and Halland, as well as in SE Norway. Further to the east, ' dioica-samhalle (Wallin 1973), Alno-Ulmetum (Klotzli alder-ash forests become less common. On bland and ' 1975a), Alno incanae-Piceetum (Bj0rndalen 1980a). Gotland, no forests of this type were found; accordingly, Many studies of Alnus incana-rich forests in various several of the characteristic species of alder-ash forests communities were carried out in W and N Norway, e.g. by are rare or lacking on these islands, as well as on the Fremstad & 0vstedal (1978), Fremstad (1979), Fottland eastern mainland,such as Carex remota, Stellaria nemo rum (1980), Blom (1980), Berthelsen (1980), Klokk (1980, and Matteuccia struthiopteris. Alder-ash forests also oc­ 1982), Fremstad &Normann (1982) and 0vstedal (1985). cur in the Boreo-nemoral zone of Russia and the Baltic Alnus glutinosa forests of different successional stages states (Linkola 1916, 1929; Perttula 1950; Rtihl 1960; cf. were described by Fremstad ( 1983) from W Norway. Schwabe 1985). Alder-ash forests and related communi­ Synecological investigations of fern-richvegetation with ties are most commonand widespread in the Boreal zone

Acta Phytogeogr. Suec. 80 60 M. Diekmann

and the Western broad-leaved and pine forest region (see eating a comparatively higher frequency of more conti­ above). nental and borea1 species. However, in view of the frequent Thus, forests on wet soils composed of Fraxinus occurrence of alder-ash forests in western Scandinavia excelsior and/or Alnus spp. occur almost in the whole of (and Britain), it must be doubted in this case that T- and C­ Fennoscandia, although with a widely varying species figures give reliable indications. R- and N-figures are, on composition. Related communities are fo und in central average, somewhat lower than those for the elm-ash for­ and eastern (southeastern) Europe, Alnus glutinosa­ ests, whereas all M-figures are much higher, pointing at Fraxinus excelsior forests in the lowlands and Alnus the importance of moisture for the differentiation of these incana forests in montane and alpine areas (e.g. Schwabe two eutrophic forest types. 1985; Ellenberg 1986). Alder-ash forests, particularly the The importance of moisture is also evident from the grey alder forests, thus represent a group of communities correlation coefficients between the releve scores on the of an azonal distribution. CA axes and the environmental variables (Table 20). Axis 1 (eigenvalue 0.355) mainly expresses a moisture gradi­ Environment ent: the releve scores on this axis are significantly posi­ Alder-ash forests can be found on a variety of sites, tively correlated with M-figures (rs 0.933), yearly pre­ = provided that the soil is sufficiently wet. They grow on cipitation and de Martonne's index. They also show a level (flat orundulating) or gently to moderately sloping strong positive correlation with L-figures, suggesting that ground, at the foot of slopes and on alluvial plains along stands on the wettest soils have a more open canopy, brooks and river banks. The wettest forest sites are peri­ probably due to a hampered growth of trees. Besides, odically inundated during winter and spring. Usually, the annual mean temperature and mean temperature in Febru­ stands occupy only small areas which are often formed as ary are positively correlated with the releve scores on axis narrow but long strips, neighbouring elm-ash forests on 1, since the stands occurring on the wettest soils are one side and river bank vegetation without trees on the located in the southern, warm provinces of the study area. other side. The releve scores on axis 2 (eigenvalue 0.281) show Typical for alder-ash forests are moist to wet mineral particularly high correlations with L-, M- and R-figures soils; there is no great tendency for the accumulation of and annual mean temperature. Light is thus of greater peat (as in alder swamp forests). When occurring in the importance for the differentiation of alder-ash forests than same area, Alnus incana often prefers coarse-grained for the other forest types. Soil moisture and soil reaction sediments, while A. glutinosa is mostly found on clayey (fertility) show opposite trends. As previously observed subsoils. The soils are nutrient- and base-rich, due to the for the other forest types, T- and C-figures do not corre­ following environmental conditions: 1. Many stands grow spond well with the climatic variables. on morainic tills and glaciofluvial sediments primarily rich in nutrients. 2. Stands growing on flushed slopes and Dynamics along brooks are influencedby seepage and moving water As all other deciduous forest types of the area, the alder­ constantly supplying allochthonous nutrients and bases. ash forests have been influenced by cutting, logging and Also oxygen is transported, supporting the plant roots and grazing. Where Alnus spp. and Fraxinus excelsior occur facilitating the decomposition of organic material which together, heavy disturbance generally favoursAlnus at the is hampered due to the wetness of the soil. 3. Both species expense of Fraxinus, because (1) Fraxinus is very much of Alnus have nutrient-rich litter favouring the occurrence liked by grazing animals and sensitive to grazing, whereas of nitrophilous herbs and grasses. The trees live in sym­ Alnus incana is avoided by animals (Sjors 1967; Schwabe biosis with microorganisms of the genus Frankia 1985), and (2) both alder species are able to reproduce (Actinomycetes), forming woody root nodules in which vegetatively after cutting, Alnus incana by root suckers the atmospheric nitrogen is fixed (Ellenberg 1986). and both Alnus incana and A. glutinosa by basal shoots The stands are found on brown earths, gleys and 'Aue' (Sjors 1967; Oberdorfer 1983). Fraxinus, on the other soils or related soil types (cf. Klotzli 1975a, Kielland­ hand, can stand repeated harvest of its twigs and leaves, Lund 1981). The pH is usually between 5.0 and 6.5 but suffers from cutting of the main shoots. (Wallin 1973; Kielland-Lund 1981), the base saturation After grazing has ceased, Alnus spp. usually are the is, with more than 80 %, the highest in all deciduous forest first to colonize the ground, while Fraxinus becomes communities studied by Klbtzli (1975a). However, ac­ more dominant in later stages of succession (cf. Kielland­ cording to Kielland-Lund (1981), base saturation values Lund 1981). In young woodlands in the outer archipelago may vary between 44 and 96 % for different communities of Stockholm in the Baltic Sea, Alnus glutinosa becomes or their sub-units. successively replaced by Fraxinus (Tapper 1992). Thus, The Indicator values (Table 19) differ considerably in alder-ash forests, Fraxinus behaves like a 'climax' from those for the elm-ash forests. Throughout, the T­ species, in contrast to its 'pioneer' behaviour in elm-ash figures are lower, whereas the C-figures are higher, indi- forests (see Chapter 5). In alder-ash forests in SE Norway,

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoralScandinavia 61

Table 19. Eutrophic alder-ash forests. Number of vascular plants (Vas), number of bryophytes (Bry) and total number of species (Tot), as well as indicator figures for light (L), temperature (T), continentality (C), moisture (M), reaction (R) and nitrogen (N) for different sub­ units. Average values are given. The sub-units correspond to: 1-3. Fraxinus excelsior-Prunus padus community ( 1. Hepatica nobilis sub­ community, Typical form; 2. Hepatica nobilis sub-community, Polygonatum verticillatum form; 3. Dryopteris dilatata sub-commu­ nity); 4-5. Fraxinus excelsior-Alnus glutinosa community (4. Poa trivialis sub-community; 5. Equisetum sylvaticum sub-community).

Sub- Number of species Indicator figures unit Vas Bry Tot L T c M R N

1 23.4 5.5 28.9 4.7 4.8 4.0 5.8 6.3 6.2 2 31.1 8.2 39.3 4.8 4.6 4.2 5.7 6.1 5.8 3 17.7 4.7 22.4 4.4 4.7 4.2 6.0 6.4 6.2 4 26.2 6.7 32.9 5.2 4.6 4.0 6.8 6.1 6.0 5 29.3 5.8 35.0 5.0 4.5 4.3 6.9 6.2 5.6

Prunus padus behaves similarly to Fraxinus: it is sensi­ Alnus glutinosa behaves as a pioneer on abandoned heath­ tive to grazing and a slow colonizer, but becomes more and farmland and forms forests on wet hillsides and along frequent in later successional stages, partly due to its brooks (Fremstad 1983). In W Norway, Acer pseudo­ ability to form vegetative polycorms (Kielland-Lund 1981). platanus, a recently naturalized escaper from gardens, is Alnus incana also functions as a pioneer species and rapidly spreading and forming mixed stands with Fraxinus, becomes very abundant on abandoned pastures and cleared Alnus spp. and Prunus padus (cf. Kielland-Lund 1981; forest sites (Aune 1973; Kielland-Lund 1981). At the Fremstad 1983). Picea abies may substitute Alnus spp. on Bothnian coast in Sweden and Finland, Alnus incana not too wet sites (cf. Sjors 1967), but can hardly establish (further to the south to an increasing degree also Alnus on very wet soils and steep, slippery slopes due to its glutinosa, Skye 1965; Ericson & Wallentinus 1979) can shallow root system (cf. Klotzli 1975a, Klokk 1980). be found close to the seashore, forming a low but dense forest fringe (Julin 1965; Havas 1967). Due to the consid­ Fraxinus excelsior-Prunus padus community erable land upheaval in this area, this grey alder belt has a short life time on the same spot: in the lower part, Alnus The forests of this community on moist soils connect the invades the young terrestrial soil (here, Picea is probably elm-ash forests on mesic to moderately moist soils and the excluded due to the (low) salt contents of the soil, Skye alder-ash forests of the Fraxinus excelsior-Alnus glutinosa 1965; cf. Schwabe 1985), while it becomes progressively community on wet soils. Several species differentiate replaced by conifers towards the upper part (Julin 1965). from the last-mentioned community, avoiding too high In the hyper-oceanic parts in SW and W Norway, also soil moisture, such as Acer platanoides, Actaea sp icata and Dryopteris filix-mas. In accordance with this, the M­ figures (Table 19) are much lower than for the Fraxinus excelsior-Alnus glutinosa community, but higher than for Table 20. Eutrophic alder-ash forests. Spearman rank correla- the elm-ash forests. The L-figures are among the lowest of tion (rs) between sample plot scores on the first two CA-ordina- tion axes and 14 environmental variables. The number of obser- all community types, indicating a strongly shading canopy. vations is 4 7. For further explanation, see Table 6. Two sub-communities are distinguished: - Hepatica nobilis sub-community (releves 1-21), in stands Variable Axis 1 Axis 2 from different regions, differentiated by Lonicera rs p rs p xylosteum and several species in the field layer, e.g. Hepatica nobilis and Viola riviniana, indicating a moder­ L 0.605 0.001 -0.707 0.001 T -0.298 0.05 0. 148 n.s. ately high soil moisture. Two forms can be separated: a c -0.167 n.s. 0.264 n.s. Typical form (relev�s 1-12) in stands from different parts M 0.933 0.001 -0.499 0.001 of Sweden, and a Polygonatum verticillatum form (releves R -0.287 n.s. 0.493 0.001 N 0.050 n.s. 0.325 0.05 13-21) in stands from Vastergotland and Norway, differ­ TEM YEAR 0.380 0.01 -0.553 0.001 entiated by Polygonatum verticillatum, Pulmonaria TEM FEB 0.507 0.001 -0.166 n.s. officinalis, etc. TEM JULY 0.176 n.s. 0. 144 n.s. - Dryopteris dilatata sub-community (releves 22-28), in PRE YEAR 0.556 0.001 -0.239 n.s. MAR IND 0.344 0.05 0.054 n.s. stands from Norway, positively characterized by Dryop­ CON IND -0.339 0.05 0.151 n.s. teris dilatata and the occurrence of some differential LAT -0.218 n.s. 0.428 0.01 species for the Fraxinus excelsior-Alnus glutinosa com­ LON -0.151 n.s. -0.170 n.s. munity, such as Stellaria nemorum and Chrysosplenium

Acta Phytogeogr. Suec. 80 62 M. Diekmann

altemifolium. The sub-community thus has an intermedi­ and with high abundance of Equisetum sylvaticum, E. ate position, which is also expressed by the comparatively arvense and E. pratense, in stands from Norway (3) and high M-figures (Table 19). In the ordination diagram in Sodermanland (1). Fig. 18, the releves of this sub-community (sub-unit 3) are This community, as well as related ones, have been fairly well separated from those of the Hepatica nobilis studied in detail in many different parts of Scandinavia. sub-community (sub-units 1-2). As synonyms can be mentioned: There are only a few studies of this community; how­ 'Filices-Typ' or 'Filices-Hepatica-Typ (Tapio 1953; ever, a comparison of releves from the literature with Makirinta 1968, 1982), 'Stellaria nemorum' -samhalle ' releves of Table 18 reveals the following synonyms and (Wallin 1973), Hillside Alnus incana forests (Fremstad & corresponding communities: 0vstedal 1978), 'Askskog' (groups 7-8, Brunet 1991), 'Athyrium-Oxalis-Typ' (Makirinta 1968), 'Geranium Alno incanae-Prunetum (Aune 1973; Fremstad & 0vstedal robertianum-Urtica dioica-samhalle' (Wallin 1973), 1978; Klokk 1980, 1982; Kielland-Lund 1981; Fremstad 'Askskog' (groups 5-6, Brunet 1991), Alno incanae­ & Normann 1982; Fremstad 1983), Ranunculo (jicariae) Fraxinetum (typicum and scrophularietosum) (Kielland­ -Ulmetum geetosum rivale (Crepis variant) and Alno­ Lund 1971, 1981; Berthelsen 1980), Ranunculo (jicariae) Ulmetum (Klotzli 1975a), Alno-Ulmetum glabrae, -Ulmetum geetosum rivale (in a Typical variant, KlOtzli Matteuccia struthiopteris variant (Fremstad 1979), Alno 1975a),Alno-Ulmetum glabrae, Typical variant (Fremstad incanae-Piceetum (Bj�mdalen 1980a), Equiseto sylvatici­ 1979), Corylo-Ulmetum glabrae (Blom 1980), Eurhyn­ Fraxinetum (Kielland-Lund 1981), Alnus incana forest chio-Fraxinetum (Blom 1980, 1982; 0vstedal 1985). (Odland 1981), Alno incanae-Fraxinetum chrysosple­ The forests occur on level or sloping ground, usually nietosum (Kielland-Lund 1981). not in connection with open (running) water. Less often, The stands of the Equisetum sylvaticum sub-commu­ they grow on terraces along small brooks. Where found nity (particularly releve 46) show some similarity with the on slopes, preferably the lower parts are occupied. Only Equiseto sylvatici-Fraxinetum described by Kielland-Lund occasionally, the ground becomes inundated. In the Alno (1981). This community is found at nutrient-rich spring incanae-Fraxinetum in SE Norway, the soils are devel­ sites enjoying a fairly warm and rainy climate; according oped as brown earth, pseudogley-brown earth or pelosol­ to Kielland-Lund (1981), it is restricted to SE Norway. In brown earth (Kielland-Lund 1981). Base saturation values other areas of Norway, and in Sweden, it is hardly possi­ vary between 44 and 61 %, the pH-values are about 5.0. ble to distinguish this community from other alder-ash forest types, since the character species given by Kielland­ Lund (Carex remota, C. sylvatica and Equisetum hyemale) Fraxinus excelsior-Alnus glutinosa community have a wider ecological and sociological amplitude here. The forests of this community are floristically well char­ An oceanic variant is the Carex remota-Alnus glutinosa acterized. Apart from the occasional occurrence of Ribes community described by Fremstad (1983) from western rubrum in the shrub layer, several constant species in the Norway (cf. also Blom 1980). field layerdifferentiate from theFrax inus excelsior-Prunus The forests are found on level or gently to moderately padus community, e.g. Stellaria nemorum, Chryso­ sloping ground on terraces along rivers and brooks, often sp lenium altemifolium and Ranunculus repens. Solanum at the bottom of a valley or ravine. Correspondingfo rests dulcamara and Galium palustre represent elements of in Norway occur on steep hillsides and talus slopes. In alder swamp forests. Matteuccia struthiopteris is often winter and spring, the ground is regularly inundated and dominant which, when reaching high cover values, causes has a high soil moisture even during summer due to the a decrease in species richness (cf. Fremstad & 0vstedal high ground water table caused by the proximity to the 1978; Odland 1981) (Fig. 19). Differential species are watercourse (cf. Aune 1973; Fremstad & 0vstedal 1978). also found among the bryophytes, e.g. Brachythecium In the Alno incanae-Prunetum in SE Norway, soils of the rivulare and Mnium hornum. All of the named species following types were found: brown 'Aue' soil, pseudogley­ require high soil moisture, which is also indicated by the brown earth, pelosol-brown earth and pseudog1ey-pelosol very high M-figures.Probably due to these extreme mois­ (on clayey, marine sediments) (Kielland-Lund 1981). Both ture conditions, several species that frequently occur in pH values (5.1-5.5) and base saturation values (61 -77 %) less wet eutrophic and mesotrophic forests are rare or are higher than for the Alno incanae-Fraxinetum (more or absent, such as Quercus robur, Corylus avellana and less correspondingto the Fraxinus excelsior-Prunus padus Hepatica nobilis. community). Two sub-communities are separated: a Poa trivialis sub-community (releves 29-43), weakly positively differ­ Regional comparison entiated by Poa trivialis, in stands fromBohuslan, Halland and Sodermanland, and an Equisetum sylvaticum sub­ In order to compare alder-ash forests and related commu­ community (releves 44-47), differentiated by Alnus incana nities from different regions in Fennoscandia, releves

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 63

Fig. 19. Eutrophic alder-ash forest (Fraxinus excelsior­ Alnus glutinosa community) in a ravine, late spring. Conspicuous are Matteuccia struthiopteris and Ranuncu­ lus ficaria, the latter species densely covering the ground. Svartedalen, Bohusliin,June 1990. Photo M. Diekmann.

from the literature were combined with the releves from the stands in Skane are characterized by some species not Table 18 in a new table. A CA-ordination of the total of occurring further to the north, such as Lamiastrum releves is presented in Fig. 20. It can be seen that the galeobdolon, Th alictrum aquilegifo lium and Cirsium releves from different regions are fairly well separated. oleraceum. Within the Boreo-nemoral zone, the SE Nor­ Stands from Finland are located in the right part of the wegian forests do not differ very much from the Swedish diagram, those from Skane in the upper left corner. The ones, except for Alnus incana largely replacing Alnus stands from the Boreo-nemoral zone in Sweden and SE glutinosa in the canopy. In the Boreal zone and in W Norway are concentrated in the middle, the W Norwegian Norway, the species composition changes considerably in ones in the lower middle. Table 21 gives the correlation all vegetation layers. Alnus incana is the dominant tree coefficients between the releve scores on the two first CA species. Betula pubescens and Picea abies (except in the axes and some environmental variables. Most correla­ western broad-leaved and pine forest region) are impor­ . tions are highly significant, with respect both to axis 1 and tant codominants. Here, Alnus incana shows a broad 2. The releve scores on axis 1 ( eigenvalue 0.365) are ecological amplitude and forms, apart from alder-ash strongly negatively correlated with R- and N-figures and forests, swamp forests on very wet soils, pioneer stands positively correlated with M-figures; thus soil moisture on cleared and grazed sites on immature soils, as well as shows an opposite trend to soil reaction and fertility, as in transitional communities on comparatively dry soils re­ Table 20. The strongest positive correlation with the placing conifer forests (Fremstad & 0vstedal 1978; cf. releve scores on axis 1 is shown by longitude. The abso­ Schwabe 1985); the species is considered as a 'weed' in lutely highest coefficients are found for the releve scores central Norway (Fremstad 1979). In the most rainy and on axis 2 (eigenvalue 0.342), showing strong positive oceanic parts of W Norway, Alnus glutinosa can be very correlations with the variables connected with tempera­ frequent as well (Fremstad 1983; 0vstedal 1985). Many ture (annual mean temperature, length of growing sea­ Nemoral, southerly distributed species are rare or lacking, son), consequently a negative correlation with latitude, especially in the field layer, such as Ranunculus ficaria and strong negative correlations with yearly precipitation and Mercurial is perennis. On the other hand, occur there and de Martonne's index. Axis 1 thus expresses the longi­ many boreal species which are not found further to the tudinal W-E gradient, whereas axis 2 expresses the latitu­ south or in corresponding forests in central Europe, e.g. dinal S-N gradient. Calamagrostis purpurea (cf. Schwabe 1985). Also ele­ The forests of the Nemoral zone in Skane resemble ments of conifer forests are frequent, e.g. Trientalis those of the Boreo-nemoral zone in Sweden, particularly europaea and Va ccinium spp. Apart from these common with respect to the tree layer dominated by Fraxinus fe atures, the forests of different regions within the Boreal excelsior, Ulmus glabra and Alnus glutinosa. However, zone show somewhat different species compositions. The

Acta Phytogeogr. Suec. 80 64 M. Diekmann

2 4 \ • 4 • 4 .4

4 "2 s. �

I � N (/) x 0 <(

-1

-2

2 -1 0 2 - AXIS 1

Fig. 20. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of 452 releves of eutrophic alder-ash forests. Numbers plotted onto the sample plot positions represent releves from this study and literature releves. Arrows indicate outliers not visible in the diagram. Finland: 1. Tapio (1953) 11 releves, Koponen (1967) 12 releves, Miikirinta (1968) 37 releves; Sweden: 2. 47 releves from Table 18; 3. Wallin (1973) 11 releves; 4. Brunet (unpubl.) 29 releves from Skfme; Norway: 5. Kielland-Lund (1981) 59 releves; 6. Aune (1978) 6 releves, Fremstad (1979) 39 releves, Klokk (1980) 68 releves, Klokk (1982) 38 releves; 7. Fremstad & 0vstedal (1978) 95 releves.

forests in the western parts of Norway are characterized phytosociological system in central Europe, even if it had by a few herbs (Cicerbita alpina and Viola biflora), but been possible in view of an occasionally similar species above all by the richness in bryophytes, both mosses and composition. Since the Boreo-nemoral zone is a large, hepatics. Frequent genera with several species are e.g. well-delimited area, this principle is preferable to an Brachythecium, Eurhynchium, Plagiomnium, Rhytidia­ attempt to squeeze the northern communities into a sys­ delphus (e.g. R. loreus) andPlagiothecium (P. undulatum). tem created for and adapted to the Nemoral and Mediter­ Correspondingfore sts in continental Finland have a much ranean zone in Europe (and which, in some cases, is based poorer bryophyte flora and lack many oceanic and sub­ on priority rather than critical revision). oceanic species. Instead, they show relatively high fre­ In the attempt to place the forest communities in the quencies of some boreal, continental plant species such as phytosociological system of the Braun-Blanquet School, Viola selkirkii.

Table 21. Eutrophic alder-ash forests: regional comparison. Spearrnan rank correlation (rs) between sample plot scores on 4.6 Syntaxonomy the firsttwo CA-ordination axes (Fig. 20) and 12 environmental variables. The number of observations is 452. For further expla­ This chapter aims at discussing the syntaxonomy of the nation, see Table 6.

Boreo-nemoral deciduous forest communities within the Variable Axis 1 Axis 2 framework of the Braun-Blanquet system. To fit the rs p rs p Scandinavian communities into this system is a difficult L 0. 181 -0.385 task, because the latter was based on phytosociological 0.001 0.001 T -0.430 0.001 0.542 0.001 studies in central and southern Europe, while vegetation c 0.390 0.001 -0.265 0.001 types in northern Europe were not taken into considera­ M 0.473 0.001 0.281 0.001 R -0.567 0.278 tion. 0.001 0.001 N -0.692 0.001 0.035 n.s. The Braun-Blanquet approach has only been applied TEM YEAR -0.390 0.001 0.726 0.001 in a few phytosociological studies and classifications of PRE YEAR -0.012 n.s. -0.697 0.001 GRO PER -0.224 0.827 deciduous forests in northern Europe (see Chapter 1). A 0.001 0.001 MAR IND 0. 103 0.05 -0.786 0.001 common feature in these classifications was that associa­ LAT 0.271 0.001 -0.836 0.001 tions have not been adopted from the already existing LON 0.505 0.001 -0.021 n.s.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 65

the nomenclatural rules given by Barkman et al. (1986) tremulae by Tiixen 1951) and a Melico-Quercetum (Table and Weber (1988) were followed. In order to do this, the 22). Both associations lack good character species, but are concept of character species, differential species and the floristically well separated from each other as well as general, characteristic species composition, including con­ from Fagetalia communities of the same area. Outside the stants (Westhoff & van der Maarel 1973), has been ap­ distribution area of Picea abies in S Norway, Trientalis plied. Efforts were made to retain as many already-de­ europaea, Plagiothecium undulatum, P. laetum and scribed associations as possible, in order not to create Dicranum majus are considered as locally restricted char­ even more nomenclatural confusion. For higher units of acter species of the Populo-Quercetum (A. BjlZ)mstad the Braun-Blanquet system, i.e. classes, orders and alli­ 1971 ). In this study, these associations were distinguished ances, the central European scheme of syntaxa has been on a sub-community level, while the main division was a adopted. Weakly characterized communities will simply geographic one. However, Table 4 reveals some striking be referred to as communities without specific rank. similarities between the sub-communities on very poor In the following, the syntaxonomy of the Boreo­ soils on one side and those on comparatively richer soils nemoral communities will be briefly discussed with re­ on the other side. Therefore, BjlZ)mstad's classification spect to the characterization by character species, the shall be adapted. In conclusion, the oligotrophic oak for­ syntaxonomical nomenclature and hierarchical ranking ests in eastern Sweden (cf. Berglund 1963; Riihling & according to the Braun-Blanquet system, and correspond­ Tyler 1986) and those in western Sweden and Norway (cf. ing communities in other areas. From the vast amount of Ivarsson 1962; A. BjlZ)mstad 1971 and 0vstedal 1985) are phytosociological literature from central Europe, only a syntaxonomically treated as geographic races of Populo­ few modem, comprehensive publications were chosen, Quercetum and Melico-Quercetum (eastern and western namely Oberdorfer (1992) and Pott (1992). For compari­ races, respectively, Table 22). son with forest communities in Britain, mainly Klotzli From northern central Europe, two very similar asso­ (1970), Birse (1982) and Rodwell (1991) were quoted. ciations have been described (Table 22): the Betulo­ Quercetum roboris (corresponding to the Populo­ Oligotrophic oak forests Quercetum petraeae) and the Holco mollis-Quercetum ( Fago-Quercetum petraeae or Violo-Quercetum, corre­= The oligotrophic oak forests are floristically well charac­ sponding to the Melico-Quercetum). Recently, it has been terized by a number of acidophilous species, whereas proposed to unite all birch-oak forests of this region on many basiphi1ous and nitrophilous species are absent. Pleistocene sandy soils in the association mentioned first Corresponding forests in central and western Europe were (Heinken 1993). Towards the east, the oligotrophic oak early recognized as so-called 'Birken-Eichenwalder' forests in central Europe show a transition primarily to (birch-oak forests) and assigned to the order Quercetalia oak forests with an increasing abundance of conifers robori-petraeae containing only one alliance, Quercion (mainly Pinus sylvestris) and continental species, to be­ robori-petraeae. The oligotrophic oak forests lack 'good' come progressively replaced by almost pure pine forests character species, since almost all of the acidophilous of the alliance Dicrano-Pinion (Ellenberg 1986). Such a species also occur in (or prefer) conifer or poor beech gradual transition is hardly observable in northernEurope. forests, forest fringes and clearings on acid soils, as well In Britain, oligotrophic oak forests have a wide distri­ as poor heathlands and meadows (Ellenberg 1986; bution and occur in several communities (cf. Rodwell Oberdorfer 1992). The problem mightbe overcome by a 1991). Particularly in the western parts of the country, restriction of character species to structural vegetation these are characterized by a high frequency of oceanic types or formations (such as vegetation formed by one or species, both vascular plants (/lex aquifo lium, Ulex spp., several layers of tall phanerophytes, i.e. forests and Hyacinthoides non-scripta, Erica cinerea, etc.) and bryo­ shrublands), as proposed by Dierschke (1992, 1994). In phytes (Bazzania trilobata, Diplophyllum albicans, that case, the following species could be regarded as Campylopus spp., Hypnumjutlandicum, etc.). As British character species of the Quercion robori-petraeae: counterparts to the Populo-Quercetum, the following can Melampyrum pratense, Lathyrus montanus, Polypodium be mentioned: Querco-Betuletum (Klotzli 1970), Birch­ vulgare and Hieracium vulgatum. Regionally, also other oak and Birch woodland (Peterken 1981), Deschampsia taxa may be regarded as character species, such as Frangula flexuosa-Pteridium aquilinum-type (Bunce 1982) and alnus, Carex pilulife ra, Polytrichum fo rmosum and Quercus spp.-Betula spp.-Deschampsia flexuosa wood­ Lophocolea heterophylla in S Norway (A. BjlZ)mstad 1971). land (Rodwell 1991) from the English lowlands; Blechno­ The differentiation of oligotrophic oak forests into one Quercetum (KlOtzli 1970; Birse 1982) and Quercus community on very acid and poor soils and another on less petraea-Betula pubescens-Dicranum majus woodland acid and somewhat nutrient-richer soils (cf. Chapter 4.2) (Rodwell 1991) from the western and northern, often was suggested by BjlZ)mstad (1971), who distinguished a higher elevated parts of Britain. The Melico-Quercetum Populo-Quercetum (first described asQuer ceto-Populetum corresponds to the Quercus robur-Pteridium aquilinum-

Acta Phytogeogr. Suec. 80 66 M. Diekrnann

Table 22. Deciduous forest communities, their syntaxonomical equivalents (associations, communities) and placement in orders and alliances, as well as corresponding communities in central Europe.

Community Association Alliance Corresponding (community) communities in central Europe

Quercetalia robori-petraeae Tx. (1931) 1937 Oligotrophic oak-forests Que reus robur-Betula pendula community Populo-Quercetum Tx. Quercion Betulo-Quercetum Trientalis europaea sub-community 1951 nom. inv., western race robori-petraeae Tx. 1930 nom. inv. (Malcuit 1929) Tx. 1974 Quercus petraea-Frangula alnus community western race Br.-Bl. 1932 Trientalis europaea sub-community

Quereus robur-Betula pendula community Melico-Quercetum Holco mollis­ Viola riviniana sub-community Bjfi)rnstad 1971, eastern race Quercetum (robori­ petraeae) Lemee 1937 Quercus petraea-Frangula alnus community western race corr. et em. Oberd. Viola riviniana sub-community

Fagetalia sylvaticae Pawl. 1928 Mesotrophic mixed deciduous forests Quereus robur-Euonymus europaeus community Ulmo-Tilietum Kielland­ Tilia platyphyllis­ Aceri platanoidis­ Tilia cordata sub-community Lund in Seibert 1969 em. Acerion pseudo­ Tilietum platyphyllis Diekmann 1994, bland race platani Klika 1955 Faber 1936/ Querco petraeae­ Quercus robur-Tilia cordata community mainland race Tilietum platyphyllis Rlih1 1967

Quercus robur-Fraxinus excelsior community Quercus robur-Fraxinus excelsior community Quercus robur-Euonymus europaeus community Oxalis acetosella sub-community, Corylus avellana Quereus robur-Corylus avellana form community Carpinus betulus Carpinus betulus community Carpinion betuli Stellario holosteae­ form Issler 1931 em. Carpinetum betuli Oberd. 1957 Oberd. 1957

Eutrophic elm-ash forests Ulmus glabra-Fraxinus excelsior community Ulmo-Fraxinetum Sjogren Tilia platyphyllis­ Fraxino-Aceretum 1971 ex Diekmann 1994, Acerion pseudo­ (W. Koch 1926) Tx. em. mainland race pia/ani Klika Th. Muller 1966/ 1955 Querco-Ulmetum Ulmus minor-Fraxinus excelsior community bland race minoris Issler 1924

Eutrophic alder-ash forests Fraxinus excelsior- Fraxinus excelsior­ Alno- Ulmion Pruno-Fraxinetum Prunus padus community Prunus padus community Br.-Bl. & Tx. 1943 Oberd. 1953

Fraxinus excelsior- Alno incanae-Prunetum padi Stellario nemorum-Alnetum glutinosae Lohm. 1957 I Alnus glutinosa community Kielland-Lund 1971 Carici remotae-Fraxinetum W. Koch exAune 1973 1926 ex Faber 1936/A lnetum incanae Liidi 1921

Rubus fr uticosus woodland of lowland Britain and to the trophic oak forests are absent or restricted to certain sub­ Quercus petraea-Betula pubescens-Oxalis acetosella units. On the other hand, it is more difficult to place these woodland of northernand westernBritain (Rodwe11 1991 ), communities in an alliance or association. Most character both occupying somewhat richer soils than the communi­ species of the alliances are absent or rare within the ties mentioned previously. Both woodland types have a Boreo-nemoral zone, particularly with respect to the large number of synonyms in the literature (cf. Rodwell Fagion sylvaticae and Carpinion betuli. The mesotrophic 1991). forests will be divided into two groups for further discussion: Mesotrophic mixed deciduous forests 1. Forests moderately influenced by human activity: mesotrophic mixed deciduous forests with linden; Both mesotrophic and eutrophic forests can easily be 2. Forests strongly influenced by former land use: oak-ash assigned to the order Fagetalia, based on a large number forests on Gotland, oak-hazel forests and hornbeam forests. of species, e.g. Fraxinus excelsior, Ulmus glabra, Tilia cordata, Acer platanoides, Dryopteris filix-mas, Pulmo­ 1. It is close at hand to refer the northern Tilia cordata­ naria offi cinalis, Mercurialis perennis, Milium eff usum, rich forests of this group to the alliance Tilia platyphyllis­ etc. Most character and differential species of the oligo- Acerion pseudoplatani (Table 22), for two reasons: firstly,

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 67

several character species of this alliance reach high fre­ on groundwater-influenced gley soils and has a species quencies, namely the trees Ulmus glabra and Acer composition much different from the Ulmo-Tilietum. No platanoides, as well as Ribes alpinum and Actaea sp icata. floristically similar counterparts to the Fennoscandian On the other hand, character species of the other alliances mesotrophic forest exist in Britain. are much less frequent or absent, except for Tilia cordata 2. The forests of this group 'replace' the Ulmo-Tilietum which is considered as a Carpinion-species, although where the anthropogenic influence has been particularly fairly common in Tilio-Acerion communities (Oberdorfer strong. The oak-hazel forests on bland and the mainland, 1983). Secondly, the mesotrophic forests are usually found as well as the oak-ash forests of Gotland, lack many in regions with a favourable macroclimate (e.g. bland) or species characteristic for the Tilio-Acerion or Ulmo­ on more or less unstable screes and slopes, preferably of Tilietum. They are therefore called Quercus robur-Corylus southern aspect (mainland), which are favoured by a avellana community and Quercus robur-Fraxinus locally warm and continental climate; in central Europe, excelsior community, respectively (Table 22). these sites are usually occupied by Tilio-Acerion forests Corresponding communities in central Europe can (cf. Oberdorfer 1992). This is also expressed by the occur­ hardly be found, since forest management and other land rence of some Quercetalia pubescentis or Trifolio­ use practises there have been widely different from north­ Geranietea species such as Lathyrus niger, Campanula em Europe (cf. Ellenberg 1986); particularly wooded persicifolia and Polygonatum odoratum. The position of meadows did not have such a great importance. The the mesotrophic forests within the Tilio-Acerion is also hazel-rich shrub-forests on screes as described by, e.g. emphasized by A. Bj !Zimstad( 1971), Kielland-Lund ( 1971, Hofmann (1958), Winterhoff (1965) and Oberdorfer (1992) 1981) and Aune (1973). Mesotrophic forests were first from central Europe, differ considerably from northern described by Kielland-Lund in Seibert (1969) as Ulmo­ oak-hazel forests. Structurally, however, these resemble Tilietum and later studied by e.g. A. Bj!Zimstad (1971), the hazel-groves as found on, e.g. bland, or in W Norway. Aune (1973) and Kielland-Lund (1981). However, the The Carpinus betulus forests on bland lack several of circumscription of this association is somewhat emended the character species of the Tilio-Acerion and Ulmo­ here to comprise the mesotrophic forests only and not the Tilietum. They are therefore simply referred to as Carpinus eutrophic elm-ash forests (see below). In the strict sense, betulus community. Structurally, they are suggestive of this association has no character species. However, in the beech forests, while the high but local abundance of Boreo-nemoral zone, several taxa can serve as regional hornbeam suggests a placement in the alliance Carp inion. character species, e.g. Tilia cordata, Lathyrus vernus, Carpinus forests with a beech forest structure are also L. niger and Bromus benekenii. Besides, when restricting known from, e.g. eastern Poland (Bialowieza), situated the character species to formations, also some forest fringe outside the distribution area ofFagus (cf. Ellenberg 1986). species are confinedto the Ulmo-Ti lietum, e.g. Laserpitium latifolium and Melampyrum nemorosum. Some of the Eutrophic elm-ash forests named species can also occur in conifer (mainly pine) forests on calcareous bedrock, as described from SE Nor­ Like the mesotrophic forests, the eutrophic elm-ash for­ way and different parts of Sweden (Kielland-Lund 1971, ests can be referred to the alliance Ti lio-Acerion, due to 1981; Bj!Zimdalen 1980b, 1985). However, the extent to high frequencies ofUlmus glabra, Acerplatanoides, Ribes which these pine forests have been only slightly modified alpinum, Campanula latifolia and Actaea sp icata, and or, as a whole, have become favoured by human activity also by Ribes uva-crispa which is not native, but has (e.g. grazing, cf. Kielland-Lund 1981), is not yet clear. spread from cultivation. Character species and differen­ From central Europe, forests with a similar species tial species of the alliance Alno-Ulmion are rare and composition are not known. However, two associations confined to certain sub-units, except for some taxa (e.g. can be named as counterparts on ecologically similar sites Prunus padus, Gagea lutea, Elymus caninus and Plagio­ (Table 22): the Aceri platanoidis-Tilietum platyphyllis on mnium undulatum) which are very frequent in different unstable, steep and boulder-rich slopes with more or less kinds of mesotrophic and eutrophic forests in Scandina­ base-rich soils, and the Querco petraeae-Tilietum platy­ via, even on mesic soils. In common with other Tilio­ phyllis on similar sites, but on siliceous bedrock with Acerion communities in Europe, the northern elm-ash acidic soils (Oberdorfer 1992). In these central European forests occupy the most nutrient-rich sites within the communities, Tilia platyphyllos(nearly absent in Scandi­ Boreo-nemoral zone and thus also ecologically corre­ navia) is generally more common than T. cordata, but the spond to this alliance. The communities can be assigned latter becomes more abundant on nutrient-poor, acidic to the association Ulmo-Fraxinetum, first named by sites. These forests are mainly found in mountainous Sjogren in Kielland-Lund (1971). Among the species areas and are rare or absent in the lowlands (cf. Ellenberg differentiating from the mesotrophic forests, Campanula 1986). However, Iversen (1958) described a fairly natural latifolia and Stachys sylvatica cannot serve as character Tilia cordata forest from Denmark, but this stand occurs species, since both are also frequent in alder-ash forests

Acta Phytogeogr. Suec. 80 68 M. Diekrnann

on moist or wet soils. Allium ursinum and Corydalis The phytosociological position of alder-ash forests bulbosa can be regarded as regional character species, but and related black or grey alder forests is unclear, due to are not very common and restricted to certain sub-units. the wide distribution and great variability of these forests, Finally, Alliaria petiolata has its optimum in shady forest particularly in western and northern Scandinavia. From fringes on fertile soils. Norway, several associations have been described which The corresponding community in central Europe (Ta­ are often restricted to certain geographic areas, e.g. by ble 22) is the Fraxino-Aceretum pseudoplatani, a mixed Kielland-Lund (1981), Fremstad (1979, 1983) and deciduous forest community rich in tall herbs which is 0vstedal (1985). The Fraxinus excelsior-Prunus padus found on fertile, often stony soils, both on slopes in shady community has no character species and is, in general, and humid gorges and on alluvial sediments along hill­ weakly characterized. It cannot be assigned to a separate sides and brooks (Oberdorfer 1992; Pott 1992). This association. In contrast, the Fraxinus excelsior-Alnus association shows high frequencies of many of the Ulmo­ glutinosa community is characterized by several taxa Fraxinetum species, as well as of many other nitrophilous which, within the Boreo-nemoral zone, are more or less species not occurring in northern Europe. Floristically, restricted to this forest type, namely Chrysosplenium the elm-ash forests with Ulmus minor on bland are also alternifolium, Matteuccia struthiopteris, Cardamine amara suggestive of the 'Eichen-Ulmen-Auwald' (Querco­ and Carex remota. It is assigned to the association Alno Ulmetum minoris, assigned to the Alno-Ulmion ), found in incanae-Prunetum, first described by Kielland-Lund the lowlands of central and easternEurope along the large (1971) and later treated by, e.g. Aune (1973), Fremstad & rivers (cf. Oberdorfer 1992). However, true 'Auwald' 0vstedal (1978) and Klokk (1980, 1982). communities do not seem to occur at all in northern The Fraxinus excelsior-Prunus padus community cor­ Europe. The northern elm-ash forests have counterparts in responds to the Pruno-Fraxinetum in central Europe (Ta­ Britain where the base-rich soils of the lowlands are ble 22) which grows on plains and in depressions with a occupied by mixed deciduous forests dominated by fairly high ground water table (not as high as in Alnus Fraxinus excelsior, Acer campestre, A. pseudoplatanus, swamp forests, but with more pronounced fluctuations, Carpinus betulus, Tilia spp. and Ulmus spp., etc. These cf. Oberdorfer 1992). The Alno incanae-Prunetum has a have been described as, e.g. Dryopterido dilatatae­ few counterparts in central Europe which, however, are Fraxinetum and Querco-Fraxinetum (KlOtzli 1970), partly not very well differentiated: the Carici remotae­ Querco-Ulmetum glabrae (Birse 1982) and Fraxinus Fraxinetum on very wet, often inundated soils (usually on excelsior-Acer campestre-Mercurialis perennis woodland calcareous bedrock) along small brooks and springs in (Rodwell 1991). Both Birse (1982) and Rodwell (1991) montane areas (similar to the Equiseto sylvatici­ distinguished a sub-unit characterized byAllium ursinum. Fraxinetum, cf. Kielland-Lund 1971, 1981); the Stellario nemorum-Alnetum glutinosae on less wet and poorer soils (usually on siliceous bedrock) along brooks and rivers in Eutrophic alder-ash forests submontane-montane areas, and the Chrysosplenio The alder-ash forests can be assigned to the Alno-Ulmion, oppositifolii-Alnetum glutinosae at base-rich springs in based on high frequencies of both character species the Pleistocene lowlands in northern Germany (Moller (Equisetum pratense, Stellaria nemo rum) and differential 1979; Pott 1992; Oberdorfer 1992). Besides, the Alnetum species (Geum rivale, Athyrium fi lix-femina, Crepis incanae, found on immature, sandy or gravelly soils along paludosa, etc.) of this alliance. In the Fraxinus excelsior­ montane brooks and rivers in the Alps and surrounding Alnus glutinosa community, further character species can areas, shows some similarities to alder-ash forests, par­ be found (e.g. Matteuccia struthiopteris, Cardamine amara ticularly to the more or less pure Alnus incana forests in and lmpatiens noli-tangere). The placement of alder-ash northern Europe(Klo kk 1980; Schwabe 1985; Oberdorfer forests, as well as the west Norwegian and boreal grey 1992). There are only few descriptions of alder-ash for­ alder forests, in the Alno- Ulmion has been emphasized by ests from Britain, e.g. as Pellio (epiphyllae)-Alnetum many authors, e.g. Kielland-Lund (1971, 1981), Aune (Klotzli 1970), as Alder woodland (Peterken 1981), Crepis (1973), Wallin (1973), Klokk (1980), Odland (1981) and paludosa-Alnus glutinosa association (Birse 1982) and as 0vstedal (1985). However, Alnus incana forests in Nor­ Alnus glutinosa-Fraxinus excelsior-Lysimachia nemo rum way at higher altitudes are closely related to subalpine woodland (Rodwell 1991). tall-herb communities. Several species of the alliance Lactucion alpinae are fairly common, such as Aconitum septentrionale,Athyrium distentifolium, Cicerbita alpina, Myosotis decumbens and Ranunculus platanifolius (cf. Fremstad & 0vstedal 1978; Fremstad 1979 and Odland 198 1).

Acta Phytogeogr. Suec. 80 5 The population structure of trees and forest dynamics

5.1 Introduction In this static approach, successional trends are in­ ferred!.fro m the age an&br size, structure· of the major tuee It has been pointed out in Chapter I that an deciduous populations of a forest Demographic data on species .. u forests in the Boreo-nemoral zone of Fennoscandia.have composition reveru intormation abot: me, forest history, in some way been influencedby hmmm activity. Due to the rate of sacces:s]OJ11la] dr:mg,e amr

Acta Phytogeogr. Suec. 80 70 M. Diekmann

will also differ in their age/diameter ratio. This problem Unfortunately, many valuable stands had been par­ might be overcome by introduction of a species-specific tially or completely cleared before the start of the struc­ growth constant, but due to differing environmental con­ tural investigations. The research concentrated on the two ditions, the intra-specific variation is high (Goff 1968). main types of deciduous forests in the study area, Information on the relationship between age and size or mesotrophic mixed deciduous forests and eutrophic elm­ height has been collected for North European conifer and ash forests. Only a few oligotrophic oak forests were trivial deciduous tree species (e.g. Hytteborn et al . 1987; included due to their rareness in eastern Sweden. In total, Hofgaard 1993a), but is hardly available for deciduous 69 stands were studied: 30 on Gland, 13 in Smaland and hardwood tree species, even if some basic autecological 26 in Uppland/Sodermanland. data for the latter have been compiled by Prentice & Helmisaari (1991). The inference of forest dynamics from Sampling procedure size rather than age is open to criticism, but the alterna­ tives - age determination and periodic re-measurements ­ In each selected stand, a transect of 10 m width was laid are very laborious and time-consuming. In the framework out at random. However, in some small stands, the transect of this typological-ecological study of many forest stands had a width of only 5 m and/or had to be placed in a certain it was not possible to perform systematic age determi­ direction in order to avoid that neighbouring stands would nations. Besides, the approach assumes that the relative be included or exert influence. After having determined a proportion of a species in a size class is not significantly starting-point, the first 100 (in some cases 50) trees with a altered as stems 'move' into larger size classes, i.e. that girth of minimally 4 cm at breast height (ea. 1.4 m height), there are no interspecific differences in rates of mortality. corresponding to ea. 1.3 cm in diameter, were sampled. This is certainly not true for seedlings and smaller sap­ Parameters recorded were species identity, vitality and lings (see e.g. Hytteborn 1986), but may still be valid for girth at breast height. Tree specimens with several stems - larger size classes. Differences in the potential size and due to e.g. resprouting resulting from former cutting - age can also be important, since they affect the ability of a were sampled separately for each stem. Girth values were species to outshade or outlive another species. converted into diameter values (DBH) for further analyses. Despite these problems, the static approach yields valuable ecological insight. It offers a means to indicate Data treatment successional trends of both the past and near future. Therefore, a tree size analysis was carried out. In conclu­ Some tree species are understorey species, which are sion, the study aims (1) to analyze the population structure usually not able to reach the upper tree layer, such as Malus of trees in the most common communities of deciduous sylvestris, Prunus padus andSorbus aucuparia. They were hardwood forests in eastern Sweden, (2) to infer the therefore excluded from further analyses, the more so as successional trends of the main tree species from their they, with a few exceptions, were present only in small population structure, and, finally, (3) to interpret these numbers. Then, the total number of remaining trees and the trends against the background of the environment and total basal area were determined for each stand. Basal area history of the forests. is defined as the area outline of a plant near the ground surface (Mueller-Dombois & Ellenberg 197 4 ). For practi­ cal purposes, the area outline at breast height is usually 5.2 Methods taken as basal area, although the real basal area is some­ what larger. All diameter values were divided into six arbitrarily defined DBH classes: class 1: 1.3-8.9 cm, 2: 9- Selection of forest stands 16.9 cm, 3: 17-24.9 cm, 4: 25-32.9 cm, 5: 33-40.9 cm, 6: For the investigation of forest structure and dynamics, more than 41 cm. Now, the number of trees in each DBH three areas were selected: The Baltic island of Gland, class could be calculated, as well as the percentage of trees eastern Smaland and the region around Lake Malaren (number of trees in a DBH class/total number of trees). belonging to the provinces ofUppland and Sodermanland. Betula pendula and B. pubescens were treated to­ In all stands studied, vegetation analyses had been carried gether, and so were Que reus petraea and Q. robur (main­ G out in an earlier stage of the project. Beyond the criteria land) and Ulmus glabra and U. minor ( land). The fol­ for the performance of vegetation analyses (see Chapter lowing parameters were determined for each species/ 3), selected stands had to meet the following standards: genus in a stand: - habitat uniformity and structural and compositional ho­ - basal area and relative basal area, i.e. basal area/total mogeneity on a ground area large enough to sample basal area of trees included; demographic data on trees; - number of trees and relative density, i.e. number of trees/ - absence of recent (i.e. during last the 20-30 yr) logging total number of trees included, for both the whole stand of larger trees, and of recent thinning of the understorey. and each diameter class.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 71

Table 23. Survey of forest stands included in the investigation of forest structure and dynamics.

Forest type Region Community Number of stands

Oligotrophic Mainland Quereus robur-Betula pendula community 4 oak forests bland 1

Mesotrophic Mainland Quercus robur-Tilia cordata community 17 forests bland Quercus robur-Euonymus europaeus community 13

Eutrophic Mainland Ulmus glabra-Fraxinus excelsior comm. 18 elm-ash forests bland Ulmus minor-Fraxinus excelsior comm. 16

Since the aim was to reveal general successional trends, 5.3 Results rather than to describe single forest stands, the forest stands were clumped to groups. First, the material was Relative basal areas divided into stands from bland (n 30) and those from = the mainland (n 39), since (a) the two regions consider­ A survey of the forest stands is given in Table 23. While ably differ in their= deciduous forest communities (see 34 stands represented elm-ash forests, 30 stands belonged Chapter 4), and (b) silviculture and land use may have to mesotrophic forest communities. Only five stands rep­ been different. Then, the classificationof the stands from resented oligotrophic oak forests. The relative basal areas Chapter 4 was adopted. For the main clusters, the means of the most important tree species were compared be­ were calculated of the percentage of trees in different tween mesotrophic and eutrophic forests (Table 24). The DBH classes and of the relative densities of species in comparisons confmn the general results obtained by the different DBH classes. Since many of these relative den­ cluster analysis presented in Chapter 4. As expected, sity values for rarer species were zero and their distribu­ Fraxinus excelsior and Ulmus spp. showed significantly tion was not normal, the functional relationship between higher values in elm-ash forests compared with meso­ mean relative density and DBH class for a species was not trophic forests, both on bland and on the mainland. In analyzed with regression analysis, but with the Spearman contrast, Quercus spp. had significantly lower values. rank correlation test (Siegel & Castellan 1988). Acer platanoides did not show any clear trend, whereas Differences in the relative basal area of species be­ Tilia cordata tended to be more frequent in mesotrophic tween the main clusters were checked with Student's forests. This relationship between the relative basal areas t-test. In order to further test the relationship between of tree species and the species composition as expressed species composition of field and bottom layer and basal by the position of the releve scores on CA axis 1 is areas of tree species, the releves were ordinated using confirmed by the correlation tests presented in Table 25. Correspondence Analysis (CA, ter Braak 1987), exclud­ ing woody species from the analysis. Values for the Diameter distribution of trees relative basal area of species were then correlated with the resulting releve scores on CA axis 1, using the Spearman The diameter distributions of trees of all species together rank correlation test. show the same pattern in eutrophic elm-ash forests and The successional stability of a stand was assessed mesotrophic forests on both bland and the mainland (Fig. using both density and basal area data. If the average diameter of a species in a stand differs from that of other species, relative density and relative basal area of this species will differ. The successional stability can there­ Table 24. Comparison of relative basal areas (rba) of the most fore be computed by means of differential weighting important tree species/genera between mesotrophic (Meso) and using the following formula (Goff 1968): eutrophic elm-ash forests (Eu). Given are t-values of t-tests. p < 0.05; p < 0.01; p < 0.001. = * = ** = ***

CC sum RDi - RBi 2, Acer Fraxinus Que reus Tilia Ulmus = (I I) I

Oland: where CC is the compositional change (corresponding to rba (Meso) 0.042 0.082 0.579 0.216 0.035 successional stability), RDi the relative density and RBi rba (Eu) 0.053 0.306 0.245 0.099 0.218 t-value (df 27) -0.506 - 2.930 3.520 ** 1.720 -2.410. the relative basal area of the ith species. CC is = •• dimensionless and can vary between 0 (no compositional Mainland: rba (Meso) 0.072 0.004 0.662 0.141 0.033 change) and 100 (maximal compositional change). rba (Eu) 0.105 0.160 0.233 0.085 0.276 t-value (df 33) -0.774 - 3.733 4.428 0.787 -3.65 1

= ••• ••• •••

Acta Phytogeogr. Suec. 80 72 M. Diekmant1;1fl

21a-d): the mean number (or mean percentage) nf trees Table 25.. Spearman rank correlation (rs) of relative basal areas was highest in the smallest DBH class and decueased of species/genera of releves with releve scores on CA axis 1. p 0.05; p < 0.01; p < 0.001. steadily with increasing diameter, but the value for DBH < * = ** = *** = class 6 was usually slightly larger than the one for class 5. Acer Fraxinus Quere us Tilia Ulmus With respect to the mean densities ofparticular species in bland 0.103 0.6 19 0.599 0.373 0.636 different DBH classes, usually ;a similar pattern of .a • Mainland 0.147 0.730 ••• -0- .709 ••• --0. 181 0. 635 ••• falli.Qg curve could be observed for all :species except ••• ••• ••• Quercus spp., which, in most stands, .displayed a rising curve with particularly high values in DBH class 6. The Compositional change oligotrophic oakfor ests on the mainland showed a differ­ ent pattern, with, on average, 2/3 of all trees in DBH dass The histograms of the compositional change (CC) are 1 and low figures for other DBH classes (Fi,g. 21e). shown in Fig. 22. For bland (Fig. 22a), many stands had However, not all stands revealed the kind of diameter CC values between 10 and 20 and between 30 and 40, distribution described above. In Fig. 21f, 11 .stands from respectively. In four stands representing mesotrophic for­ bland and the mainland were compiled, representing oak­ ests CC exceeded 60. Two stands showed lower values hazel forests, a structurally .special sub-type of the than 10: an oligotrophic oak forest and a eutrophic elm­ mesotrophic forests. The mean number of trees was more ash forest on moist, nutrient-rich soil, very much domi­

similar in all DBH classes than in the communities men­ nated in all DBH classes by Quercus robur and Ulmus tioned above, with a fairly high value in class 6 and only a minor, respectively. For the mainland forests (Fig. 22b ),

:slightly larger value in class 1. In some cases, hardly any CC values between 20 and 40 were most common. A regeneration of trees could be observed.

50 a 50 b

40 40 ., gQ) J 0 30 0 30 � � E � 20 � 20

c c Ql <11 � 10 � 10

0 0 2 3 4 5 6 2 3 4 5 6

50 c d

50 40 "' Q) Q) Sl40 0 30 � 0 30 Q) .0E � � 20 E a e 20 Fig. 21. Mean number of trees ::.\!Q) 10 c �"' 10 in DBH classes, only hardwood trees included. a. Eutrophic 0 0 2 3 4 5 6 b 3 4 5 6 elm-ash forests, land (N 16); b. Mesotrophic mixed decidu-= 60 50 ous forests, bland (N 13); e f = c. Eutrophic elm-ash forests, 50 40 mainland (N 18); d. Meso-

"' "' = �40"' gQ) trophic mixed deciduous for- 0 0 30 ests, mainland (N 17); e. Oli- 30 2l � = 2 � gotrophic oak forests, mainland 20 (N 4); f. 'Oak-hazel' forests, a 2o c:: = "' bland + mainland (N 11). � � 10 10 DBH classes: 1: 1.3-8.9= cm,

0 0 2: 9-16.9 cm, 3: 17-24.9 cm, 2 3 4 6 2 3 4 5 6 4: 25-32.9 cm, 5: 33-40.9 cm,

DBH class DBH class 6: > 40.9 cm.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 73

1 00 Qrurcus 0.25 bland a robur a 90 0.20 )8 g 80 0.15 -@ 70 8 0.10 >- �1! 60 0.05 � � 50

0 20 40 60 80 100 � 4() 30 CC class fil � 20 Mainland 10 0.25 b 0 --'------r------"""1-__...... _ 0.20 2 3 4 5 6

c.o 0. 15 i 100 Fruxinus �xc�lsior 8 0. 10 90 b 0.05 80 � g 70 0 > 20 40 60 80 100 -� 60 CC class "0 50 .,CD � 40 Fig. 22. Frequency distribution of forest stands in classes of � 30 compositional change (CC). a. bland (N 30); b. Mainland (N fil � 20 = = 39). 10

0 2 3 4 5 6

100 A platanoules fairly large number of stands showed extreme values. In 90 cer c seven stands CC exceeded 60. These belonged to both g 80 mesotrophic forests and eutrophic elm-ash forests. CC 70 >- values lower than 10 were shown by three stands, one � 60 � 50 eutrophic elm-ash forest with dominance of Ulmus glabra � 40 in all DBH classes, as well as two oak-hazel forests, � 30 composed almost exclusively of and with 20 Quercus robur 2 a very low number of trees in smaller size classes. 10 0 2 11 ... 3 4 5 6I

Relative densities of species 1 00 Ulmus spp. 90 d

Mean relative densities (MRD) of species were calculated g 80 separately for regions and forest types. Normally, only the 70 > five most important species/genera were examined, which � 60 � 50 are Quercus spp., Fraxinus excelsior, Ulmus spp., Acer � 40 platanoides and Tilia cordata. §; 30 Eutrophic elm-ash forests on bland (16 stands, Fig. 2CD 20 23a-e) and showed high MRD values : Que reus Fraxinus 10 0 in DBH classes 4 to 6, and values declined rapidly with 2 3 4 5 6 decreasing diameter. The correlations between MRD and 1 00 Tilia cordala DBH were statistically significant for both species (r5 = 90 e 80 0.448, p < 0.001 and r5 = 0.267, p < 0.05, respectively, N g 70

= 93; note that is less than 16 6 because some DBH >- N x 60 classes are empty). In contrast, significant negative corre­ � � 50 lations were shown by Acer (r5= - 0.594,p < 0.001), Tilia � 40 (r5 = - 0.243, p < 0.05) and Ulmus (r5 = - 0.341,p < 0.001) a; 30 with higher MRD values in the three smallest DBH classes. � 20 Mesotrophic forests on bland (13 stands, Fig. 24a-e): 10 0 All trends were statistically significant. For Quercus, 3 4 5 6

MRD values were continuously increasing with increas­ DBH class ing DBH class (r5= 0.811, p < 0.001, N = 75), while they Fig. 23. Frequency distribution of mean relative density over were decreasing for the four other species. While Acer DBH class for five deciduous hardwood tree species in eutrophic 0.619,p 0.001), 0.288,p 0.05) (r5= - < Fraxinus (r5=- < elm-ash forests on bland (N 16). a. Que reus robur; b. Fraxinus and Ulmus (r5= - 0.302, p < 0.01) were not very common excelsior; c. Acer platanoides;= d. Ulmus spp.; e. Tilia cordata. even in DBH classes 1 to 3, Tilia reached values between DBH classes as in Fig. 21.

Acta Phytogeogr. Suec. 80 7 4 M. Diekmann

100 Quercu.• robur a 40 and 50 % (rs= - 0.474, p < 0.001). 90 Eutrophic elm-ash forests on the mainland ( 18 stands, g 80 70 Fig. 25a-e): Quercus again showed a significant, though

> 1 ea not continuous increase of MRD values towards larger 50 g? size classes (rs = 0.370, p < 0.001, N = 99), but it was less i 40 common than in eutrophic elm-ash forests on bland. � 30 was the most abundant species. It reached � 20 Ulmus glabra 10 MRD values between 40 and 55 % in the smaller classes 0 ..L.----�--.---- and significantly lower values (15 - 25 %) in the larger 3 4 5 6 DBH classes Crs= - 0.323, p < 0.01). Acer (rs =-0.170, 100 Fraxinus excelsinr n.s.), 0. 105, n.s.) and 0.074, 90 b Fraxinus (rs= - Tilia (rs =- n.s.) did not show clear trends. g 80 70 Mesotrophic forests on the mainland ( 17 stands, Fig. -l'5-� 60 26a-e): Quercus and Acer showed the same pattern as in � 50 the mesotrophic forests on bland, i.e. a significant posi­ 40 � tive correlation between MRD values and DBH class for � 30 � 20 the former (rs= 0.488,p < 0.001, N = 91) and a significant 10 negative correlation for the latter (rs= - 0.555,p < 0.001). 0 _,____..,..___j For Ti lia, there was only a weak negative correlation 3 4 5 6 (rs= - 0.209, p < 0.05), caused by its low MRD values in 1 00 A cer plalanuides the two largest DBH classes. and both 90 c Fraxinus Ulmus had very low overall abundance, but showed slightly g 80 70 falling curves (rs= - 0.408, p < 0.001) and Crs =-0.281, >- � 60 p < 0.01). � 50 Oligotrophic oak forests on the mainland ( 4 stands, � 40 Fig. 27a-f): Quercus was dominating in all strata. How­ � 30 � 20 ever, it showed a slight, but significant decrease of MRD 10 values towards smaller DBH classes (rs = 0.547,p < 0.01, 0 • 21). and were compara­ 3 4 5 N = Fraxinus, Acer, Tilia Ulmus I 6I tively rare and, except the latter species, absent from the 100 Ulmusspp. two largest DBH classes. The regeneration of Pieea abies 90 d was remarkable (rs=-0.619,p < 0.01). In the studied oak g 80 - forest stand on bland (not shown), Quercus had relative 70 >- 1 60 densities between 90 and 100 % in all DBH classes. g? 50 � 40 � 30 � 20 5.4 Discussion 10 Relative basal area 2 3 4 5 6

100 cordata Tilia It has been shown that the relative basal areas of some tree 90 e species differed between the main forest types, and that g 80 70 they were correlated with the vegetational composition of � 60 field and bottom layers. It is probable that these tree � 50 species respond to the same environmental factors as the

� 40 non-woody species, and, hence, that their basal areas � 30 � 20 partly depend on edaphic (pH, nutrients, etc.) factors and/ or climatic conditions (temperature, precipitation, etc.). 10 Another possible explanation is that many non-woody 0 2 3 4 5 6 species respond positively or negatively to the dominance DBH class of certain tree species which create a favourable or unfa­ Fig. 24. Frequency distribution of mean relative density over vourable environment, respectively. However, the results DBH class for five deciduous hardwood tree species in mesa­ trophic mixed deciduous forests on Gland (N 13). a. Quercus justify the separate analysis of the main forest communi­ robur; b. Fraxinus excelsior; c. Acer platanoides;= d. Ulmus spp.; ties. e. Tilia cordata. DBH classes as in Fig. 21.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 75

100 Quercus 100 Fra.rinu.• spp. a e:rcelsior b 90 90

80 80 g g 70 70

>- >- � 60 � 60 50 50

:;::;(!) (!) :> 40 �:> 40 �"' � 30 30 c c �"' 20 �al 20 10 10

0 0 2 3 4 5 6 3 4 6

100 100 Ulmus Acer plaJanoides glabra 90 c 90 d 80 80 g g 70 70

>- l 60 � 60 50 50 � (!) �:> � 40 40 � � 30 30 c c al �"' � 20 20 10 10

0 0 2 3 4 5 6 3 4 5 6

DBH 100 Tilia cordala class e 90

80 g 70

>- � 60 50 Fig. 25. Frequency distribution of mean rela- :;::;(!) :> 40 tive density over DBH class for five decidu- �"' 30 ous hardwood tree species in eutrophic elm- c �al 20 ash forests on the mainland (N 18). 10 a. Quercus spp.; b. Fraxinus excelsior; c.= Acer 0 • ' ... ' • platanoides; d. Ulmus glabra; e. Tilia cordata. 2 3 I4 5 6 DBH DBH classes as in Fig. 21. class

Tree density in different DBH classes 6 and partly due to the deviating behaviour of Quercus, which usually showed an increasing absolute (and rela­ The diameter distributions of the total of tree stems, as tive) density with increasing diameter. Malmer et al. well as those shown by most single tree species, were ( 1978) described diameter distributions of the most im­ usually of a falling logarithmic type, i.e. there was a high portant tree species for Dalby Soderskog. Here, Quercus number of young, small-sized individuals and rapidly showed the same pattern of high densities in larger DBH decreasing numbers of trees with increasing size. Similar classes, whereas the other species more or less displayed patterns have been described by e.g. Jones (1945) and the typical falling logarithmic curve. Quercus is often Leibundgut ( 1982) to occur in certain developmental stages represented by a fairly large number of very old and of virgin forests. For primeval boreal forests in northern impressive trees (see also Ryberg 197 1), particularly in Europe, such diameter distributions have been demon­ oak-hazel forests. The poor regeneration of all tree spe­ strated for Betula pubescens, Picea abies and Sorbus cies in this community probably results from low light aucuparia by Hytteborn et al. (1987) and Hofgaard levels in the interior of the stands, created by Corylus (1993a,b). However, this patternis also well known from avellana forming dense sub-canopies below the tree canopy managed 'selection forests' (German 'Plenterwald') in (see Chapter 4.3). In the Mittlandsskogen area on bland, which isolated scattered stems are felled and trees of all Corylussometimes grows in pure shrub-woods (so-called ages are present. Selective logging of large trees has 'hasslen' ), similarly dark and unsuitable for tree regenera­ probably taken place in most of the stands studied up to tion. It will probably take a long time for oak-hazel the recent past. The slightly larger number of trees in forests, as well as for the hazel-groves mentioned, to DBH class 6, compared with class 5, was conspicuous in achieve a more natural structure and species composition. all forest types and is partly due to the larger size of class

Acta Phytogeogr. Suec. 80 76 M. Diekmann

1 00 Que spp. a Successional stability 90 reus g 80 70 >- The value for compositional change (CC) summarizes the � 60 � 50 amount of differences in tree species composition be­

� 40 tween overstorey and understorey of a forest in one figure, � 30 � 20 which is thought to indicate the compositional instability. 10 A comparison with CC values calculated for a large number 0 3 5 6 of stands in Wisconsin (Goff 1968) shows much higher

class figures for the Swedish forests, both with respect to the DBH overall average and maximum values. The same result 1 00 Fraxinus excelsior 90 b was obtained from calculating the percentage similarity g 80 between overstorey and understorey as used by Fralish 70 ( 1988). The majority of the stands are compositionally � 60 � 50 instable, i.e. the species composition of trees - and subse­

� quently that of other layers too - will probably change 3400 �c 20 towards more mature and site-specific, natural condi­ :::! 10 tions. It is probable that the forests retain structural and compositional characteristics of the former cultural land­

DBH class scape. Even if the overstorey may partly represent site­ specific species belonging to the pioneer tree generation, 00 1 Acer piDlanoides c 90 it mainly reflects the composition of former wooded pas­

g 80 tures and meadows from which the forests have origi­ � 70 nated. The human impact on deciduous forest structure -2l 60 and species composition in Sweden has already been � 50 � 40 emphasized by, e.g. Almquist ( 1929) and Lindquist ( 1934). 30 � 20 A very low CC value was found for Viisterstads lund � 10 on bland. This stand is dominated throughout all layers by Ulmus minor and represents one of the most unaf­ 3 4 5 6

D8H class fected, 'primeval' -looking deciduous forests in Sweden (Selander 1955). However, a low CC value does not 100 Ulmus glabra d always indicate a forest's naturalness. It may also result 90 g 80 from edaphic or historical conditions allowing only one or 70 a few tree species to establish and survive, as is the case � 60 for the oak forest on acid soil on bland. Besides, it can be � 50 . 40 m the result of an extreme 'unnatural' species composition � 30 � 20 due to human impact, as in some oak-hazel forests. Here, hardly any regeneration of trees occurs, which causes a

5 6 slow compositional change.

DBH class

Relative densities of species and successional trends 1 00 Ti/ia cordata e 90 The figures for the mean relative densities of particular g 80 70 species in different DBH classes (Figs. 23-27) show some

i 60 general trends. The pattern is most distinct for Quercus 50 � 40 which increases in relative density with increasing diam­ � 30 eter. Acer exhibits the opposite pattern of decreasing � 20 abundance in larger DBH classes. Interpreting these pat­ 10 terns as future replacement series, Acer will increase in 5 6 importance while Que reus will decrease, if not disappear DBH class from the studied forests. This is well in accordance with the forest-historical background given in the introduction Fig. 26. Frequency distribution of mean relative density over in Chapter While Quercus generally was one of the DBH class for five deciduous hardwood tree species in mesa­ 1. trophic mixed deciduous forests on the mainland (N :::: 17). most favoured tree species in wooded meadows and pas­ a. Quercus spp.; b. Fraxinus excelsior; c. Acer platanoides; tures, Acer was disfavoured among the nemoral decidu­ d. Ulmus glabra; e. Tilia cordata. DBH classes as in Fig. 21. ous tree species. The present overrepresentation and fu-

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 77

100 Quen:'tiSspp. a 100 Fraxinust:rcel�or b 90 90 g 80 g 80 70 70 >- � 50 i eo - "' 50 � 50 ';i;> 40 <0 40 £i '-' 30 c 'c 30 Ul " ::?:0 20 ::>0 20

1 0 10

0 0 -·--- 2 3 6 6 2-,------,----·---,-·,--,- 3 � 6 6

100 A plalanoide�· c 100 lilmu�gltzhm d cer 90 98 g 80 g 80 70 70 ""' >- � 50 � 60 "' bO - "' 50 > �-=-"' '§ 1i! 40 � 40 30 30 �"'L �"'c ?0 20 10 10 I • J :.1 ""'"" I I 0 I ------,--- ? 3 4 5 6 � 3 4 5 6

Tilia 1 00 - cordaJa e 100 Picea abies f 90 90 g AO g 80 7!) 70 > � 60 � 60 50 "' 50 "" 4() 1\0 �"" e; 30 '- 30 "'c "' � 20 2U> 20

10 10 -�- 0 0 - Ill .I I I I ---,-- r·------,- I 1 2 3 4 5 6 2 3 4 5 6

DBH closs DBH �:ass Fig. 27. Frequency distribution of mean relative density over DBH class for six different tree species in oligotrophic oak forests on the mainland (N 4). a. Quercus spp.; b. Fraxinus excelsior; c. Acer platanoides; d. Ulmus glabra; e. Tilia cordata; f. Picea abies. DBH classes as in F=i g. 21. ture decline of Quercus in northerndeciduous forests has to survive on other sites with extreme edaphic conditions, been demonstrated or suggested by many authors, e.g. as on thin, dry calcareous soils (locally observed on Lindquist (1934, 1938), Sjors (1965b), Ryberg (1971), bland) and on periodically inundated river banks (e.g. at Klotzli (1975a), Malmer et al. (1978) and Brunet (1991). the river DaHilven, Almquist 1929; Skoglund 1989). The However, oak takes advantage of its wide ecological wide distribution and present underrepresentation of Acer amplitude compared with other deciduous hardwood trees. has not been paid attention to, since it is uncommon as an On very acid and nutrient-poor soils, as in the studied overstorey tree and only rarely has achieved dominance. oligotrophic oak forest stand on Oland, Quercus (robur) Ti lia shows a similar, but weaker pattern as compared is the dominating tree species in all layers and will prob­ with Acer, with decreasing mean relative densities to­ ably not be replaced by other species. An excellent regen­ wards larger DBH classes, except for the mainland eration can also be observed in oligotrophic oak forests in eutrophic elm-ash forests. This positive successional trend the suboceanic parts of Sweden (Halland and BohusHin), is most distinct on bland, which was also suggested by where Quercus (petraea) also re-invades many defor­ Sterner (1986). An increase of Tilia and a decline of ested areas. Besides, it probably would be more common Quercus have also been reported from the Bialowieza in many coniferous, especially spruce, forests where it has forest situated in the Boreo-nemoral zone of Poland (Pigott often been removed during the past, but where it can be 197 5; Faliri.ski 1986). Here, a reduced grazing pressure on observed to regenerate well (Lindquist 1934; Sjors 1965b; the forest has facilitated the successful regeneration of Ryberg 1971; KlOtzli 1975a). Moreover, Quercus is able Tilia.

Acta Phytogeogr. Suec. 80 78 M. Diekmann

Regarding elm-ash forests, the responses of Fraxinus fire for the successful establishment of Pinus and Picea excelsior and Ulmus spp. are particularly interesting. On and suggest that the deciduous forest has possibly only the mainland, Fraxinus does not show a clear trend, survived where no comparable disturbance has occurred. whereas, on bland, it has much higher mean relative According to Bradshaw & Hannon, the conifers are best densities in the three larger DBH classes. In both regions, suited to the present climate, whereas the deciduous for­ Ulmus shows the opposite trend with comparatively higher ests just have survived in a non-equilibrium state. How­ values for the three smaller DBH classes. Thus, Ulmus ever, at comparatively warm and base-rich sites, the com­ will probably gain dominance at the expense of Fraxinus. petitive ability of Picea becomes weakened due to mor­ This succession has been observed in Dalby Soderskog by phological and physio-pathological reasons (KlOtzli Lindquist (1938), Malmer et al. (1978) and S. Persson 1975a), and consequently it becomes outshaded by the (1980). Leemans (1992), in a computer simulation study faster growing deciduous tree species. Only in oligotroph­ of Dalby Soderskog, predicted the same trend for the ic forests, can it take advantage of its greater tolerance to future. Also in other eutrophic elm-ash forests in Skane, nutrient-poor conditions. In the oligotrophic oak forests this succession will probably take place (Brunet 1991), on the mainland, Picea in fact shows a positive succes­ although attacks of the Dutch elm disease may reverse this sional trend.For two oligotrophicconif erous forests in the trend. Hytteborn(1 986) demonstrated the partial replace­ Boreo-nemoral zone, Bradshaw & Hannon (1992) and ment of Fraxinus by Ulmus for the Vardsatra forest in the Bradshaw (1993) showed an increase of Picea during the vicinity of Uppsala. The present overrepresentation of last 200 years due to a reduction of the fire frequency. Fraxinusis partly due to the patronage of this species for Other tree species, such as Betula spp. and Populus leaf fodder in former wooded meadows. On the contrary, tremula, can be regarded as light-demanding pioneer spe­ Ulmus is likely to regain a position which the species cies, not able to survive under mature stands in the long might have had in the natural landscape. Fraxinus is, run, although quite old, large birch and aspen trees do however, a rapid colonizer and can invade abandoned occur. Corylus avellana, which was not included in the pastures much faster than Ulmus. Thus, many old ashes structural analysis, has been considered to be a strong may represent survivors from a pioneer tree generation. In competitor in deciduous forests (Lindquist 1938; Ryberg mesotrophic forests, both Fraxinus and Ulmus are not 1971). For Dalby Soderskog, Lindquist predicted the fu­ very common, but show slightly positive successional ture formation of a climax hazel shrub forest in parts of the trends. In both eutrophic elm-ash forests and mesotrophic stand. However, Malmer et al . (1978) demonstrated that, forests, Fraxinus is much more abundant as seedling and in the meantime, Corylus has markedly decreased in this small sapling in the field layer than Ulmus, but suffers forest. Probably hazel is able to achieve dominance in an from a high mortality (cf. Hytteborn 1986). This may be early successional stage, e.g. after abandonment of a due to the sensitivity of this species to grazing (Lindquist wooded meadow, but is later replaced by more tolerant, 1934; Ryberg 197 1), today mainly by roedeer, but results shade-bearing trees. A high abundance of shrub species also from a change in the physiological demands during during a certain phase of succession has also been men­ its ontogenetic life cycle (Gatsuk et al. 1980): while the tioned by Sjors (1967). juvenile plant has a high shade-tolerance, the light re­ In contrast to many other forest succession studies, data quirement increases sharply at the immature state. Thus, of comparatively small detail have been sampled from a this species has a highrate of turnoverof young individu­ large number of stands, instead of data of great detail from als. While, under shady conditions, most young ash plants a few stands. The results, outlining the secondary succes­ never grow up to the shrub layer or canopy, the population sional trends of deciduous forests and its tree species, are can be replenished when conditions are more favourable. confirmed by the few long-term permanent plot studies Prentice & Helmisaari ( 1991) suggested an increasing available for northern deciduous forests. They also corre­ shade intolerance with increasing age for both Fraxinus spond to the trends which could be expected in view of the excelsior and Acer platanoides, but, in this study, this was forest history and its impact on the tree species composi­ not observed for the latter species. tion of the forests. However, the current knowledge about Picea abies is rare in both eutrophic elm-ash forests the deciduous forest vegetation is not sufficient to make and mesotrophic forests, with respect to both older indi­ precise predictions of the natural vegetation of the studied viduals and saplings. This may be partly due to former sites, i.e. the vegetational composition of 'climax' commu­ human influence aiming at an exclusion of spruce from nities. Besides, the application of the climax concept is at wooded meadows. Bradshaw & Hannon (1989) give an­ least questionable in an area where the environmental other possible explanation in a study from two edaphically conditions may change quickly due to anthropogenic proc­ similar islands in Lake Malaren, one with dominating esses such as exploitation, global warming and acidifica­ coniferous forest (Picea and Pinus), the other one with tion. The predictions given in this Chapter are thus limited dominating deciduous forest (Tilia, Ulmus, Quercus). to the secondary succession currently in progress in de­ They emphasize the importance of disturbance through ciduous forests of the Boreo-nemoral zone.

Acta Phytogeogr. Suec. 80 6 Environmental studies

6.1 Introduction seedling survival, flowering, fruit ripening, etc. (Larcher 1980; Woodward 1987). There is often good agreement The present structure and species composition of decidu­ between thermal thresholds and major vegetation zones ous forests is influencedby many different factors which (Waiter 1979; Schroder 1983; Woodward 1987) or distri­ can be classified in abiotic, biotic and historical factors, butional borders of single taxa, as has been shown for which impact on the vegetation and dynamics of decidu­ many species in northernEurope (e.g. Iversen 1944; Dahl ous forests has already been emphasized in Chapters 4 and 1951, 1992; Skre 1979). In northernEurope, temperatures 5. Here, abiotic components will be treated in more detail. limit thermophilous species towards the north and higher Special emphasis will be given to the climatic conditions, elevations because of their requirement of a sufficiently as only a few climatic studies of northern deciduous warm and long growing season, i.e. of an effective tem­ forests have been made, compared with studies on the perature sum (Tuhkanen 1980). It has been assumed that significance of edaphic factors. Therefore, this introduc­ temperatures may restrict such species by their effect on tion is mainly concerned with the relationship between ATP production through dark respiration (e.g. Dahl & climate and vegetation. Mork 1959; Skre 1979; Dahl 1992; Holten 1993). Plant The pattern of vegetation zones is primarily deter­ species distribution may thus be explained in terms of a mined by climatic factors, such as temperature and pre­ respiration sum. On the other hand, high late autumn and cipitation. Climate has a direct influence on plants, but winter temperatures may limit boreal and continental also an indirect impact via soil properties which partly species towards the south and coastal areas, e.g. Picea depend on the climate. abies; the physiological mechanisms responsible for this Three levels of climate are generally distinguished, phenomenon are still unclear (Dahl 1992), but may in macroclimate, mesoclimate and microclimate (Stoutjesdij k some way be related to winter temperatures (Holten 1993). & Barkman 1992). The macroclimate is defined as the With respect to extreme temperatures, low winter tem­ (average) weather situation over a long period independ­ peratures are the limiting factor towards colder regions ent of topography, soil and vegetation. Mesoclimate (or for many frost-sensitive plants (Woodward 1987). There topoclimate) is a local variant of the macroclimate caused are many examples of northeastern distribution limits of by topography, human impact or structure of the vegeta­ oceanic plants corresponding to isotherms of the coldest tion (Stoutjesdijk & Barkman 1992). Finally, the climatic month (e.g. Dahl 1992). Occasionally, a combination of conditions in the lowermost 2 m of the atmosphere and in climatic factors gives a better correlation with plant distri­ the upper soil are called microclimate. Microclimate is bution, e.g. the mean monthly temperatures for both Janu­ considerably influenced by the nature of the vegetation. ary and July (Iversen 1944; Hintikka 1963). Species are Microclimatic variables which may deviate from the not necessarily excluded from an area only due to purely macroclimatic situation are: light intensity, extreme tem­ physiological mechanisms, but also by competition from peratures, average temperature, air humidity, evapora­ other plant species (Woodward 1987). In general, forests tion, wind velocity, etc. (Geiger 1961; Stoutjesdijk & reduce temperature fluctuations(S toutjesdijk & Barkman Barkman 1992). As the climatic conditions in a tall forest 1992). This moderating effect is most pronounced at the are a matter of both mesoclimate and microclimate, from forest floor, as there is a vertical gradient below the now onwards the term 'forest climate' will be used. Even canopy (e.g. Heckert 1959; Hytteborn 1975; Stoutjesdijk if macroclimatic isolines (e.g. isotherms) are often well & Barkman 1992). Thus, especially plants in the field and reflectedin boundaries of the distribution areas of plants bottom layers are subjected to a temperature climate con­ and animals, it is the forest climate that is decisive and siderably deviating from the macroclimate. which, for example, governs the occurrence of southerly distributed taxa at favoured localities at northern lati­ Air humidity and evaporation tu des. Humidity is another limiting factor for plant distribution (Holten 1993). Air humidity and evaporation are interre­ Temperature lated with temperature: Evaporation depends on both air Both average and extreme temperatures control plant humidity and temperature but may in turn influence both distribution. Often, these temperature effects are crucial (Stoutjesdij k & Barkman 1992). Evaporation rate of soil for certain phases of the life cycle, such as seed germination, and water-bodies and transpiration rate of plants and

Acta Phytogeogr. Suec. 80 80 M. Diekmann

animals (together called evapotranspiration) are of great slope, the greater the difference between north and south ecological importance for all organisms; vascular plants slope. These climatic differences can be highly determi­ must compensate for the water losses through water up­ nable for the distribution of different forest types (e.g. take from the soil. A prolonged, high evaporation rate J. Parker 1952; Heikkinen 1991), and they also influence thus may imply a stress situation for many plant species. the phenology of plants. In general, the air within the forest is relatively moister The special 'southern' vegetation of slopes facing S than outside, due to evaporation and, during the day, and often also W or E in Scandinavia has already been lower temperatures. mentioned in Chapter 4. Climatological aspects as well as the flora and vegetation of steep hillsides in northern Sweden have been thoroughly described and discussed by Light climate J. Lundqvist (1968), who also gave a review on the For photosynthesis, green plants can use light with wave­ literature. Lundqvist particularly emphasized differences lengths between 400 and 700 nm, the so-called 'photo­ in the temperature climate. synthetically active radiation' (PAR) which makes up In this study, the forest climate and some physiographic about half of the energy provided by the total solar radia­ and edaphic factors of selected forest stands in the prov­ tion (Stoutjesdijk & Barkman 1992). Light reaching the ince of Uppland have been investigated in order to show vegetated surface of the earth consists of two compo­ differences between forest types and to relate these differ­ nents: direct light coming directly from the sun and dif­ ences to abiotic factors and the structure and species fuse light arriving from all other sources. Different effects composition of the forests. As in Chapter 5, mainly of light can be distinguished: light colour (spectral com­ mesotrophic mixed deciduous forests and eutrophic elm­ position), light quality (e.g. red/far red ratio), and light ash forests were considered, as these are the only decidu­ intensity. The latter is crucially important as all green ous forest types of wide distribution within Uppland. plant species have a 'compensation light intensity' where net photosynthesis is zero. For a sufficiently long time, light must exceed a plant's compensation point to allow it 6.2 Soil analysis and climatic measure­ to survive. As light passes through the forest canopy, not ments only its intensity but all its components are altered (Stoutjesdij k & Barkman 1992). General reviews of the Climatic measurements light climate of forests have been given by, e.g. Salisbury (1936), Tranquillini (1960), M.C. Anderson (1964a,b), For climatic measurements, nine forest stands with 17 Evans (1966), Eber (1971, 1972) and Stoutjesdijk & plots were selected in the vicinity of Uppsala, represent­ Barkman ( 1992). Most light measurements have been ing different community types. Several of the plots were concerned with light intensity, particularly with the rela­ situated close to Lake Malaren. Additionally, two conifer­ tive intensity of diffuse light transmitted through different ous forest plots were selected for comparative tempera­ kinds of forests under cloudy conditions, the so-called ture measurements. Temperature and evaporation were 'daylight factor' (Atkins et al. 1937). At an early date, registered on 29 days in 1992 (between 3.4. and 9.11.) and several attempts were made to correlate the spatial distri­ on 14 days in 1993 (between 26.5. and 27.9.). On some bution of plant species to definiteranges oflight intensities, occasions, soil samples for determination of moisture in deciduous forests, e.g. by Ellenberg ( 1939) and were collected. The plots were always visited in a fixed Blackman & Rutter ( 1946), in a wooded meadow by Sja rs order and with roughly the same time intervals between (1954). However, in deciduous forests, there is a strong visits at the different plots. seasonal variationin relative light intensity (M.C. Anderson Minimum and maximum temperatures were recorded 1964b; Coombe 1966; Pons 1983), and also the spatial with mercury Min/Max-thermometers. In each of the 17 variation within a forest stand is considerable, both verti­ plots, three thermometers were laid out at the soil surface. cally and horizontally (cf. Salisbury 1936; Evans 1966). To avoid exposure to the sun and therewith a large radia­ tion error (Stoutjesdijk & Barkman 1992), all thermom­ eters were placed in maximal shade close to a tree or shrub Aspect and inclination and covered with a thin litter layer. Temperatures were Slopes are subjected to a special climate affecting plants, recorded with 0.5 oc accuracy. For data analysis, an animals and soil development. Both aspect and inclina­ average value was calculated from the three recordings tion of the slope have significant effects (Geiger 1961). for minimum and maximum temperature, respectively. The differences between north and south slopes are well­ An average value was also calculated for the temperature known and concern solar radiation, temperatures, air hu­ range (difference between maximum and minimum tem­ midity, evaporation, etc. (Stoutjesdijk & Barkman 1992). perature). Only one thermometer was placed in each of Inclination intensifies the effect of aspect: the steeper the the pine forest plots.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 81

Evaporation was determined with Fiche-evaporim­ injection analysis, given as mg N (NH/ -N + NOr -N)/kg eters. These were hung up at a height of 2.4 m on the dry soil; northernside of a tree (in order to avoid direct sunshine) in - pH(H20) was determined by mixing 10 g dried sample the centre of the plot. Distilled water was used for filling with 25 g distilled water. pH was measured with a pHM83 and re-filling the tubes. Recordings had an accuracy of 1 Autocal pH meter, Radiometer, Copenhagen; mm. The daily evaporation for a certain time period was - pH(KCl) was determined as above, but with a 0.2 M calculated as the absolute evaporation divided by the KCl-solution instead of distilled water; number of days in that time period. - soil moisture was determined gravimetrically as men­ Comparative light measurements in the selected forest tioned above by weighing the fresh and dry samples stands were carried out in July and August 1991 and (water contents/fresh soil sample, in %). August 1993, as the transmission of the fully developed Additional measurements, especially light measure­ canopy is more or less constant during these months (cf. ments, were carried out in five forest stands on Gland in M.C. Anderson 1964b; Eber 1971; Pons 1983). Measure­ July 1991. In each stand, light was recorded at 1-hour ments were made of photosynthetically active radiation intervals during one day (8.00 a.m. - 5.00/6.00 p.m.) un­ (PAR) under overcast skies (diffuse light). Day time der more or less clear sky (direct and diffuse light). Here, varied between 8.00 a.m.- 4.00 p.m., since the relative a transect had only 20 subplots, and only one measure­ light intensity under cloudy conditions is rather constant ment was carried out at the height of the field layer. The during this time interval (Eber 1971), fairly independent hourly light measurements were complemented by regis­ of the light intensity in the open (Nageli 1940). Measure­ tration of air temperature [T(lOO)] and air humidity 1 m ments were made simultaneously inside and outside the above-ground, using an Assmann-psychrometer, and by forest, using two light meters (PAR 'Special' Sensors, measurements of the ground temperature [T(O)] and the SKP 210, Skye Instruments Ltd.), registrating quantum soil temperatures at 5 cm [T(-5)] and 15 cm [T(-15)] flux density in einstein/m2/s. The relative response of the depth, using mercury thermometers (three at each depth). instruments, i.e. the sensitivity to different wavelengths, corresponded to the absorption spectrum shown by the Data analysis plant pigments. Both sensors were regularly compared with each other to display the same light intensity; from For pairwise comparison of temperature and evaporation time to time, a zero adjustment was made. In each plot, a data of the forest plots, the Wilcoxon signed ranks test transect was established of 25 subplots with 1 m distance (Siegel & Castellan 1988) was used. Data for 1992 and from each other. Two measurements of PAR were made 1993 were analyzed separately. The data for 1992 were both at the height of the field layer (60 cm) and at ground divided into two periods, (a) the time of full leaf develop­ level (5 cm), 30 cm to the left and the right of the centre, ment (20 observations), from the beginning of June until respectively. The sensor outside the forest was located at the end of September, and (b) the time before and afterfull some distance from the stand in order to receive full leaf development (nine observations). For pairwise com­ radiation. Both sensors were always kept in a horizontal parisons, only period (a) was considered. The Wilcoxon position. signed ranks test was also used for comparison of the soil moisture data. It was not possible to use the original recordings of the environmental factors named above as Soil analysis input for further analysis. Sums or means could not be Soil samples for chemical and physical analyses were calculated because of missing values and different time taken with a 200 cm3 (5 cm high) metal cylinder. In each intervals. Instead, each plot was assigned a score between plot, three mixed soil samples were collected; each sam­ 1 and 8 for each factor on the basis of the pairwise ple contained five cores. Samples for determination of comparisons counted as + or -. The score 8 was assigned soil moisture were collected on nine occasions during to 16 +10 - and 15 +11 - (meaning that a stand had higher 1991, 1992 and 1993, always after at least one day without values of the measured factor in all 16 comparisons or in precipitation. Here, only one sample (containing fivecores) 15 out of 16 comparisons, respectively), the score 7 was was taken and kept in a waterproof plastic bag until assigned to 14 +/2 - and 13 +13 -, etc., down to score 1 for analysis. Generally, methods for soil analyses followed 2 +1 14 - and lower. Scores for 1992 and 1993 were Balsberg-Pahlsson (1990). averaged, and the resulting mean was used as ordination - dry bulk density was determined as weight/volume in input. In order to compare the forest climatic conditions g/cm3; with the macroclimate of the same time period, data from - organic matter was analyzed as loss on ignition by the Ultuna Meteorological Station in Uppsala were used as ashing the samples at 600 oc in a muffle furnace; reference (provided by the Swedish University of Agricul­ - soil mineral N was extracted by shaking 20 g soil with tural Sciences). For chemical and physical soil properties, 100 ml 0.2 M KCl for two hours and determined by flow the values for the three mixed samples were averaged.

Acta Phytogeogr. Suec. 80 82 M. Diekmann

Light data were analyzed in the following way. The - topographic factors: inclination (ORAD), heat index relative light intensity (RLI, in %) was calculated as the according to K.C. Parker (1988) (HEATIND); ratio of the absolute light intensity (ALl) inside the forest - edaphic factors: organic matter contents (ORGANMAT), and the total light intensity (TLI) outside the forest. Occa­ mineral nitrogen contents (NITROGEN), pH(H20) sionally, there was an influence of direct sunshine during (pHWATER), pH(KCl) (pHKCl), dry bulk density (DEN­ a series of measurements; then, values were excluded SITY), soil moisture (MOISTURE); before data analysis. For the bland-plots, RLI and ALl - climatic factors: RLI in height of the field layer under a clear sky were calculated in two ways: once (LIGHTFLD), evaporation (EVAPOR), minimum tem­ including all measurements [RLI(incl) and ALI(incl)], perature (MINITEMP), maximum temperature (MAXI­ once excluding sunflecks from the analysis [done by TEMP), temperature range (DIFFTEMP), canopy cover excluding all values exceeding a value arbitrarily defined (CANOPCOV), exposure to wind (EXPOSURE). as RLIunder diffuse light + 2 SD for each particular stand, RLI(excl) and ALI(excl)]. The canopy cover was calculated as the sum of the 6.3 Results cover percentages of the upper tree layer (T1), lower tree layer (T2) and shrub layer (S). The tree cover of single A short survey ofthe selected forest plots is given in Table species was calculated as the sum of the cover-abundance 26. Six plots (from five stands) represented mesotrophic values of T1, T2 and S. Exposure was estimated subjec­ forests, belonging to the Que reus robur-Tilia cordata tively in three degrees (++, +, -). community, Geranium sylvaticum sub-community. The To relate the environmental measurements to the vari­ three forms of this sub-community were represented each ation in vegetation, the ordination programs Correspond­ by two plots. All were located on slopes of different ence Analysis (CA) and Canonical Correspondence Analy­ inclination, and mostly with a western aspect. Ten plots sis (CCA) of the program package CANOCO (ter Braak (from five stands) represented eutrophic elm-ash forests, 1987, 1990) were used. The original releves of the forest located at level ground or gentle slopes of maximally 5° plots were used and arranged in a new releve table. Only inclination. Only one plot (Vardsatra-2) represented species in field and bottom layers were considered for the eutrophic alder-ash forests of the Fraxinus excelsior­ ordination. No new cluster analysis was carried out, but Prunus padus community. This plot was treated together the classification as described in Chapter 4 was adopted. with the eutrophic elm-ash forests. A releve table of the Environmental variables are (abbreviations used in the plots is given in Table 27. The two coniferous forests tables given in brackets): selected for comparative temperature measurements were

Table 26. Description of the forest stands selected for environmental measurements. Communities: 1. Quereus robur-Tilia cordata community, Geranium sylvaticum sub-community; a. Geranium robertianum form; b. Deschampsiaflexuosa form; c. Lathyrus vernus form. 2. Ulmus glabra-Fraxinus excelsior community, Gagea lutea sub-community. 3. Fraxinus excelsior-Prunus padus community, Hepatica nobilis sub-community. 4. Eu-Piceetum.

Plot Forest stand/ Geographic Height Aspect Inclination Wind Community -plot location (longitude/latitude) (m a.s.l.) ( ) exposure 0 1 Valloxen-1 17°50'15"/59°44'45" 25 5-10 lb w ++ 2 Valloxen-2 1 r50'15"/59°44'45" 25 E 5-10 1b 3 Ragnhildsvik 17°38'45"/59°42' 15" 10 10 w ++ le 4 Skokloster- 1 17°38'00"/59°42'45" 5 E 5 2 5 Skokloster-2 17°38'00"/59°42'45" 10 2 + 6 Skokloster-3 17°38'00"/59°42'45" 10 2 ++ 7 Skokloster-4 17°38'00"/59°42'45" 15 15 la w + 8 Hj alstaviken 17°24'30"/59°39'30" 10 10-20 wsw + le 9 Parnassen- 1 17°23'15"/59°39'30" 5 SW 5 2 ++ 10 Parnassen-2 17°23'30"/59°39'30" 5 SE 5 2 11 Djurgarden 17°22'15"/59°39'30" 10 2 12 Vik-1 17°28'45"/59°44'30" 15 2 + 13 Vik-2 17°28'30"/59°44'15" 15 2 14 V ardsatra- 1 17°38'00"/59°47'15" 5 SW 3-5 2 + 15 Vardsatra-2 17°38'00"/59°47'15" 5 w 3-5 3 16 Vardsatra-3 17°37'45"/59°47'30" 5 w 1-2 2 17 Kungshamn-Morga 17°39'30"/59°47'00" 10 SE 25 la 18 Pine forest- 1 17°40'30"/59°42'30" 25 4 ++ 19 Pine forest-2 17°33'15"/59°42'30" 20 4 ++

Meteorological Station (Ultuna) 17°39'/59°49' 15 ?

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 83

characterized by a fairly open upper tree layer composed ciduous forest plots were significantly lower during the of Pinus sylvestris and by Picea abies in the sub-canopy. main growing season (p < 0.01). However, this pattern is They can be assigned to the association Eu-Piceetum reversed in early spring and late autumn, as shown by Fig. (Kielland-Lund 1971). 28b. Maximum temperatures were also considerably lower than the reference values from Ultuna (determined at 1.5 m height), except for some higher values in April. Temperature The highest temperature minima were shown by plot 7 The differences in maximum temperature between plots (Skokloster-4), plot 14 (Vardsatra-1) and plot 10 (Parnas­ were not the same in 1992 and 1993. For 1992, no clear sen-2). For 1993, also plot 17 (Kungshamn-Morga) had differences existed between mesotrophic and eutrophic high values. Particularly low minima were shown by plot forests, and both the highest and lowest values were 1 (Valloxen- 1), plots 4 and 5 (Skokloster- 1 and -2) and shown by elm-ash forests. For 1993, the patternwas very plot 9 (Parnassen-1), for 1992 also by plot 2 (Valloxen-2). different: the highest maximum temperatures were shown Thus, in both years, there were no clear differences be­ by mesotrophic forests, except for plot 7 (Sko.kloster-4) tween mesotrophic and eutrophic forests. In Fig. 29a, the with fairly low values. The course of maximum tempera­ course of minimum temperatures in 1992 is given for the tures in 1992 for three selected plots is given in Fig. 28a. selected plots. In both years, plot 9 had significantly lower Plot 9 (Parnassen-1) had significantlyhigher values than values than plots 7 and 10 (p < 0.01), which were not plot 7 (Sko.kloster-4) and plot 10 (Parnassen-2), both in significantly different from each other. In all deciduous 1992 and in 1993 (p < 0.01). Plots 7 and 10 did not forest plots except plot 2 (Valloxen-2), the minima for significantly differ from each other. Compared with the 1992 were significantlyhigher than in the two pine forest two pine forest plots, maximum temperatures of all de- plots (p < 0.05, most plots p < 0.01). Again, this trend

25 a 15 a

0 20 0 10

0 \ ! -� I --� •

Pa nass 1 ·>i'C - - - r en- - - Parnassen- 1 - Parnassen-2 * - - - Parnassen-2 10 +-----�----�----�----� • Skokloster-4 • - o - 0 O- Skokloster-4 A s A s 1992 1992

40 b ?.0 b

30 10

e:> 20 0 2�

Ea. � 10 - 10

• UltUrlB * --- - Pir1e forest- 1 * ··- Ultur1a • - -- Pine forest- 1 Valloxen- 1 • - o -- · o- Valloxen- 1 A M J A S 0 N A M A S 0 N 1992 1992

Fig. 28. Course of maximum temperatures in 1992 for selected Fig. 29. Course of minimum temperatures in 1992 for selected plots. a. Skokloster-4 (plot 7), Pamassen-1 (plot 9) and Pamassen- plots. a. Skokloster-4 (plot 7), Pamassen-1 (plot 9) and Pamassen- 2 (plot 10), from the beginning of June until the end of Septem­ 2 (plot 10); b. Valloxen- 1 (plot 1), pine forest- 1 and Ultuna. ber; b. Valloxen-1 (plot 1 ), pine forest-1 and Ultuna, from the Time periods as in Fig. 28. beginning of April until mid-November.

Acta Phytogeogr. Suec. 80 84 M. Diekmann

Table 27. Releve table of selected forest plots. Running numbers Elymus caninus ..22 .31. .4...... Crataegus spp . ..2 ...... 2.2 .12 correspond to the plot numbers in Table 26. (See Table 3.) Polygona tum odoratum 1.2 ..... 3 ...... 3 Lathyrus vernus ..31 23 ...... Rubus saxat ilis ..31 ...2 ....3 .... n i Ru n ng number 12345678911111111 Aegopodium podagraria ...3 ....666 ...... 01234567 Anthriscus sylvestris .....2 ...1.1 1 .. . Cover % Tl (xlOl 34466763757346557 Dactylis glomerata 33 ....1 ...... % T2 (XlO) 2262511136616545- Lathyrus montanus 33 .....3 ...... % s (XlO) 42243458122763344 Veronica chamaedrys 21 .....2 ...... % F (X10) 6675253 4888777686 Deschampsia flexuosa 43 ...... 2 Rubus idaeus 23 ...... 2 % B -1-1111-155115221 Calamagrost is arundinacea 2 ...... 3 ....2 .... 0 0 Milium effusum ..4 .1.... 2 ...... Ranunculus cassubicus ...... 444 ...... T1 Lilium martagon ...... 444 ...... Ulmus glabra . . 4644 ..55443 5454 Carex montana ...... 112 ... . Quercus robur 553 ..3. 4.4 344 ...6 Corydalis intermedia . . 42 Fraxinus excelsior ...344 ..546 . 5434 ...... 1 Polygonatum mu l tiflorum ...... 543 . Acer platanoides . 3.355 .443 ...... 3 Ribes uva-crispa ...... 1.1 2 Betula pendula 33 ...... 3 ...... Dryopteris carthusiana 11 ...... Tilia cordata . . 5 ...6 ...... Luzula pilosa . Alnus glutinosa ...... 45 .. 11 ...... Campanula rotundifolia 1 ...... 1 Campanu la persicifolia . 1 ...... T2 1 Ranunculus repens ...2 ...... Acer platanoides 34534 .33.34343 .4. .1 Polypodium vulgare . 2 ....1 ...... Ulmus glabra 4 .. 453 .. 46533554 . Rosa spp . . 2 ...... Fraxinus excelsior . . 34 .33.344 .43 .4. 1 ...... Tilia cordata ..2 ...4 ...... Quercus robur 33..... 3 ...... Hieracium murorum ...... Sorbu s aucuparia 44 ...... 2 ...... 1 4 Anemone ranunculoides . Tilia cordata . . 53 ..3 ...... 44 ...... Deschampsia cespitosa ...... 1 ..1 . Prunus padus . 3 ...... 4. Ma ianthemum bi fol ium ...... 24 .

.s. Corylus avellana 54344446 .44643 .4. .la Eurhynchium hians ...4241 . 44412434 . Fraxinus excelsior 2.3 .. 4532 .3443234 Brachythecium rutabulum ...2 ....1121 2 . 111 Ulmus glabra 4 ..4434 . 242 . 44433 Plagiomn ium undulatum ...2 ...... 1. .2.2. Acer platanoides 3343 .2.4 ..2 ..32 .4 Cirr iphyl lum piliferum ...... 1 ...2. 3. Sorbu s aucuparia 32 .2.322 ....23 .24 Ribes alpinum .2222 .232 ..23 ...3 Additional species occurring only once : Prunus padus 22 ...... 344 ..344 . Lonicera xy losteum ..242242 ....42 .. (T1 ) Fagus sylvatica 9:3, Pinus sylvestris 17 :4, Sorbus Tilia cordata ..22 ..3 ..2 ...... aucuparia 17 :4 (T2 ) Fagus sylvatica Picea abies Ribes uva-crispa .....2 ...... 2. 22 . 9:4, 2:3, ( S) Amelanchier confusa 1: 4, Betula pendula 1:2, Quercus robur . 3 .....2 ...... Juniperus communis 1:2, Prunus avium 17 :4 (F) Athyrium Picea abies 2 ..2 filix-femina 15 :2, Betula pendula 17 :1, Carex muricata Crataegus spp ...... •.....3 2 agg . 8:1, Carex pilulifera 2:1, Daphne me zereum 14:1, ....•• •.•• Fragaria mo schata 10 :3, Galium aparine 9:1, Gal ium .E boreale 1: 1, Geranium robertianum 7:2, Hieracium umbe llatum Acer platanoides 34433423323332122 1:1, Hypericum maculaturn 1:1, Lapsana communis 4:1, Anemone nemorosa 4565555544466556. Lathraea squamaria 3:2, Melampyrum nemorosum 8:2, Ulmus glabra 1.24333 . 112123222 Neottia nidus-avis 13 :1, Oxalis acetosella 13 :4, Picea Hepatica nobilis 44544444 .425 .2.44 abies 1:1, Populus tremula 1:1, Primula veris 4:1, Geum urbanum .14223 ..331332244 Prunus avium 8:1, Pteridium aquilinum 2:4, Rubus Fraxinus excelsior ..44234 2 . 23444444 fruticosus agg . 17 :1, Solidago virgaurea 1:1, Trifolium Poa nemoralis 32322444 .1.13 ..34 medium 1:2, Vaccinium myrtillus 2:3, Vaccin ium vitis­ Paris quadrifolia 12343 ..1 .2344444 . idaea 1:4, Veronica officinalis 17 :2 (B) Eurhynchium Vicia sepium 333 .2222 . 1223 ..22 angus tirete 16 :4, Pleuroziu m schreberi 2:1, Ac taea spicata ..3233 .31.244314 . Rhytidiadelphus triquetrus 17 :1. Ranunculus auri comus agg . 1 .4. 343444444 .... Geranium sy lvaticum 432323 .1.11.2 ..2. Sorbus aucuparia .211.232 ..1222 .. 3 Ranunculus ficaria ..324 ...66 ..24222 Viola riviniana 4231 . .23 .....2 . 24 Melica nutans 443 . 12 . 2 ....2 ..13 Convallaria maj alis 3333 . . . 4 . 2.3 ..2. Prunus padus 2 .. 1. .1.1.. 1. .222 . Viburnum opulus ..122 .11 ....222 .. was reversed at the time before and after full leaf develop­ Corylus ave llana 222 ..131 ...... 2. ment (Fig. 29b ). The minimum temperatures of all studied Dryopteris filix-mas 24 ..2.3 3 ...... 14 . Taraxacum officinale agg . .111 . .1. .2.1. ..1. plots were much higher than those at Ultuna. Allium oleraceum .. 4423 ...... 2.4 2 In 1992, the widest temperature ranges were shown by Gagea lutea ..34 ....33 ...3. 32 Lonicera xy losteum ..1 . 1132 ....22 .. . plots 4 and 6 (Skokloster- 1 and -3), plot 9 (Parnassen-1) Mycelis muralis 3211 ...... 23 .... and plots 1 and 2 (Valloxen-1 and -2). The pattern was Ranunculus acris 231 . .2.1. ..1. ... . Carex digitata 132 ...33 ...... 3 similar in 1993, except for plot 5 (Skokloster-2) which Fragaria vesca 23 ...212 ...... 4 had high values instead of plot 4. Small temperature Ribes alp inum .21. .1.31. ... 2 .... . Viola mirabilis ..233 .....445 .... ranges were found in three eutrophic forests, namely plot Campanula trachelium ...32 ...... 111 .1. 10 (Parnassen-2), plot 11 (Djurgarden) and plot 14 Stachys sylvatica ...2 ....443 .. 32.. Quercus robur 32 ....11 ...... 3 (Vardsatra- 1), as well as in one mesotrophic forest, plot 7

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 85

(Skokloster-4 ). Fig. 30a shows the course of temperature eutrophic forest plots, plot 10 (Parnassen-2) and plot 11 ranges in 1992 for the selected plots. Again, both in 1992 (Djurgarden). Fig. 31 gives the course of daily evapora­ and in 1993, plot 9 had significantly higher values than tion for the selected plots for 1992 and 1993. Plot 7 had plots 7 and 10 (p < 0.01), which were not significantly the highest values, followed by plot 9 and plot 10. Differ­ different from each other. Compared with the pine forest ences are significantfor 1993 (p < 0.05), but not for 1992. plots, all deciduous forest plots showed significantly In both years, all three plots had a lower daily evaporation smaller temperature ranges during the time period of full than plot 1 (p < 0.01). For the selected plots, evaporation leaf development (p < 0.01 for all plots), but higher ranges is significantly positively correlated with maximum tem­ before and after (Fig. 30b ). The corresponding reference peratures in the plots (Table 28). Besides, there is a values from Ultuna were considerably higher. positive correlation with average temperature (but not with wind velocity) and a strong negative correlation with air humidity at the reference station in Ultuna. Evaporation

The results for evaporation do not differ very much be­ tween 1992 and 1993. In both years, three mesotrophic forest plots, namely plot 1 (Valloxen-1 ), plot 3 (Ragnhilds­ vik) and plot 7 (Skokloster-4), showed the highest values. a However, two other mesotrophic forest plots, plot 2 10 Parnassen- (Valloxen-2) and plot 17 (Kungshamn-Morga), had fairly * · 1 ·-- · · · · Parnassen-2 low values. The lowest evaporation was found for two • - o - - Skokloster-4 8 * - Valloxen- 1 >- CC! D '-. E 6 20 2 c a 0 � 0 4 0. 15 wCl) 6 ,, 0 2 �; • -;;; 10 (jj .. �a. 0 1- 5 A S 1992 1 b * - - - Parnassen- • - - Parnassen-2 0 Parnassen- 1 o - Skokloster-4 6 * -- Parnassen-2 A s • -

1992 o - - Skokloster-4 5 a 1 * - V lloxen- 30 b

Cl)>- 4 D '-. E 2 c 3 �0 0 0. � Cl) 2 > � w 1- 10

l!:· --- Ulluna • - - Pine forest- 1 0 +------,------,,------,------� o - Valloxen- 1 A s A M A S 0 N 1993 1992

Fig. 30. Course of temperature ranges in 1992 for selected plots. Fig. 31. Course of daily evaporation in four selected plots a. Skokloster-4 (plot 7), Pamassen-1 (plot 9) and Pamassen-2 [Skokloster-4 (plot 7), Pamassen-1 (plot 9), Pamassen-2 (plot (plot 10); b. Valloxen-1 (plot 1), pine forest- 1 and Ultuna. Time 1 0) and Valloxen-1 (plot 1)] from the beginning of June until the periods as in Fig. 28. end of September. a. 1992; b. 1993.

Acta Phytogeogr. Suec. 80 86 M. Diekmann

Table 28. Spearman rank correlation ( ) between daily evaporation and climatic variables for some selected plots. Maximum rs temperature determined in the plot, average temperature, wind velocity and air humidity taken from the Meteorological Station at Ultuna. p significance level; n.s. not significant. = = Plot Forest stand Maximum-T Average-T Wind velocity Air humidity rs p rs p rs p rs p

1 Valloxen-1 0.723 0.001 0.781 0.001 0.309 n.s. -0.815 0.001 7 Skokloster-4 0.732 0.01 0.779 0.01 0.267 n.s. -0.870 0.001 9 Parnassen-1 0.607 0.01 0.736 0.001 0.204 n.s. -0.831 0.001 10 Parnassen-2 0.520 0.05 0.668 0.01 0.208 n.s. -0.885 0.001

Light 5.9) and 5.0-6.0 (mean 5.6), respectively. The lowest values were shown by plots 1 and 2 (Valloxen-1 and -2). RLI showed large differences between communities (Ta­ The forest types differed from each other also with ble 29). Light levels in the eutrophic forests were signifi­ respect to soil mineral N, but the variation was fairly high cantly lower than in the mesotrophic forests (Mann­ both within mesotrophic and eutrophic forests, and the Whitney U-test, p < 0.001 for both light at 60 cm and 5 difference was just not significant (Mann-Whitney U-test, p 0.07). For mesotrophic forests, N varied between 23.6 cm). The variation within plots was small, as can be seen = from the fairly small standard deviations. Within the and 67.2 mg/kg (mean 37.5 mg/kg), for eutrophic forests mesotrophic forests, RLI-values varied between 2.4 and between 30.0 and 113.2 mg/kg (mean 48.6 mg/kg). The 8.1 % at 60 cm and 1.9 and 7.0 % at 5 cm. The two plots lowest values were found for plots 8 (Hjalstaviken) and 2 of the Deschampsia jlexuosa form showed the highest (Valloxen-2), the highest ones for plots 9 and 10 values, followed by the plots of the Geranium robertianum (Parnassen-1 and -2). form and those of the Lathyrus vernus form. RLI-values Organic matter contents varied between 10.8 % for in eutrophic forests varied between 0.9 and 2.4 % at 60 cm plot 1 (Valloxen-1) and 20.8 % for plot 12 (Vik- 1). The and 0.8 and 1.8 % at 5 cm. Out of 136 pairwise compari­ means for the mesotrophic and eutrophic forest plots were sons of RLI-values between plots at both heights, 119 exactly the same (14.2 %). Also for dry bulk density, no were significant (Student's t-test, p < 0.05) at 60 cm and clear differences were found between mesotrophic and 121 at 5 cm. Thus, clear differences in light were also eutrophic forests: 0.68 g/cm3 and 0.71 g/cm3, respectively. observed on the level of forest forms. Only for five comparisons (60 cm) and one comparison (5 cm), respec­ tively, were values for the eutrophic forests not signifi­ Table 29. Relative light intensity of diffuse light at the cantly lower than for the mesotrophic forests. (RLI) height of the field layer (60 cm) and at the height of the bottom RLI-values at 60 cm are positively correlated with the layer (5 cm) for the selected forest plots. Given are means and cover values of Quercus robur (Spearman rank correla­ standard deviations (SD), as well as the percentage of light tion, r 0.600, p < 0.05) and negatively correlated with 5 = reaching the bottom layer, relative to the light at 60 cm (PER). those of Ulmus glabra (r 0.759, p < 0.001). No sig­ s =- N is the number of observations. nificant correlations were found for Fraxinus excelsior, 60 cm (%) 5 Plot Forest stand N RLI at RLI at cm (%) PER Acer platanoides and Tilia cordata. SD(+/-) SD(+/-) (%) The percentage of RLI reaching the bottom layer Mean Mean relative to the field layer varied between 47 % and 89 % 1 Valloxen-1 44 5.17 2.5 1 4.62 2.28 89 2 Valloxen-2 50 8.08 1.28 6.98 1.32 86 (Table 29). The highest values were shown by mesotrophic 3 Ragnhildsvik 50 2.40 0.50 1.94 0.69 81 forest plots. There is a significant negative correlation 4 Skokloster- 1 48 0.95 0.28 0.80 0.25 84 between these percentage values and the cover of the field 5 Skokloster-2 43 0.85 0.44 0.70 0.33 82 6 Skokloster-3 50 0.99 0.27 0.82 0.23 83 layer (rs 0.487, p < 0.05). 7 Skokloster-4 50 3.45 0.76 2.28 0.71 66 = - 8 Hjalstaviken 49 2.39 0.81 1.87 0.54 78 9 Parnassen-1 50 1.87 0.27 0.88 0.26 47 Edaphic factors 10 Parnassen-2 50 1.30 0.32 0.88 0.26 67 11 Djurgarden 50 1.94 0.51 1.31 0.44 68 Soil acidity differed significantly between forest types 12 Vik- 1 50 2.24 0.34 1.50 0.42 67 [Mann-Whitney U-test, p < 0.01 for both pH(H 0) and 13 Vik-2 50 2.43 0.74 1.78 0.55 73 2 14 Vardsatra-1 50 1.38 0.35 0.98 0.30 71 pH(KCl)], but there was some overlap. For the mesotrophic 15 Vardsatra-2 50 1.52 0.36 1.25 0.31 82 forests, pH(H20) varied between 4.1 and 5.8 (mean 5.1), 16 Vardsatra-3 50 1.97 0.42 1.41 0.29 72 pH(KCl) between 3.6 and 5.6 (mean 4.7). The corre­ 17 Kungshamn-Morga 50 2.67 0.66 2.31 0.56 87 sponding values for eutrophic forests were 5.4-6.4 (mean

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 87

Table 30. Results of the CCA-ordination ofreleves and environmental variables, compared to data on the CA-ordination ofreleves only.

CCA axis CA axis 2 3 4 2 3 4

Eigenvalue 0.487 0.309 0.23 1 0.219 0.489 0.321 0.231 0.220 Sum of eigenvalues 2.215 2.3 17 Cumulative percentage variance of species data 21.0 34.4 44.3 53.8 21.0 34.8 44.7 54.2 Cumulative percentage variance of species/environment relation 22.1 35.9 46.4 50.3 Species-environment correlation 0.998 0.998 0.999 0.998

Eutrophic forest plots showed both the highest value (0.81 ronmental variables thus accounted for the main variation g/cm3, plot 6) and the lowest value (0.62 g/cm3, plot 13). in the floristic composition (cf. Jongman et al. 1987). Fig. The highest soil moisture scores were found for the 32a reveals a diagonal structure regarding the position of eutrophic forest plots 12 (Vik-1), 11 (Djurgarden) and 4 releves and arrows for the environmental variables. The (Skokloster-1), the lowest scores for the mesotrophic reason for this is that the ordination axes tend to form the forest plots 1 (Valloxen-1) and 8 (Hjalstaviken). How­ diagonals of the space of the main environmental gradi­ ever, also some eutrophic forest plots showed low scores, ents (cf. Loucks 1962). The mesotrophic forest plots are such as plots 5 and 6 (Skokloster-2 and -3), as well as plot situated in the right part, with plots 1 and 2 (Valloxen-1 14 (Vardsatra- 1). In general, forest types did not differ and -2) well separated from the other plots. The eutrophic with respect to their moisture scores, nor with respect to elm-ash forest plots appear close together in the lower left absolute soil moisture values on any of the days. At the corner, except for plots 9, 10 and 11 (Parnassen-1 and -2, day with the lowest values, mesotrophic forests had a Dj urgarden) situated in the upper left corner. Plot 15 mean soil moisture content of20.8 % (range 15.9-24.5 % ), (V ardsatra-2), representing an alder-ash forest, lies close eutrophic forests of 22.9 % (range 15.4-29.1 %). to the two other plots ( 14 and 16) from the same forest stand. Two gradients are of great importance along axis 1: a Ordination light gradient (increasing values oflight, decreasing values The ordination diagrams with axes 1 and 2 of the CCA are of canopy cover towards the right) and a complex-gradi­ shown separately for sample plots/environmental vari­ ent in nutrient status [increasing values of pH(�O)/ ables (Fig. 32a) and species (Fig. 32b ). Eigenvalues were pH(KCl) and nitrogen towards the left]. This is also 0.487 for axis 1 and 0.309 for axis 2, and they were almost expressed by the inter-set correlations between the envi­ as high as the corresponding eigenvalues of the un­ ronmental variables and the site scores, given in Table 31. constrained CA-ordination (Table 30). The species-envi­ The arrows for inclination, maximum temperature and ronment correlations were very high. The measured envi- temperature range point in the same direction as light, but are shorter. Soil moisture and density increase towards the upper left hand corner, evaporation in the opposite direc­ tion. The other variables (minimum temperature, heat Table 31. Inter-set correlations of environmental variables with CCA axes 1 and 2. index and exposure) are represented by very short arrows and are the only variables with higher inter-set correla­ Variable Axis 1 Axis 2 tions with axis 2 than with axis 1. Correlations between

GRAD 0.5 1 0. 11 the environmental variables are given in Table 32. HEAT!ND 0.05 -0.07 The species diagram in Fig. 32b reveals a pattern EXPOSURE 0. 15 0. 19 which is in accordance with the results of the general pHWATER -0.81 -0.27 pHKCl -0.81 -0.30 vegetation survey in Chapter 4. Differential species of ORGANMAT -0.18 -0.04 oligotrophic sites (e.g. Va ccinium myrtillus, V. vitis-idaea NITROGEN -0.69 0.47 and Hieracium umbellatum) are situated to the right in the MOISTURE -0.45 0.11 DENSITY -0.31 0.23 diagram. Species of slightly oligotrophic-mesotrophic for­ LIGHTFLD 0.75 0.36 ests (e.g. Campanula persicifolia, Dactylis glomerata, CANOPCOV -0.60 -0.44 Veronica chamaedrys and Calamagrostis arundinacea) MAXI TEMP 0.55 0.30 MINITEMP -0.19 -0.33 are concentrated in the right half. Finally, species with DIFFTEMP 0.43 0.27 their highest frequencies in eutrophic forests, i.e. those EVAPOR 0.34 -0.24 indicating high pH and amounts of soil nutrients, can be

Acta Phytogeogr. Suec. 80 88 M. Diekmann

a 3 N � X

2

10

1 Nitrog�n 11 Li9hltld

0

-1 Conopcov 16 15

-2 AXIS 1

-2 -1 0 1 2

b 2 Ane ronv Sol ,virg m Fro mosc Cor pilu Pop t r!""'e acu Lil mart ;:ass 0 Hyp • Ron 0 p� .abie o:� b�;edi N podo eo., • Aeg ry � � Hie m be viti a Vac • , C m X Luz p1IO rotu Pie schr • Pte oqui• 1 id

found in the left half of the diagram, such as Galium occasions of a permanently clear sky. ALI(excl)-values aparine, Aegopodium podagraria and Stachys sylvatica. were more or less constant during the day, except for OL­ plot 3, where they were slightly rising. ALI(incl)-values were usually only slightly higher than ALI(excl)-values, Climatic characteristics of the Oland plots but showed a somewhat greater variation between hours. General information about the bland plots (OL-plots) is An exception was OL-plot 5, where both the values and given in Table 33. Results of the light measurements are their variation were much higher than in other OL-plots. shown in Fig. 33a-e. TLI-values usually showed a typi­ RLI (incl)- and RLI(excl)-values had similar curves, ex­ cally parabolic patternin the course of a day, particularly cept for OL-plots 4 and 5 where RLI(incl) fluctuated very in OL-plots 3 and 5 where measurements were made on much. In some plots, both showed a reverse pattern as

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 89

Table 32. Spearman rank correlation (rs) for pairs of environmental variables, given to the right of the diagonal. Significance levels are given to the left of the diagonal. n.s. = not significant.

6 10 11 12 13 14 15

I GRAD 0. 176 -0.358 -0.302 -0.03 1 -0.25 1 -0.126 -0.171 0.425 -0.416 0.439 -0.005 0.230 0.3 14 0.660 2 HEATIND 0.439 -0.202 -0.140 -0.045 -0.191 -0.271 0.366 0.263 -0.4 17 0.391 -0. 124 0.340 0.01 0.612 3 EXPOSURE -0.239 -0.120 -0.172 0.431 -0.361 0. 189 0.007 -0.382 -0.171 n.s. n.s. 0.527 0.532 0.591 4 pHWATER 0.075 0.297 0.235 0.425 -0.359 0.027 -0.227 -0.138 n.s. n.s. n.s. 0.957 -0.693 0.527 5 pHKCI 0. 152 0.362 0.334 0.397 0.433 -0.221 -0.126 -0.080 -0.096 n.s. n.s. n.s. 0.001 -0.686 6 ORGANMAT 0.358 0. 181 0.091 -0.101 -0.309 -0.089 n.s. n.s. n.s. n.s. n.s. 0.858 -0.516 -0.492 NITROGEN 0.297 0.391 -0.335 0.220 -0.254 -0.079 -0.185 -0.131 7 n.s. n.s. n.s. n.s. n.s. n.s. 8 MOISTURE -0.267 0.01 1 0.049 -0.456 -0.235 -0.238 -0.304 n.s. n.s. n.s. n.s. n.s. 0.001 n.s. 9 DENSITY -0.458 0.141 0. 139 0.021 0.059 -0.167 n.s. n.s. n.s. n.s. n.s. 0.05 n.s. n.s. 10 LIGHTFLD -0.416 0. 141 0.072 0.015 0.220 n.s. n.s. n.s. 0.01 0.01 n.s. n.s. n.s. n.s. 11 CANOPCOV 0.325 -0.201 n.s. n.s. n.s. 0.05 n.s. n.s. n.s. n.s. n.s. n.s. -0.652 -0.649 12 MAXITEMP -0.455 0.330 n.s. n.s. 0.05 n.s. n.s. 0.05 n.s. n.s. n.s. n.s. 0.01 0.895 13 MINITEMP 0.016 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. -0.679 14 DIFFrEMP 0.309 n.s. n.s. 0.05 n.s. n.s. n.s. n.s. n.s. n.s. n.s. 0.01 0.001 0.01 15 EVAPOR n.s. 0.05 0.05 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

compared with TLI, with comparatively high values at were reached in late afternoon (54 % ). In OL-plot 1, the early morning (most pronounced in OL-plot 2) and late fluctuation was particularly small, and values did not fall afternoon(most pronounced in OL-plots 3 and 5), and low below 87 %. values at noon. The temperature-curves showed a similar patternin all OL-plots (Fig. 34a-e). Temperatures were increasing in 6.4 Discussion the order T(-15), T(-5), T(O) and T( 100); only occasion­ ally T(O) exceeded T( 1 00). In OL-plots 1, 3 and 4, T( -15) Temperature and T(-5) were very close to each other. The below­ ground temperatures were much lower and less variable Deciduous forests have a moderating effect on tempera­ than temperatures above-ground. During the course of a ture fl uctuations. Compared with the open, maximum day, the below-ground temperatures slightly rose, but the temperatures are lower, minimum temperatures are higher, absolute ranges between morning andlate afternoon were and, thus, the temperature range is much smaller (Fig. not larger than 1.5 °C for T(-5) and 0.8 oc for T(- 15), 28b, Fig. 29b, Fig. 30b). This has been shown in many except for OL-plot 5 with corresponding values of 2.0 oc studies (e.g. van der Poel & Stoutjesdijk 1959; Heckert and 1.8 °C, respectively. T(O) and T(IOO) showed much 1959; Pahlsson 1969). Fluctuations over the day are very larger fluctuations during the day; they rose in the morn­ small a few cm below-ground (cf. Hytteborn 1975), some­ ing until noon to remain more or less stable or slightly what larger at ground level and considerably larger at 100 falling during the afternoon. OL-plot 5 formed an excep­ cm height (Fig. 34). This vertical gradient was described tion, with more or less constantly rising above-ground for differentfo rest types by, e.g. van der Poel & Stoutjesdijk temperatures (and high absolute values). (1959), Nihlgard (1969) and Stoutjesdijk & Barkman In general, air humidity was a mirror image of the (1992). In this study, the moderating effect was more above-ground temperatures (Fig. 34a-e). In OL-plots 1-4, pronounced in the deciduous forests than in the pine values were highest in the morning (between 91 and 97 %) forests, which had a more open canopy and understorey and lowest at noon or in early afternoon.However, in OL­ (cf. Geiger 1961). In a comparison of ten forest types in plot 5, the maximum was only 84 %, and the lowest values the Bialowieza forest in Poland, Matuszkiewicz (1977) showed that coniferous forest in the course of a year had later frosts in spring, earlier frosts in autumn and a consid­ erably shorter frost-free period than deciduous forests. Table 33. Short characteristics of forest plots on bland. Given are also the mean relative light intensities (RLI) measured under However, Nihlgard ( 1969) showed that a beech forest had overcast conditions (diffuse light only). higher surface and soil temperatures than a spruce forest during summer despite very similar light levels, caused by OL- Forest stand Forest type Aspect Wind RLI (%) a higher heat accumulation due to a higher water content plot exposure in the upper soil of the beech forest (cf. Geiger 1961). 1 Stora Dalby eutrophic elm-ash forest 1.68 + Outside the growing season, the evergreen coniferous 2 Gillsattra mesotrophic forest 2.73 Eriksore mesotrophic forest 2.61 forests show smaller fluctuations than the deciduous for­ + 4 GIOmminge mesotrophic forest 1.89 ests. In early spring, the maximum ground temperatures 5 Stora Vickleby oligotrophic oak forest 3.88 in deciduous forests are much higher than in coniferous

Acta Phytogeogr. Suec. 80 90 M. Diekmann

20 200 a 20 200 b All All TU TU RLI RU � 15 150 15 150

.. _""\' ·- -"' 10 100 10 100 .• J!;

* * '* 5 50 5 50

0 0 0 0

7 8 9 10 11 12 13 14 15 16 17 18 19 7 8 9 10 11 1 2 13 14 15 16 17 18 19

Hour Hour 20 200 c d 20 200 All All TLI RLI TU RLI 15 '*"-"""' 150 ,.. 15 ·* · � 150 � f Jl \ lt\ • 10 100 * ,* ' 10 • 100

* ·* *. • *, * 5 50 5 50

0 0 � 0 0 7 8 9 10 11 12 13 14 15 1 6 17 1 8 19 7 8 9 10 11 12 13 14 15 16 17 18 19 Hour Hour

20 200

All e TU • RL1 (excl) RLI - • - RU (incl) 15 150 0 - ALI (excl) D All (incl)

* -- TLI 10 100

Fig. 33. Light intensity (PAR) over a day in five forest plots on bland. Shown are total light intensity outside the forest (TLI) 5 50 and absolute light intensity inside the forest (ALl), scales in 10 x microeinstein/m2/s on the left vertical axis, and relative light �= : :: =� intensity inside the forest (RLI), scale in % on the right vertical 0 0 axis. a. Stora Dalby (OL-plot 1); b. Gillsattra (OL-plot 2); c.

7 8 9 1 0 11 1 2 13 14 15 16 17 18 19 Eriksore (OL-plot 3); d. Glomminge (OL-plot 4); e. Stora O Hour Vickleby ( L-plot 5).

forests (cf. Nihlgard 1969), and may even exceed those in Mesotrophic forests had significantly higher maxi­ the open, as the litter layer is warmed up by the sun (cf. mum temperatures than eutrophic forests (Mann-Whitney Stoutjesdijk & Barkman 1992). This has a positive effect U-test, p < 0.05), whereas no significantdif ferences were on many early forest species, particularly the spring found for minimum temperature and temperature range. geophytes (Anemone spp., Gagea spp., etc.), which are Both maximum temperature and temperature range are photosynthetically most active during this period before negatively correlated with canopy cover (Spearman rank < < tree leaf development. correlation,rs =- 0.652,p 0.01 and rs =� 0.649,p 0.01,

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 91

30 100 a 30 100 b

. - -· - . 90 .. 90 . ·- - . ' •. 25 .. 25 ?; 80 '6 E ... - -��.

::J � .c � 20 70 20 7 0 3 iU 3 '(\i crl crl (i; (i; .� ED. ED. eo Q) Q) I- I- 60 � 15 15

e e � 50 .:r-G a e e e 50

1 0 +----.----.--,--.---,--.----.--,�-.---.--.---+ 40 7 8 9 10 11 12 13 14 15 16 17 18 19 7 8� 9 10 11 12 13 14 15 16 17 18 19

Hour Hour

30 100 c 30 . 100 d

90 90

25 25

•, 80

-· � ,• 20 70 � 3 .3 20 70 ·ro Qjcrl Cll w ·� g> D. E 60 D. � · l .. fQ)· Q) - 60 & I- 15 15 �---· 50 a e e o 50

1 0 - 40 -- ,-- 7 B 9r· --, 10 11 12 13 14 16 17 18 19 10 15 40 7 8 9 101 111- 12 13 14 15 16 17 18 19

Hour Hour

30 100 e

• - - Rei. air humidity 90

*- T( 1 00) 25

* - T(O)

e-T(-5) 0 - T(- 15) � 20 70 3 ro crl05 g> ED. � � 60 & 15 Fig. 34. Relative air humidity, above-ground and below-ground 50 temperatures over a day in five forest plots on Gland. For abbreviations, see text. a. Stora Dalby (OL-plot 1); b. Gillsattra 10 +-··· ·-.-.-.-.----.----.�-.-�---.----+ 4 0 (OL-plot 2); c. Eriksore (OL-plot 3); d. GlOmminge (OL-plot 4); 7 8 9 10 11 12 13 14 15 16 17 Hl 19 e. Stora Vickleby (OL-plot 5). Hour

respectively, Table 32) which is significantly lower in provides an exchange of air masses between the forest and mesotrophic forests than in eutrophic forests (Mann­ the open, which results in higher temperature fluctua­ Whitney U-test,p < 0.05). However, these two factors are tions. In OL-plot 5, representing an oligotrophic oak strongly influenced by some physiographic factors: they forest, the high temperature fluctuations (cf. Fig. 34e) are positively, though not significantly, correlated with were probably caused by the open canopy and the subse­ inclination and heat index, and they are significantly quently high light level, as the wind exposure is low positively correlated with wind exposure (r 0.527, (Table 33). Minimum temperatures are influenced by s = p < 0.05 and rs 0.532, p < 0.05, respectively). Wind canopy cover and wind exposure (Table 32), but also by =

Acta Phytogeogr. Suec. 80 92 M. Diekmann

topographic location. Particularly high minimum tem­ influenced by the cover of certain tree species, namely peratures were shown by plots on more or less steep Quercus robur (positive correlation) and Ulmus glabra slopes (e.g. plot 7, Skokloster-4) (cf. J. Lundqvist 1968) (negative correlation). Ellenberg (1986) classified tree and by plots located very close to Lake Malaren (plots 14- species on their ability to produce shade in a stand on a 5- 16, Vardsatra- 1, -2 and -3). On slopes, the cold air is degree scale (5: much shade, 1: little shade). Figures for flowing downwards during the night, and the nearby lake main tree species were: Quercus robur: 2, Fraxinus moderate the temperature decline at night. excelsior: 3, Ulmus glabra, Acer platanoides and Tilia cordata: 4. The results of this study are thus partly in accordance with Ellenberg' s classification.Howeve r,Acer Evaporation and Tilia do not seem to create the same degree of shade as In general, evaporation depends on temperature, satura­ Ulmus. It must be noted that the canopy openings not only tion deficit and wind velocity, which is confirmed by this control light intensity, but also throughfall precipitation study. Daily evaporation is significantly correlated with (Stoutjesdijk & Barkman 1992), which can be decisive for maximum temperature (Table 28), and maximum tem­ the response of the field layer (R.C. Anderson et al. 1969). perature was often,but not always, higher in mesotrophic Some RLI-values at 60 cm lay below 1%and were forests than in eutrophic forests (see above). Accordingly, thus extremely low (cf. Tranquillini 1960), even in com­ evaporation in mesotrophic forests did not always exceed parison with beech and oak/hombeam forests, two forest that in eutrophic forests either. This is caused by the types that are known for low light intensities on the overruling effect of the physiographic factors heat index ground (Eber 1972; Ellenberg 1986). However, former and wind exposure, which are positively correlated with studies were usually concerned with light of all wave­ r 0.612 ,p < 0.05 and r evaporation (Table 32, s = s 0.591, lengths, not only PAR, and light of wavelengths > 700 nm = p < 0.05, respectively). According to Parker's formula is much better transmitted through the canopy than PAR (K.C. Parker 1988), the heat index of slopes increases light (Larcher 1980; Stoutjesdijk & Barkman 1992). A with inclination and is maximal at a SW aspect. In fact, low RLI-value goes parallel with a decreased variation particularly high evaporation was shown by plots 1, 3, 7 (see Table 29), which was also observed by, e.g. Nageli and 8, which all were W- or WSW-exposed, and which (1940). The standard deviations of RLI and RLI are had a moderately high to high wind exposure. On the significantly positively correlated (Spearman rank corre­ other hand, plot 17 (Kungshamn-Morga) had low evapo­ lation, r 0.802,p < 0.001 , at 60 cm; r 0.894,p < 0.001, s = s = ration values, as it was SE-exposed and had a low wind at 5 cm) (cf. Eber 1972). RLI-values in a stand may vary exposure. The average wind velocity shows only a weak from year to year due to variation in the number, size and positive correlation with evaporation (Table 28), since it thickness of the leaves of particular species. Quercus becomes overruled by the local wind exposure (Table 26). robur, for example, varies appreciably in leaf development The differences in evaporation between forest types caused due to attack by the insect To rtrix viridana (Fam. Tortri­ by temperature and wind may be accentuated by differ­ cidae, Lepidoptera) (cf. M.C. Anderson 1964b; Stoutjes­ ences in relative air humidity, which decreases with in­ dijk & Barkman 1992). Some RLI-values determined in creasing temperature. The measurements in the bL-plots 1992 by C. Gustafsson (1992) from the same plots dif­ (Fig. 34) revealed a good correlation between saturation fered considerably from the values given in Table 29. deficit and temperature, which was also demonstrated by The values of the ratio RLI (5 cm)IRLI (60 cm) were Pahlsson (1969) and Nihlgard (1969). This has also been fairly high, ranging from 47 to 89 % (mean 75 %, Table shown for vertical profiles in deciduous forests (e.g. 29). However, measurements in some bland forests Heckert 1959; Stoutjesdijk & Barkman 1992) as well as in (Diekmann, unpubl.) gave lower values of 30 to 40 %. a comparative way for different types of deciduous forests The vertical gradient and decrease of light is not only (Stoutjesdijk & Barkman 1992). In the bL-plots, the dependent on the total cover of the field layer, but also on smallest fluctuations were observed in the eutrophic for­ its architecture and the occurrence of particular species est plot, low absolute values and large fluctuations in the (cf. Tranquillini 1960). Examples of strongly shading oligotrophic forest plot, and intermediate values in the species are Pteridium aquilinum (Salisbury 1936) and mesotrophic forest plots. Mercurialis perennis (common in the investigated bland stands). Instantaneous measurements of diffuse light do not Light consider direct sunlight and do not show the important Light is of maj or importance for the diffentiation of de­ strong seasonal variation of the light climate (Evans 1956; ciduous forests. Eutrophic forests have a closer canopy M.C. Anderson 1964b). The importance of the light spring and lower light levels than mesotrophic forests. Signifi­ phase for the distribution and density of species was cant differences in RLI also exist on the level of forest demonstrated for Hyacinthoides non-scripta (not occur­ forms. As has been previously shown, RLI is particularly ring in Sweden) by Blackman & Rutter (1946) and for

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 93

other species by Eber (1972). Differences in RLI of dif­ intensity due to sunflecks, also other aspects of the fuse light between different deciduous forest types and microclimate change (cf. Nageli 1940; M.C. Anderson their impact on plant distribution have been demonstrated 1964b): temperature and species transpiration rise, which in many studies (e.g. Eber 1972; Fekete 1974). Also the may negatively influence photosynthesis (cf. Ellenberg spatial distribution of species within a stand is often 1986). Also in early spring, a period which contributes a correlated with the distribution of RLI measured on over­ high proportion of the total light for many forest plants cast days (Ellenberg 1939; Fekete 1974; C. Gustafsson due to high absolute light intensities, the light distribution 1992), though the impact of sunflecks may also be empha­ on overcast days is fairly similar to that in summer (Eber sized (see below). 1971). RLI of diffuse light may not always explain plant distribution and the differentiation of communities, but Edaphic factors direct light may be of great importance. However, a comparison of the light measurements under clear and Soil acidity is usually considered one of the most impor­ cloudy conditions in the OL-plots showed that absolute tant causes offloristicvariation in forests (e.g. Sjors 1961; light intensities hardly differed during midday, despite Tyler 1989; R.H. 0kland & Eilertsen 1993). Base satura­ much higher total light intensities outside the forest under tion and the contents of minerals such as Ca have been sunny conditions. This is due to the absorbance and reflec­ shown to be more or less strongly correlated with pH in tion of direct light by the forest canopy. Therefore, RLI­ Scandinavian deciduous forests (A. Bj�rnstad 1971; values measured under sunny conditions were much lower Kielland-Lund 1981; Rtihling & Tyler 1986; T. 0kland than those measured under cloudy conditions, which has 1988; Brunet & Neymark 1992), as well as in coniferous also been shown by Nageli ( 1940), Sjors ( 1954), Tranquillini forests (e.g. Rydgren 1993). The influence of pH on the (1960) and Fekete (1974). Still, RLI-values under sunny species composition and community differentiation in conditions showed similar differences between communi­ northerndeciduous forests has been demonstrated in vari­ ties as RLI-values under overcast conditions (see above). ous studies (e.g. A. Bj�rnstad 1971; Fremstad 1979; T. This was also demonstrated by Eber (1971, 1972) for the 0kland 1988; Brunet 1993). Also in the present study, spatial distribution of RLI within a stand, whereas Evans pH(H20) and pH(KCl) have the highest inter-set correla­ ( 1966) showed substantial differences. tions with axis 1 in the CCA-ordination (Table 31). How­ However, direct light which is transmitted results in ever, it is not clear which physiological processes affect sunflecks which dramatically increase the light level at the response of the vegetation. There may be a direct the ground. Sunflecksmay be important for the plants of influence of hydrogen ion activity, but secondary changes the forest floor as they may represent the only periods are probably more important: pH influences the availabil­ during the day when the light intensity exceeds the com­ ity of mineral nutrients, the solubility of toxic metals, the pensation point of a plant species (Evans 1956). They soil fauna composition, the litter decomposition and N 'move' over the ground, and both their number and dura­ mineralization rate (cf. Tyler 1989). In the present study tion at a particular place are important. The measurements (cf. Table 32), nitrogen is positively correlated with in the OL-plots showed clear differences between com­ pH(H20)/pH(KCl), which was also shown by A. Bj�rnstad munities. InOL-plot 5, an oligotrophic oak forest, sunflecks (1971), Aune (1973) and Fremstad (1979). A close corre­ were very frequent and often rose to more than ten times lation between N availability and base saturation/pH was the light intensity in the shade (more than 2/3 of TLI), also demonstrated for northern coniferous forests by, e.g. sometimes exceeding a quantum flux density of 1 milli­ Dahl et al. (1967), Rydgren (1993) and R.H. 0kland & 2 einstein/m /s (Fig. 33e). They were less frequent in the Eilertsen (1993). This complex-gradient in nutrient status other OL-plots and rarely rose to more than five times the is probably the most important factor for the differentia­ shade intensity (Fig. 33a-d) (cf. Tranquillini 1960). Apart tion of northern deciduous hardwood forests. This is also from one extreme value, sunflecks were least frequent in expressed by the terms 'oligotrophic', 'mesotrophic' and OL-plot 1, the eutrophic elm-ash forest: 6 out of 11 hourly 'eutrophic', used for the description of the main forest measurements did not include any sunflecks, whereas the types. Nutrient characteristics of the mineral soil were corresponding figures for the three mesotrophic forests also shown to be a primary cause of the floristic variation are 01 10, 211 1 and 2110. Thus, also the sunfleck probabil­ in northern coniferous forests (e.g. Kuusipalo 1985; T. ity decreases with decreasing RLI-values of diffuse light. 0kland 1990; R.H. 0kland & Eilertsen 1993; Rydgren The good correlation between plant distribution and RLI 1993). measured under overcast conditions may be attributed to No clear differences between eutrophic and mesa­ the fact that the duration of sunflecks is severely limited trophic forests were found for organic matter contents by cloudiness (cf. Atkins et al. 1937) and that the sunflecks which varied between about 10 % and 20 %. Similar or move, diminishing their assimilatory value for a particu­ slightly higher figures were given by A. Bj�rnstad(19 71), lar plant (Salisbury 1936). Besides, with increasing light Fremstad (1979), Kielland-Lund (1981) and Rtihling &

Acta Phytogeogr. Suec. 80 94 M. Diekmann

Tyler ( 1986) for corresponding communities. Much higher occurrence of certain tree species and forest commu­ values can be found in oligotrophic deciduous forests (cf. nities, but will, on the other hand, be changed by them. Sj ors 1961; A. Bjl2Srnstad 1971; Olsson 1974; Kielland­ R.H. 0kland & Eilertsen (1993) showed that topogra­ Lund 1981; Riihling & Tyler 1986; T. 0kland 1988). For phy was the most important underlying environmental an oak forest stand on acid soil on Oland, organic matter factor in a Norwegian coniferous forest on poor soil, contents varied between 32 % and 55 %. affecting species composition via its influence on soil Moisture as determined in this study did not account depth and nutrient contents. The relative importance of for much of the floristicvariation. It tended to be higher in physiographic factors is also well documented for conif­ eutrophic than in mesotrophic forests, but the differences erous esker vegetation (e.g. Oksanen 1983; Heikkinen were not clear. In general, all forests were moderately dry 1991) and for the (deciduous) vegetation on south- (west­ to mesic, i.e. the range in soil moisture was fairly narrow. /east-) facing slopes as pointed out earlier (see Chapter 4). However, differences in soil moisture may still have a Except on very steep slopes, physiographic factors have large impact on the vegetation of the stands studied, mainly an indirect effect, due to their influence on both because average soil moisture conditions were used as an climatic and edaphic factors. In the present study, inclina­ environmental variable in this study, while the frequency tion, aspect (heat index) and wind exposure caused an and duration of extremely low soil moisture values may increased evaporation and more extreme temperatures be of great significancetoo (cf. R.H. 0kland & Eilertsen (cf. Table 32). These in turn may have affected soil 1993). However, a CCA-ordination using moisture values moisture and nitrogen mineralization. after a long dry period instead of average values resulted The CCA-ordination revealed that pH and light were in insignificantly higher eigenvalues and species-envi­ the primary causes of floristic variation in the field and ronment correlations. In the selected plots, canopy inter­ bottom layers (Fig. 32a). The importance of soil acidity ception does not seem to have a large impact on soil and canopy cover for the field layer composition were moisture, as canopy cover and moisture were practically also emphasized by Tyler (1989) in a study of oak/ uncorrelated (Table 32). Moisture shows a strong positive hornbeam forests in S Sweden. Correlations between correlation with organic matter, and it is also positively species occurrences and these factors can even be de­ 2 (though not significantly) correlated with pH(H20)/ tected on a small scale (1 m ) within forest plots (cf. pH(KCl) and nitrogen. Moisture is negatively correlated C. Gustafsson 1992). pH and light are not directly interre­ (but not significantly) with the physiographic variables, lated (cf. Tyler 1989), but there is an indirect interrelation­ indicating that steeper sites are better drained, and that the ship, indicated by the strong negative correlation between soils of south- (west-) facing and wind-exposed slopes light and pH(H20)/pH(KCl) found in this study (Table dry up relatively fast. In a SE Norwegian beech forest, soil 32). In forests with different canopy closures, and conse­ moisture was shown to be topographically conditioned quently different light levels in the understorey, throughfall (T. 0kland 1988). In general, only a few comparative precipitation and chemistry may also be different, but this studies of soil water contents have been carried out in has probably only a modifying and local effect on soil pH Scandinavian deciduous forests, mainly in Nemoral for­ and nutrients. A more probable explanation is that a high ests of southernmostSweden and Norway (P:Thlsson 1969; nutrient status, indicated by a high pH, favours tree growth F. Andersson 1970; T. 0kland 1988). Andersson's com­ in general and particularly the occurrence of Ulmus glabra prehensive investigations in an oak woodland and meadow and, at the same time, disfavours Que reus spp. by compe­ area in Skane on partly fairly wet soils showed the impor­ tition, causing a higher canopy cover and lower light tance of soil moisture characteristics for the vegetational values. On very acid soils, on the other hand, Quereus spp. gradients. The impact of soil moisture on the differentia­ are relatively favoured, and tree growth in general is tion of the vegetation of northern coniferous forests and hampered, resulting (particularly in the absence of Fagus its correlation with nutrient characteristics were demon­ and Pieea) in a lower canopy cover and increased light strated by, e.g. T. 0kland (1990), Rydgren (1993) and intensities on the forest floor. In northern coniferous R.H. 0kland & Eilertsen (1993). forests, tree cover and fertility characteristics may be only weakly linked (Kuusipalo 1985; Heikkinen 1991; R.H. 0kland & Eilertsen 1993; Rydgren 1993; but see T. 0kland The importance of abiotic factors - conclusions 1990). The fact that nutrient status and light accounted for Among the abiotic environmental factors, three types were the main floristic variation is suggestive of the theories distinguished: physiographic, edaphic and climatic. All and ideas about the dynamics and structure of plant com­ three factors were shown to differ between forest plots and munities by Tilman (e.g. 1988), who considered soil to account for part of the floristicvariat ion. However, they nutrients (a below-ground resource) and light (an above­ are often interrelated, and it may be difficult to identify the ground resource) as the major constraints upon plants. ultimate cause of the differentiation of the vegetation. The Both are nutritionally essential for plants and may be the forest climate, for example, partly determines the major determinants of community pattern.

Acta Phytogeogr. Suec. 80 7 General discussion and conclusions

This chapter is concernedwith a general discussion of the The same applies to some taxa which have their ecologi­ results presented in Chapters 4, 5 and 6. With regard to cal optima in more open vegetation types, e.g. Veronica species composition and structure, the Boreo-nemoral chamaedrys, Rubus saxatilis, Geranium sylvaticum and forests will be compared with their Nemoral counterparts. Allium oleraceum. Also in the bottom layer, several taxa At the end of each section, some concluding remarks will are more common in Boreo-nemoral forests, such as be given. Rhytidiadelphus triquetrus, Eurhynchium hians, E. angustirete, Brachythecium velutinum and Th uidium tama­ riscinum. On the other hand, e.g. Mnium hornum, 7.1 Species Polytrichum fo rmosum and Dicranella heteromalla are less common. In southernmost Sweden, both groups of species may commonly occur, but then the 'Nemoral' Species richness and general species composition species tend to occur in beech forests on poor to moder­ Regarding forest types, the total number of species (aver­ ately rich soils and the 'Boreo-nemoral' species in elm­ age of community types) varied between 29 (eutrophic ash forests on the richest soils (cf. Sjogren 1991). elm-ash forests) and 37 (mesotrophic mixed deciduous forests) (cf. Table 2). The figures for different community Life forms types ranged from 22 to 51. In order to compare these figures with figures on species richness for Nemoral for­ All forest types have similar life form spectra (Table 34): ests, 260 releves (from Forster 1981 and Hofmeister 1990) Hemicryptophytes are clearly dominating, as in N emoral of similar size of the whole spectrum of deciduous hard­ forests (cf. Ellenberg 1986) and in the Scandinavian flora wood forest communities in two areas at the northern as a whole (Raunkiaer 1934). Also phanerophytes are edge of the German 'Mittelgebirge' region were used. well represented due to the fairly large number of tree and Here, the average species number of different associa­ shrub species. Geophytes are important, with about 20 % tions ranged from 16 to 40, with the majority of figures of the species in all types except in oligotrophic oak below 30. Figures from other areas in the same region forests. The reason might be that spring geophytes only were similar. This indicates that the species number per have a short time available for photosynthesis (the period unit area of the Boreo-nemoral forests is similar to or even from early spring until tree leaf development when most can exceed that of their counterparts in at least some parts of these species start to wither), and that a sufficient of the Nemoral zone. assimilation and net production is possible only on However, in general, there is a floristic gradient in sufficiently fertile soils, i.e. mesotrophic to eutrophic (cf. central and northernEurope, with decreasing species rich­ Ellenberg 1986). As in Nemoral deciduous forests, chamae­ ness towards the north, and some differences in the gen­ phytes and therophytes are least frequent and reach a eral species composition exist between the two vegetation somewhat higher species number only in oligotrophic zones. With regard to tree species, these differences have oak forests. Among the former, Ericaceae (woody) and already been mentioned in Chapter 2.1. Concerningshrubs, Ribes alpinum is frequent in various Boreo-nemoral com­ munities, while, in central Europe, it is confined to Table 34. Life form spectra of Boreo-nemoral deciduous forest types/communities. 1. Oligotrophic oak forests, Table 4; 2-4. eutrophic, montane to sub-alpine forests. Considerable Mesotrophicmixed deciduous forests (2. Quereus robur-Fraxinus differences are found with respect to field layer species. excelsior community, Table 8; 3. Quercus robur-Euonymus Several southerly distributed taxa common in Nemoral europaeus community, Table 1 0; 4. Quereus robur-Tiliacordata forests are completely absent in northern Europe, e.g. community, Table 12); 5. Eutrophic elm-ash forests, Table 15; Aconitum vulparia, Leucojum vernum,Galium sylvaticum, 6. Eutrophic alder-ash forests, Table 18. Seneciofuchsii, etc. Besides, several atlantic species found in central Europe are largely lacking, such as /lex Life form Forest type/Community NW 2 3 4 5 6 aquifolium and Corydalis claviculata. However, some deciduous forest species are common in several of the Hemicryptophytes 39 44 43 46 45 52 Phanerophytes 32 32 29 31 31 26 Boreo-nemoral communities, whereas they are less com­ Geophytes 13 21 22 18 22 19 mon and often restricted to a few communities in the Chamaephytes 10 2 4 3 1 2 Nemoral zone, e.g. Actaea spicata and Viola mirabilis. Therophytes 6 1 2 2 1 1

Acta Phytogeogr. Suec. 80 96 M. Diekmann

Veronica-species (herbaceous) are most conspicuous. values for a comparison of climatic conditions between Among the few annuals can be mentioned Impatiens noli­ communities within the zone. Moreover, whereas the tangere and Melampyrum spp. climatic variables used describe the macroclimate, the vegetation, as expressed in the indicator figures, respond to the forest climate which can considerably deviate from Species indicator values the macroclimate (see Chapter 6). However, the use of T­ Ellenberg's indicator value system represents a quantifi­ and C-values for a comparison between zones may still be cation of ecological responses of plants to climatic and justified. edaphic factors. This system (as well as other indicator value systems, e.g. that proposed by Landolt 1977) has Plant geography been criticized because (1) it expresses complicated eco­ logical responses of species in simplified values, (2) it Here, some plant geographical aspects will be discussed, may be of value only within a restricted geographic range using Ellenberg's T- and C-values as mentioned above. (e.g. Waiter & Breckle 1983), (3) it is supposed to be Most species in Boreo-nemoral deciduous forests show inferior to measurements, and (4) strictly mathematically, T -values of 4, 5 and 6. The average T -figure for both the ordinal scale of the system does not allow a weighted mesotrophic mixed deciduous forests and eutrophic elm­ average procedure (Bocker et al. 1983). ash forests was 5.1 (see Table 2), which is exactly the Still, ter Braak & Barendregt (1986) theoretically sup­ same as calculated for the 'Mittelgebirge' region in Ger­ port the method of weighted averaging of species indica­ many previously mentioned, although the annual mean tor values. Besides, the use of indicator values has some temperature in the latter region is considerably higher advantages (see van der Maarel 1993; Dierschke 1994); (around 8.5 °C). Accordingly, the forest types mentioned for example, it may serve to gain 'integrated' information represent southernvegetation types. In contrast, the aver­ on factors that cannot be identified by single observations age T-figures of the oligotrophic oak forests and eutrophic due to temporal fluctuation. Ellenberg et al. ( 1991) give alder-ash forests are much lower (both 4.6, see Table 2). many examples of strong relations between indicator fig­ Among those species that have fairly low T-values, many ures for sites (releves) and actual measurements of corre­ are characteristic for Boreo-nemoral forests, e.g. Ribes sponding factors at the same sites. The applicability of alpinum (T 4), Geranium sylvaticum (T 4), Equisetum indicator values has also been demonstrated for western pratense (T 4) and E. sylvaticum (T 4), as well as some Europe (e.g. Thompson et al. 1993; van der Maarel 1993) bryophytes, e.g. Rhytidiadelphus triquetrus (T 3) and as well as for different types of vegetation in Scandinavia Th uidium tamariscinum (T 4). (e.g. Vevle & Aase 1980; S. Persson 1981; Borgegard & C-values vary considerably more, between C 3 (sub­ S. Persson 1990; Brunet 1991; Diekmann 1994). oceanic species) and C 6 (sub-continental species). Only In general, Ellenberg's indicator values proved help­ few species show an oceanic distribution (C 2, e.g. Melica ful for the interpretation of differences in ecological con­ uniflora, Lathyrus montanus) or a continental distribution ditions between different community types. The ecologi­ (C 7, Rubus saxatilis, Galium boreale, Equisetum pra­ cal conditions as estimated from the indicator values tense). The C-figuresfor the forest types vary between 3.9 corresponded well with observations and measurements and 4.1 (cf. Table 2), compared with 3.4 for the German in the literature. In this study, there was also good agree­ 'Mittelgebirge' region, which expresses the relative ment between the indicator figures obtained in Chapter 4 continentality of the Boreo-nemoral communities (which and results of the climatic and soil measurements in also have higher climatic continentality indices). Chapter 6 (cf. Diekmann, submitted). This is particularly true for the edaphic factors (moisture, nitrogen and reac­ Species richness and environment tion (or pH) and for light. For the two climatic factors temperature (T) and continentality (C), however, indica­ Some differences in species richness between communi­ tor values did not show a good correspondence with the ties have already been discussed in Chapter 4. Here, climatic variables, except in connection with the regional species richness will be related to soil fertility, which, in comparisons of mesotrophic forests (Table 14) and combination with light, has been shown to be of major eutrophic alder-ash forests (Table 21). The T -values de­ importance for the floristicvariation in deciduous forests. scribe the temperature demands of species on the basis of In contrast to, e.g. grasslands, where species numbers their latitudinal and altitudinal distribution in Europe. The show an unimodal response with peaks at moderately low C-values describe the tolerance of species to large tem­ or intermediate soil fertilities (e.g. Vermeer & Berendse perature fluctuations and severe frosts on the basis of their 1983), in temperate forests the peak is shifted to compara­ longitudinal distribution and distance to the sea. A reason tively high soil fertility. For North American hardwood for the poor performance of T- and C-values might be that forests, a strong positive correlation between species the study area is too small for us to be able to use these number of herbaceous species and pH/cation availability

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 97

Oligotrophic oak forests 50 40 o Vascular plants a Vascular plants b mixed deciduous * Mosotrupt1ic forests Eutrophic elm-ash forests • * Eutrophic alcior-ash forests 35 40 . (f) . Q)(f) Q) 0 . ·� 30 Q) . lA· . . 5} . " 0 30 0 .. . Q) " . z 25 .0E . § � z . 20 . 20

Light figure > S.O

10 15 2 3 5 6 2 3 4 5 6 N trogen figure Nitrogen figure i

50 Vascular plants c 10 d Bryophytes

8 40 � Fig. 35. Correlation between ·u Q)(f) Q) ·o 0 5} 6 0 number of species and nitrogen 5}Q) � 0 30 0 figures for community types. a. "' (]) a, .0 Vascular plants, all 29 commu- § .0 4 "• z § z nity types; b. Vascular plants, 20 . . 14 community types with light � figures > 5.0; c. Vascular plants, Ligh1 Ugure 5.0 * * < 15 community types with light 10 J_ 5.0 5.5 60 6.5 0 fi gures < 5.0; d. Bryophytes, all 2 3 4 5 6 Nitrogen figure Nitrogen figure 29 community types.

of the soil was demonstrated by Peet & Christensen (1988) tions are obtained. However, for the community types and Palmer (1990). A positive correlation between spe­ having light figures lower than 5.0, a strong positive cies richness and pH was also found by Helliwell (1978) correlation is found (r = 0.643, p < 0.01). in a study of some central Swedish forests. In order to Bryophytes do not show any clear trend in species obtain an indication of the relationship between species richness along the fertility gradient (Fig. 35d), nor along richness and fertility in Boreo-nemoral forests, the aver­ gradients for light, moisture and reaction. For bryophytes, age number of vascular plants was plotted against nitro­ other factors may be of more significant importance, such gen figures for all community types distinguished in the as airhumidity, thickness of the litter layer and the rate of releve tables. Fig. 35a shows a peak in diversity at nitro­ its decomposition. gen figuresbetween 5 and 6, corresponding to intermedi­ ate to moderately high, but not very high nutrient levels Species ecology (quadratic regression, correlation coefficient r = 0.598, p < 0.001, N = 29). A linear regression instead gave no A species does not always have the same ecological (r = significant correlation - 0.062). Peet et al. (1990) behaviour in different geographic areas, i.e. its ecological suggest that, along a gradient of decreasing light levels in optimum and amplitude may shift. This might be due to the understorey, the peak in diversity shifts from low to the existence of ecotypes (and thus a change in physi­ high fertility sites. This is indicated by Fig. 35b and c, ological demands), or the presence and absence of possi­ where community types are separated into those with ble competitors. In general, the successful application of light figures above 5.0 and those below 5.0, and where the Ellenberg' s indicator values in different regions of central plots are fitted by quadratic regression. Peaks for nitrogen and northern Europe as previously mentioned has proved figures are at about 4.8 (r = 0.632, p < 0.01, N = 14) and that the majority of species do not shift in their ecological

5.5 (r = 0.828, p < 0.001, N = 15), respectively. Graphs behaviour within these regions. However, with respect to with reaction figures instead of nitrogen figures do not the transition from the Nemoral to the Boreo-nemoral give clear results, despite the close correlation of soil zone, there is strong evidence for a change in biotope for acidity and fertility in Boreo-nemoral deciduous forests at least some species (Vevle & Aase 1980; L. Gustafsson (see below). Plotting the number of vascular plants against 1994; Diekmann, submitted). Luzula pilosa, for example, light figures for all community types, very low correla- which in westerncentral Europe is bound to mesotrophic

Acta Phytogeogr. Suec. 80 98 M. Diekmann

to fairly eutrophic sites, occurs in eastern and northern composition and successional relationships (cf. van der Europe also on very oligotrophic sites, even in pine for­ Maarel 1969). The classificationproposed in this study is ests on sand (cf. Passarge 195 8; Ellenberg 1986; Diekmann believed to reflectma jor discontinuities in the vegetation. 1988). Several species which in central Europe serve as In contrast to other classificationsof deciduous forests in indicators of calcareous soils, such as Campanula the Boreo-nemoral zone (e.g. Kielland-Lund 1971; persicifolia, Polygonatum odoratum and Laserpitium Bergendorff et al. 1979; Anon. 1982; Anon. 1984 ), this latifolium, can in Scandinavia also be found on fairly acid study is based on a fairly large number of releves. In soils (L. Gustafsson 1994; Diekmann 1994 ). Other spe­ addition, it is the firststudy using multivariate analysis. It cies show a change in biotope relative to light or soil does not place particular emphasis on the dominance of moisture (Diekmann, submitted). These differences in single tree species, because, except for oligotrophic oak ecological behaviour deserve more attention, but have not forests, the canopy is almost always composed of a mix­ been systematically or experimentally studied up till now. ture of different trees. Differential species of community types are found in all layers of the vegetation. However, as the total species richness of the forests is mainly deter­ Conclusions mined by the species richness of the field layer, most The Boreo-nemoral forests form species-rich communi­ differential species are found among herbs and grasses. ties, not poorer than corresponding communities in cen­ tral Europe. Although they do not have character species Ordination of community types in relation to light and which not also occur in the Nemoral zone, several taxa soil conditions reach much higher frequencies in Boreo-nemoral forests, both deciduous forest species and species of more open The community types will be discussed with respect to vegetation types. Regarding life forms, no general differ­ their demands on light, soil moisture, soil acidity and ences exist between the two vegetation zones. Ellenberg' s fertility, based on the indicator figures for these factors indicator values, although developed for central Europe, given in Chapter 4. Fig. 36a shows a strong positive can also be applied in northernEurope, particularly with correlation between R- and N-figures (Pearson's product­ respect to the factors light, moisture, nitrogen and reac­ moment correlation, r = 0.940,p < 0.001), indicating that, tion. However, there is a strong indication for a change in in Boreo-nemoral forests, decreasing soil acidity goes biotope for some species. Indicator values for temperature parallel with increasing fertility. This is somewhat differ­ and continentality should only be applied in connection ent from the situation in central Europe, where, for exam­ with regional comparisons between vegetation zones. A ple, beech forests can be found on steep slopes on dry, plant geographic comparison between correspondingfor­ calcareous, but fairly nutrient-poor soils (cf. Ellenberg ests in central and northern Europe reveals that both 1986). In northernEurope, however, similar sites seem to mesotrophic mixed deciduous forests and elm-ash forests be occupied by basiphilous pine forests (cf. Bjfi)rndalen represent southern, relatively continental vegetation types, 1980a, b, 1985). The correlation between L- and M-fig­ which is in accordance with their geographic and ecologi­ ures (Fig. 36f, r 0.305, n.s.) is comparatively weak, =- cal distributions. The total number of species is highest at whereas strong negative correlations exist between L- and r p < intermediate to moderately high soil fertilities; on fairly R-figures (Fig. 36b, =-0.599, 0.001) and particu­ r p < infertile soils, species richness increases with increasing larly L- and N-figures (Fig. 36d, = 0.692, 0.001), - fertility, while, on fairly fertile soils with low light levels, indicating that increasing fertility (connected with de­ it decreases with increasing fertility and decreasing light creasing soil acidity) causes a higher canopy closure (see availability. Chapter 6) which in turncauses a lower light intensity in the interior of the forest. The position of communities relative to moisture and reaction (r = 0.339, n.s.) is shown 7.2 Communities and environment in Fig. 36c, which represents a diagram often used for presenting the ecological distribution of plant communi­ ties (cf. Ellenberg 1986). Only on soils with moderately Classification high to high pH, are deciduous forests found to occupy a This study is not concerned with the debate on the exist­ wide moisture spectrum from moderately dry to moist ence or non-existence of plant communities. However, soils. On very acid sites, they are restricted to mesic forest although a hierarchical classification can be criticized as soils in fairly oceanic areas. On drier or wetter soils and in being artificial, it can still be motivated by practical continental areas, they become replaced by coniferous or purposes, as it facilitates the description of the vegetation deciduous forests of trivial species. A plot of N- and M­ r and of the ecology of species. In general, the combination figures results in a similar diagram (Fig. 36e, = 0.500, of classification and ordination helps to interpret p < 0.01). vegetational variation, e.g. regarding structure, species

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 99

7 .----.-----.-----.-----.----, a 6.0 ,----,------.------.------.------, b

• . * •• • Q 6 * • 5.5 Q • * Q * .. . *) Q) . • 's 5 . � 5Q) *. 2 5.0 • c .* Q) . 0> , . 0 :2 4 • • z * Q "" .,.

Q 4.5 3 Q

2 4.0 '------'------'------'------'--� 3 4 5 6 7 8 3 4 5 6 7 8 Reaction figure Reaction fi ur g e

4 .----.-----.-----.-----.----, c 6.0 d

Q Q Q 5.5 <> <> * . * Q Q • • 5 <> * Q *•• * 'sQ) � <> & ·': · .. -.::0> ·. 5.0 * � . .c . '15 ::J0> -:. "*. "' 2 6 :

4.5

7 4.0 �-�------'------'------'------' 3 4 5 6 7 8 2 3 4 5 6 7 Reaction figure Nitrogen figure

4 ,----,-----,-----.-----.----, e 4 ,-----.---�--�---� f

<> <> • . Q f • # Q . • • 5 • • • <> Q •' •* •* • ...... · . . ·.

Oligotrophic <> oak forests Mesotrophic mixed deciduous * forests Eutrophic elm-ash forests • Eutrophic ald r-ash forests 7 7 � e �--�----�-----L-----L--� 4. 4.0 5 5.0 5.5 6.0 2 3 4 5 6 Light fig ure Nitrogen figure

Fig. 36. Plots of pairs of indicator figuresfor all 29 community types. a. Nitrogen/Reaction; b. Light/Reaction; c. Moisture/Reaction; d. Light!Nitrogen; e. Moisture/Nitrogen; f. Moisture/Light.

Levels of community differentiation and underlying light (oligotrophicnight - mesotrophic/moderately dark ­ environmental factors eutrophic/dark forests), and a moisture gradient (mesic/ moderately moist elm-ash forests- moist/wet alder-ash Fig. 37 shows the differentlevels of community differen­ forests). Related to both nutrient, light and moisture gra­ tiation and their underlying environmental factors. On the dients are also forest climatic factors, such as temperature first level of forest types, the main differentiating factors and evaporation. The importance of nutrients and light for are a complex-gradient in nutrient status, connected with the differentiation of deciduous forests has already been

Acta Phytogeogr. Suec. 80 100 M. Diekmann

Level Community type Fact�r

Forest type Nutrients Oligotrophic Mesotrophic Eutrophic (Light) oak forests (0) mixed dccidu- elm-ash forests (101)

Moisture Eutrophic alder-ash 1) r (1 �o ests Quercus Quercus Quercus Quercus Quercus m Commuoity l us l Geographic/ petraca- robur· robur- robur- robur- glabra·U m minor-U Climatic Frangula Betula Fraxinus Euonymus Tilia Fraxinus Fraxi nus alnus pendula excelsi or europaeus cordata excelsior excelsior (01) c. (00) c. (10000) c. (1000 1) c. (1001) (1010) c. (lOll)l c. lJ:c. Fraxinus Fraxinus Moisture excclsior- excelsior- Prunus Alnus padus glutinosa c. (110) c. (1 11)

Viol� Viola riviaiaoa rivioiaDa NulriCflt.s sub<:omm. sub-comm. (010) (000)

Trientalis Triencalis

ucopaea c:uro�ea s•b-eomm. su.b-comm. (011) (001)

Tilia Ouli.s cordata acelosella LanduR sub-oomm. svb

Geranium Slellaria entiation and underlying environmen­ G.ognoph1<1 sylvalicum bolostea Climatic sub-comm. sub-comm. tal factors. Cluster codes of commu­ (10010) (10011) nity types given in brackets. emphasized in Chapter 6. This is also obvious from the 1 are highly negatively correlated with M-, N- and R­ correlations between environmental variables and releve figures, and positively correlated with L-figures. On this scores on the CA-ordination axes for the whole data set of level, climatic variables are of minor importance; corre­ 367 releves (Fig. 38, Table 35). The releve scores on axis spondingly, they have much weaker correlations with the

2 N (.() X

1

0

-1

-2 -1 0 2

Fig. 38. Ordination diagram with axes 1 and 2 of a Correspondence Analysis of all 367 releves of deciduous hardwood forests in Sweden and Norway. Lines encircle releves of different forest types and communities, indicated by numbers. 1. Oligotrophic oak forests; 2-4. Mesotrophic mixed deciduous forests 2. Quercus robur-Tilia cordata community; 3. Quercus robur-Euonymus europaeus community; 4. Quercus robur-Fraxinus excelsior community; 5. Eutrophic elm-ash forests; 6. Eutrophic alder-ash forests.

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 101

Table 35. Spearman rank correlation (rs) between sample plot 7.3 Geographic distribution of commu­ scores relative on the firsttwo CA-ordination axes (Fig. 38) and nities and the Boreo-nemoral zone 15 environmental variables. All releves included (N 367). For further explanation, see Table 6. = Fig. 39 shows the distribution of localities for releves of Variable Axis 1 Axis 2 different communities. A comparison with Fig. 3 in Chap­ rs p rs p ter 2.2 reveals that deciduous forests generally are con­ L 0.382 0.001 0. 176 0.001 centrated to (1) areas with calcareous bedrock and/or T 0.331 -0.385 0.001 0.001 soils, and (2) to areas close to the Swedish east coast or to c -0.168 0.01 0.083 n.s. M -0.832 0.001 0.501 0.001 the big lakes with a fairly warm and dry climate. Note that R -0.653 0.001 -0.324 0.001 the two areas partly are identical. This distribution pattern N -0.795 0.001 -0.006 n.s. TEM YEAR 0. 184 0.001 -0.142 0.01 applies particularly to mesotrophic mixed deciduous for­ TEM FEB 0. 165 0.01 -0.153 0.01 ests and eutrophic elm-ash forests. Naturally, oligotroph­ TEM JULY -0.020 n.s. -0.126 0.05 ic oak forests on acid soils are not bound to lime, but are PRE YEAR -0.163 0.01 0.423 0.001 PRE VEG 0.062 n.s. 0.234 0.001 concentrated to the relatively oceanic regions of SW and MAR IND -0.182 0.001 0.427 0.001 SE Sweden. Alder-ash forests are not restricted to calcar­ CON IND -0.162 0.01 0.032 n.s. LAT -0.241 0.001 0.160 0.01 eous areas either, as they are mainly topographically LON 0. 133 0.05 0.419 0.001 determined and, at least within the Boreo-nemoral zone, are largely confined to ravines and river valleys, particu­ larly in hilly landscapes of relatively humid areas. Hence, releve scores on axis 1. On the second level of communi­ they can mainly be found in SW Sweden and SE Norway. ties, however, macroclimatic (geographic) factors are the Distribution patterns and community differentiation of primary causes of differentiation, in particular the E-W deciduous forests in southern Scandinavia may help to (humidity) gradient. This gradient is probably partly ex­ obtain a more accurate vegetational characterization of pressed by CA axis 2, as the releve scores on this axis have the Boreo-nemoral zone. Among the four forest types high positive correlations with yearly precipitation, de distinguished, mesotrophic mixed deciduous forests are Martonne's index and longitude. A soil moisture gradient most characteristic of the Boreo-nemoral zone. Within the (M-figures) is the main cause for the differentiation of the Nemoral zone, corresponding sites are occupied by com­ eutrophic alder-ash forests into its two communities. On munities with a widely different species composition (see the third level of sub-communities, several different gra­ Chapter 4.6). In contrast, oligotrophic and eutrophic for­ dients are of importance, including geographic/climatic ests (especially alder-ash forests) occur with similar com­ and fertility factors, as well as anthropogenic factors, i.e. munities also in western and central Europe. Concerning differences in former land use. the oceanic region in western Norway, the ordination diagrams in Figs. 14 and 20 reveal that the forests of this region have a species compositon very different from that Conclusions of Boreo-nemoral forests. The classification of deciduous forests in Boreo-nemoral Regarding the delimitation of the Boreo-nemoral zone Scandinavia reflects the vegetational variation, which in (cf. Chapter 2.1 ), the following conclusions can be drawn: turncorresponds to the variation of climatic, physiographic 1. The elevated plateau in central Smaland is largely and edaphic factors. These underlying environmental fac­ devoid of deciduous hardwood forests (although the hard­ tors have different significance on different levels of wood species are present), which supports the opinion of community differentiation. Comparing community types, Ahti et al. (1968) to consider this region as a southern fertility and soil reaction are strongly positively corre­ outlier of the Boreal zone. 2. The whole of bland should lated with each other, while both factors have a negative be treated as part of the Boreo-nemoral zone, because the correlation with light. A much smaller spectrum of possi­ deciduous forests on the island show all the features of ble combinations of soil moisture and soil reaction is Boreo-nemoral forests as described . above, and because occupied by deciduous forest than in central Europe, since the remnants of deciduous forests on southernmostbland these are partly replaced by coniferous forest. At present, do not substantially deviate from those in other parts of more or less natural mature forests of mixed thermophilous the island. 3. Finally, the oceanic SW and W coast in deciduous trees and conifers can hardly be found. Thus, Norway should, in accordance with Sjors (1963) and the transition from the Nemoral via the Boreo-nemoral to Moen ( 1987), be considered as a separate phytogeo­ the Boreal zone is not accompanied by a gradually in­ graphical unit, here called the Western broad-leaved and creasing abundance of conifers in some of the deciduous pine forest region. forests, but rather by an ecological separation of conifer­ ous and warmth-demanding deciduous forests and a re­ striction of the latter to the most favourable sites.

Acta Phytogeogr. Suec. 80 102 M. Diekmann

Quercus robur-Betula pendula community Ll Ouercus petraea-Frangula alnus community 11o.. Ouercus robur-Fraxinus excelsior community IZl Ouercus robur-E;uonymus europaeus community D Ouercus robur-Tilia cordata community • Ulmus glabra-Fraxinus excelsior community 0 Ulmus minor-Fraxinus excels1or community • Fraxinus excelsior-Prunus padus community 6. Fraxinus community & excelsior-Ainus glutinosa

Fig. 39. Location ofreleves in Sweden and Norway. The lines delimit the bor­ ders of the administrative provinces in a 1-4 relcves a Sweden ('Hin'), which not always cor­ • 5-9 releves >9 rclcves respond to the landscape ('landskap') boundaries.

7.4 Structure and dynamics forests, on the other hand, are, on average, still in a premature state, having a high percentage of Que reus and The deciduous forests in the Boreo-nemoral zone form Fraxinus as remnants of the former land use (see Chapter multi-layered stands, usually with two more or less dis­ 5). These trees create a relatively open canopy, enabling tinct tree layers, a dense shrub layer and fairly species­ the establishment of fairly dense lower tree and shrub rich field and bottom layers. They are in some respects layers. Particularly Corylus could spread after abandon­ different from corresponding forests in central Europe, ing of the wooded meadows where it was abundant. which usually have a comparatively denser canopy and a Besides, during the past decades, the grazing pressure in more open shrub layer_ The reason for this is that many Boreo-nemoral deciduous forests has been comparatively beech and oak/hombeam forests (the two dominating low (although the roedeer population in southernSweden forest types) in central Europe are in a mature state and has considerably increased during later years), enabling structurally not very different from primeval forests (cf. trees and shrubs to regenerate. Ellenberg 1986). Both Fagus and Carpinus form very The population structure of tree species revealed that shady canopies, reducing the light intensity in the interior most Boreo-nemoral deciduous forests are compositionally of the forest to a level too low for many shrub species rather unstable (see Chapter 5). This not only implies a (compare the scattered occurrence of shrubs in Carpinus successional change in the canopy itself, but also changes forests on bland, see Chapter 4.3). In addition, grazing in the species composition of the other layers. Shrubs will by, e.g. roedeer, has caused severe damage to young trees probably decrease in cover due to a decreasing light and shrubs in many Nemoral forests. The Boreo-nemoral intensity caused by, in general, an increased matureness

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 103

and canopy re-closure of the forest stands, and, in particu­ Despite their small size, they are often very species-rich lar, increasing abundance of the shading species Ulmus and serve as refuges of endangered species. However, the glabra and Acer platanoides (at the expense of Quercus small size and isolation imply a high risk of local extinc­ robur and Fraxinus excelsior). Such a decrease was shown tion of taxa due to impoverished populations and a small for Corylus avellana and Crataegus spp. in Dalby chance of re-establishment. Many species have been shown Soderskog in Skfme (Malmer et al. 1978). A closer canopy to be largely restricted to ancient forests which have not also affects the field and bottom layers by increased severely suffered from forestry and animal husbandry shading and interception, changed amount or quality of (Hermy & Stieperaere 1981; Peterken & Game 1984; leaf litter (cf. S. Persson et al. 1987), and a more moderate Dzwonko & Loster 1989, 1990). For the province of forest climate. In Dalby Soderskog, a decrease in the total Skane in Sweden, this was demonstrated for the four 2 number of species and in the number of species per m grasses Hordelymus europaeus, Festuca altissima, Bromus was observed, as well as negative and positive trends for benekenii and B. ramosus (Brunet 1993). Also among individual species (cf. Malmer et al. 1978). Particularly bryophytes, some taxa serve as indicators of long forest species with ecological optima in more open vegetation continuity, e.g. Eurhynchium striatum and Mnium stellare types (e.g. Filipendula ulmaria and Geum rivale) have (Hallingback 1991). The species in question usually have diminished. On the other hand, true forest species may a limited capacity for re-colonization. This sensitivity and increase in abundance, as has been observed for, e.g. low ability to expand into new areas shows the impor­ Bromus benekenii and Cardamine bulbife ra in immature tance of ancient forests with a high continuity as species mesotrophic mixed deciduous forests on bland (Ekstam reservoirs. et al. 1984). Also in the bottom layer, successional changes The risk of extinction of plant species is particularly will occur, and certain species will diminish in the course high in stands subjected to forestry. Selective logging of succession (cf. Sjogren 1964). and, in particular, clear-cutting not only damage plants Most deciduous forests in Boreo-nemoral Scandina­ directly, but also cause changes in the environment, i.e. via are still in a premature state and structurally and changed light and temperature climates, an increased compositionally influenced by former land use, as they nitrification and, when heavy machines are used, distur­ have arisen from wooded meadows, (wooded) pastures bance and condensation of the soil. As a consequence, and arable land. In general, the following dynamic proc­ also the competitive relationships between species are esses can be recognized: an increasing canopy closure, altered. At an early stage, for example, light-demanding, increasing abundance of Ulmus glabra and Acer plata­ nitrophilous species (e.g. Rubus idaeus) may appear or noides, decreasing abundance of Quercus robur and increase in abundance and suppress forest species less Fraxinus excelsior, decreasing cover of shrubs, and adapted to the new environment. This can be observed in changes in field and bottom layers towards a more site­ many parts of the Mittlandsskogen area on bland. specific deciduous forest vegetation. These successional In conclusion, the following recommendations may changes may be counteracted or reinforced by vegetational be made: dynamics due to climatic change (cf. Sykes & Prentice - existing deciduous forest stands should be retained and 1993) or changes in soil properties due to acidification protected, irrespective of their size. This is particularly and nitrogen deposition (e.g. Falkengren-Grerup 1986, important for eutrophic forests. In that way, not only rare 1992). taxa and a high number of species are preserved, but also rare plant communities which do not necessarily contain rare species (e.g. Ulmus glabra-Fraxinus excelsior com­ 7.5 Implications for nature conservation munity). -in stands subjected to forestry on any scale, clear-cutting As a whole, deciduous hardwood forests in the Boreo­ and the use of heavy machinery should be avoided, in nemoral zone have become rare and threatened. This is order not to destroy populations of vulnerable species and particularly true for eutrophic forests: elm-ash forests, their environmental conditions. No sites should be ferti­ once widespread, have to a large extent been cut down and lized or drained (especially important regarding alder-ash transferred into arable land, while alder-ash forests, least forests), and sites subjected to drainage in the past should, widespread among the deciduous forests of this zone, if possible, be restored. have suffered from silviculture and especially drainage. - it is necessary to obtain more information on both The situation is somewhat better for mesotrophic and autecology, population ecology and synecology of spe­ oligotrophic forests which, in some parts of Sweden, have cies, particularly with respect to rare taxa. In order to increased in area since the turn of the century due to follow up the environmental changes and the species abandonment of wooded meadows and ceased grazing. responses to these changes, a system of permanent plots Many deciduous forest stands are very small in area should be established in a wide spectrum of areas and and located far away from other stands of the same type. communities described in this study.

Acta Phytogeogr. Suec. 80 104 M. Diekmann

8 Acknowledgements

First of all, I want to thank my supervisor Erik Sjogren, Fulton and Frank Z. successfully resisted the challenges who was my first scientific contact in Sweden and who of the Swedish forest. The staff of the institute was of convinced me to move to this lovely country for my great help: Folke Hellstrom visited me on bland and took doctoral studies. He helped me in all stages of the work most of the photographs in this volume. Ulla Johansson and read the first version of the thesis. We share a lot of was an effective and helpful secretary, and always willing nice memories: many interesting discussions, several ex­ to have a nice chat. Stefan Bjorklund, Marta Ekdahl and cursions, as well as nice dinner parties at his home in Willi Jungskar supported me in many ways. Thanks to my Uppsala and in his summer house on bland. room mates Ignacy Bekier and Jerry Skoglund who still I am deeply indebted to Eddy van der Maarel, head of talk to me after all these years. Also the staff of the the Department of Ecological Botany in Uppsala. In many Ecological Station on bland (a wonderful place to be) lectures and seminars, he teached me much about ecology made life easier during the many, many weeks of field and introduced me to multivariate analysis, and he also work: thank you, Lennart Agren, Per Stenberg and Jan made many valuable comments on the manuscript. He has Tengo! really created a stimulating working atmosphere, and Without my friends, I had never enjoyed my stay in through his never-failing enthusiasm to make the institute Uppsala as much as I did. Renate Huber moved to Sweden multi-linguistic and multi-cultural (also through enjoy­ at about the same time as me and, from the first moment, able open house events), I got many new friends. became a true friend and companion. Whenever there was I am very grateful to Hugo Sjors, who carefully read a problem, she helped me to solve it. Mats Anefjard the manuscript and eliminated many of my innocent ideas became a good friend; I owe him special thanks for about the vegetation of northern deciduous forests. His teaching me much about the Swedish language. I have a encyclopedic knowledge of the ecology and the vegeta­ lot of wonderful memories together with my friends Helena tion and flora of Scandinavia was of invaluable help. On Runyeon, Karin Gerhardt, Vera Noest, Graciela Rusch, several excursions, I had the pleasure to learn from his Christian Andersson, Martin Sykes and Martin Weih. rich fieldexperience. Many thanks also to my "birder"- and/or"pingis" -fellows I would also like to thank Hartmut Dierschke, who Christer Larsson, Anders Sandstrom and Dan Johansson was my teacher in vegetation science at the University of for good friendship and many hours of exciting excur­ Gottingen, and who turned me from a narrow-minded sions and matches. birdwatcher to an ecologist. He supported me in my I also wish to thank my family, particularly my par­ intention to work in Sweden and he also read parts of the ents: they have supported me with all their heart from the manuscript. very beginning (about 33 years ago), let me move to Kuno Thomasson helped with findingbooks and cor­ Sweden and helped me in many different ways during my recting the reference list. Nigel Rollison conducted the time here. Without them, I had never succeeded in writing linguistic revision. Reno Lottmann made the cover draw­ this up. ing. Jorg Brunet gave valuable comments on the manu­ And last but not least, special thanks to Cecilia, for her script. Vera N oest was fantastic in proof reading. Marijke love, understanding and indispensable practical and emo­ van der Maarel and Joost van der Maarel took care of the tional support during the last period. technical editing of the manuscript. Great thanks to all of them! This work had not been possible without the heroic The study was supported by grants from WWF Sweden, help of a number of field assistants: Patrik Andersson, Sernanders forskningsfond and the University's "ogradu­ Henrik Berg, Camilla Gustafsson, Renate Huber, Mark R. erade forskares anslag".

Acta Phytogeogr. Suec. 80 9 References

Ahti, T., Hamet-Ahti, L. & Jalas, J. 1968. Vegetation zones and skogstyper pa Nedstrand og omkringliggende distrikter. ­ their sections in northwestern Europe. - Ann. Bot. Fenn. 5: Cand. reaL-thesis, Univ. Bergen, 163 pp. 169-211. Balsberg-Pahlsson, A.-M. 1990. Handledning i kemiska metoder Akerlind, L.-0. & Sjogren, E. 1975. Halltorps Hage: Botanisk vid vaxtekologiska arbeten. - Medd. VaxtekoL A vd., Lunds inventering samt forslag till skotseUltgarder. - Medd. Univ. 52: 1-58. Vaxtbiol. Inst. Uppsala 1975 (10): 1-90. Barkman, J.J., Moravec, J. & Rauschert, S. 1986. Code of Almgren, G., lngelOg, T., Ehnstrom, B. & Mortnas, A. 1984. phytosociological nomenclature. - Vegetatio 67: 145-195. Adellovskog. Ekologi och skotsel. - Skogsstyrelsen, Behre, K.-E. 1988. The role of man in European vegetation Jonkoping, 136 pp. history. - Handb. Veg. Sci. 7: 633-672. Almquist, E. 1929. Upplands vegetation och flora. - Acta Bergendorff, C., Larsson, A. & Nihlgard, B. 1979. Sydliga Phytogeogr. Suec. 1: 1-622. lovskogsbestand i Sverige. - Statens naturvardsverk. Rap­ Ammar, M.W. 1978. Vegetation and environment on shore port. SNV PM 1278, Solna, 68 pp. ridges at Vickleby, Oland, Sweden. An analysis. - Acta Berglund, B. 1963. Vegetation pa on Senoren. 11. Land­ Phytogeogr. Suec. 64: 1-94. vegetationen. - Bot. Not. 116: 31-79. Anderson, M.C. 1964a. Studies of the woodland light climate. Berthelsen, B. 1980. Alno-Fraxinetum pa Inre Vestlandet. - 1. The photographic computation of light conditions. - Nor. Vidensk. Selsk. Mus. Rapp. Bot. Ser. 1980 (5): 151- J. EcoL 52: 27-41. 157. Anderson, M.C. 1964b. Studies of the woodland light climate. Birse, E.L. 1982. The main types of woodland in North Scot­ 2. Seasonal variation in the light climate. - J. EcoL 52: 643- land. - Phytocoenologia 10: 9-55. 663. Bj j�jrndalen, J.E. 1980a. Urterike granskoger i Grenland, Anderson, RC., Loucks, O.L. & Swain, A.M. 1969. Herbaceous Telemark. - Blyttia 38: 49-66. response to canopy cover, light intensity, and throughfall Bjj�jrndalen, J.E. 1980b. Kalktallskogar i Skandinavien - ett precipitation in coniferous fo rests. - Ecology 50: 255-263. fOrslagtill klassificering. - Sven. Bot. Tidskr. 74: 103-122. Andersson, C. 1994. Factors affecting regeneration of Quercus Bjj�jrndalen, J.E. 1985. Some synchorological aspects of robur in a temperate deciduous forest. - Doctoral thesis, basiphilous pine forests in Fennoscandia. - Vegetatio 59: Uppsala. 211 -224. Andersson, F. 1970. Ecological studies in a Scanian woodland Bjjljrnstad, A. 1971. A phytosociological investigation of the and meadow area, southern Sweden. 1. Vegetational and deciduous forest types in Sogne, Vest-Agder, South Nor­ environmental structure. - Opera Bot. 27: 1-190. way. - Norw. J. Bot. 18: 191-214. Angstrom, A. 1974. Sveriges klimat. 3. uppL - Generalstabens Bjjljmstad, O.N. 1991. Changes in forest soils and vegetation in Litograf. Anstalts Forlag, Stockholm, 188 pp. Sj�jgne,southern Norway, during a 20 year period. - Holarct. Anon. 1977. Naturgeografisk regionindelning av Norden. - Ecol. 14: 234-244. Nordiska Utredningar B 1977 (34): 1-137. Blackman, G.E. & Rutter, A.J. 1946. Physiological and ecologi­ Anon. 1982. Adellovskog. Forslag till skydd och vard. - Statens cal studies in the analysis of plant environment. I. The light naturvardsverk. Medd. 1982 (9). SNV PM 1587, Solna, 175 factor and the distribution of the bluebell (Scilla non-scripta) pp. in woodland communities. - Ann. Bot. N.S. 10: 361-390. Anon. 1984. Nordiska Ministerradet. Vegetationstyper i Norden. Blom, H.H. 1980. P1antesosiologiske undersj�jkelser av - Berlings, Arlov, 539 pp. edellauvskog og beslektede samfunn pa frisk mark i Ytre Asberg, M. & Steam, W.T. 1973. Linnaeus' Oland and Gotland Hordaland. -Nor. Vidensk. Selsk. Mus. Rapp. Bot. Ser. Journey 1741. - BioL J. Linn. Soc. 5: 1-107. 1980 (5): 134- 150. Atkins, W.R.G., Poole, H.H. & Stanbury, F.A. 1937. The meas­ Blom, H.H. 1982. Edellauvskogssamfunnene i Bergenregionen, urement of the intensity and the colour of the light in woods Vest-Norge. - Cand. reaL-thesis, Univ. Bergen, 102 pp. by means of emission and rectifier photoelectric cells. - Backer, R., Kowarik, I. & Bornkamm, R. 1983. Untersuchungen Proc. Soc. London Ser. B 121: 427-450. zur Anwendung der Zeigerwerte nach Ellenberg. - Verh. Aune, E. I. 1973. Forest vegetation in Hemne Sj�jr-Trondelag, Ges. Okol. 11: 35-56. Western Central Norway. - Norske Vidensk. Selsk. Mus. Borgegard, S.-0. & Persson, S. 1990. Vegetationsgradienteroch Mise. 12: 1-87. skotselresultat i Allekvia lOvang pa Gotland. - Sven. Bot. Austin, M.P. 1977. Use of ordination and other multivariate Tidskr. 84: 87-103. descriptive methods to study succession. - Vegetatio 35: Bradshaw, R.H.W. 1993. Tree species dynamics and distur­ 165-175. bance in three Swedish boreal forest stands during the last Bakkevig, S. 1974. Eikeskog i Ryfylke. Plantesosiologiske og two thousand years. - J. Veg. Sci. 4: 759-764. j�jkologiske undersj�jkelser av eikeskogen og beslektede Bradshaw, R.H.W. & Hannon, G. 1989. The influenceof distur-

Acta Phytogeogr. Suec. 80 106 M. Diekmann

bance on long-term successional processes in Swedish boreal historia och utveckling. - Sven. Skogsvardsfor. Tidskr. 16: forest. - Stud. Plant Ecol. 18: 36-37. 20 1-262. Bradshaw, R.H.W. & Hannon, G. 1992. Climatic change, hu­ Daubenmire, R. 1968. Plant communities. A textbook of plant man influence and disturbance regime in the control of synecology. - Harper & Row Publ., New York, 300 pp. vegetation dynamics within Fiby Forest, Sweden. -J. Ecol. de Martonne, E. 1926. Une nouvelle function climatologique: 80: 625-632. l'indice d'aridite. - Meteorologie 2: 449-458. Bnikenhielm, S. & IngelOg, T. 1972. Vegetationen i Kungsharnn­ Diekmann, M. 1988. Waldgesellschaften auf bland. Diplom­ Morga naturreservat med forslag till skotselplan. - Vaxtekol. arbeit. Georg-August-Univ. Gottingen. - Unpubl., Got­ Stud. 1: 1-94. tingen, 161 pp. Braun-Blanquet, J. 1964. Pflanzensoziologie. 3. Aufl. - Diekmann, M. 1994. En jamforande studie av nordisk Springer-Verlag, Wien, 865 pp. backangsvegetation.-Sven. Bot. Tidskr. 88: 227-236. Brenner, W. 1921. Vaxtgeografiska studier iBarosunds skargard. Diekmann, M. Use and improvement of Ellenberg's indicator -Acta Soc. Fauna Flora Fenn. 49 (5): 1-151. values in deciduous forests of the Boreo-nemoral zone in Brunet, J. 1991. Vegetationen i Skanes alm- och askskogar. ­ Sweden. (submitted) Sven. Bot. Tidskr. 85: 377-384. Dierschke, H. 1992. Zur Begrenzung des Gtiltigkeitsbereiches Brunet, J. 1993. Environmental and historical factors limiting von Charakterarten. Neue Vorschlage und Konsequenzen the distribution of rare forest grasses in south Sweden. - ftir die Syntaxonomie. - Tuexenia 12: 3-11. For. Ecol. Manage. 61: 263-275. Dierschke, H. 1994. Pflanzensoziologie. - Verlag Eugen Ulmer, Brunet, J. 1994. Importance of soil solution chemistry and land Stuttgart, 683 pp. use to growth and distribution of four woodland grasses in Du Rietz, G.E. 1917. Ekskogen vid Borgholm. - Sver. Nat. 8: south Sweden. -Doctoral thesis, Lund. 15-23. Brunet, J. & Neymark, M. 1992. Importance of soil acidity to the Du Rietz, G.E. 1925. Die Hauptztige der Vegetation der Insel distribution of rare forest grasses in south Sweden. -Flora Jungfrun. - Sven. Bot. Tidskr. 19: 323-346. 187: 317-326. Du Rietz, G.E. 1954. Sydviixtberg. - Sven. Bot. Tidskr. 48: Bunce, R.G.H. 1982. A field key for classifying British wood­ 174- 187. land vegetation. Part 1. Institute of Terrestrial Ecology, Dzwonko, Z. & Loster, S. 1989. Distribution of vascular plant Cambridge, 103 pp. species in small woodlands on the WesternCarpathian foot­ Carlsson, T. & Clemedson, C.-J. 1989. BraWn - en o med hills. - Oikos: 56: 77-86. vardefull lundflora i Sodermanland. - Sven. Bot. Tidskr. Dzwonko, Z. & Loster, S. 1990. Vegetation differentiation and 83: 283-295. secondary succession on a limestone hill in southernPoland. Cedercreutz, C. 1927. Studien tiber Laubwiesen in den Kirch­ - J. Veg. Sci. 1: 615-622. spielen KyrksHitt und Esbo in Stidfinnland. - Acta Bot. Eber, W. 1971. The characterization of the woodland light Fenn. 3: 1-181. climate. - Ecol. Stud. 2: 143-152. Conrad, V. 1946. Usual formulas of continentality and their Eber, W. 1972. Uber das Lichtklima von Waldernbei Gottingen limits of validity. - Trans. Am. Geophys. Union 27: 663- und seinen EinfluB auf die Bodenvegetation. - Scr. Geobot. 664. 3: 1-150. Coombe, D.E. 1966. The seasonal light climate and plant growth Ehrlen, J. 1992. Proximate limits to seed production in a herba­ in a Cambridgeshire wood. - Brit. Ecol. Soc. Symp. 6: 148- ceous perennial legume, Lathyrus vernus. - Ecology 73: 166. 1820-1831. Corley, M.F.V., Crundwell, A.C., Dtill, R., Hill, M.O. & Smith, Ekstam, U. 1979. Detaljerad vegetationskartering av kustland­ A.J .E. 1981. Mosses of Europe and the Azores; an annotated skapet mellan Halltorp och Solliden pa bland. - Calidris 8: list of species, with synonyms from the recent literature. - 269-3 18. J. Bryol. 11: 609-689. Ekstam, U. & Martinsson, I. 1981. Bodakronopark. Naturinven­ Dahl, E. 1951. On the relation between summer temperature and tering. - Lansstyrelsen i Kalmar lan informerar 1981 ( 4): 1- the distribution of alpine vascular plants in the lowlands of 77. Fennoscandia. - Oikos 3: 22-52. Ekstam, U. & Sjogren, E. 1973. Studies of past and present Dahl, E. 1992. Relations between macro-meteorological factors changes in deciduous forest vegetation on bland. - Zoon and the distribution of vascular plants in northern Europe. - Suppl. 1: 123-135. Univ. Trondheim Vitensk. Mus. Rapp. Bot. Ser. 1992 (1): Ekstam, U., Jacobson, R., Mattson, M. & Porsne, T. 1984. 31-59. Glands och Gotlands vaxtvarld. - Bokforlaget Natur och Dahl, E. & Mork, E. 1959. On the relationship between tempera­ Kultur, Stockholm, 336 pp. ture, respiration and growth in Norway spruce (Picea abies Ellenberg, H. 1939. Ober Zusammensetzung, Standort und Stoff­ (L.) Karst.). - Medd. Nor. Skogfors0ksves. 53: 81-93. produktion bodenfeuchter Eichen- und Buchen-Mischwald­ Dahl, E., Gj ems, 0. & Kielland-Lund, J. 1967. On the vegetation gesellschaften Nordwestdeutschlands. - Mitt. Florist.­ types of Norwegian conifer forests in relation to the chemi­ Soziol. Arbeitsgem. 5: 1-135. cal properties of the humus layer. - Medd. Nor. Ellenberg, H. 1956. Aufgaben und Methoden der Vegetations­ Skogfors0ksves. 85: 503-531. kunde. Grundlagen der Vegetationsgliederung. 1. Teil. In: Danielsson, U. 1917. A venboken pa bland. - Sven. Skogsvards­ Waiter, H. (ed.) Einftihrung in die Phytologie, Band 4. ­ fOr. Tidskr. 15: 833-839. Verlag Eugen Ulmer, Stuttgart, 136 pp. Danielsson, U. 1918. Anteckningar om de oHindska skogarnas Ellenberg, H. 1986. Vegetation Mitteleuropas mit den Alpen in

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 107

okologischer Sicht. 4. Aufl. - Verlag Eugen Ulmer, Stutt­ Acta Phytogeogr. Suec. 50: 269-280. gart, 989 pp. Froman, I. 1946. Vaxterna pa Stora Karlso. In: Pettersson, B. & Ellenberg, H., Weber, H.E., Diill,R., Wirth, V., Werner, W. & Curry-Lindahl, K. (eds.) Natur pa Gotland. - BokfOrlaget PauliBen, D. 1991. Zeigerwerte von Pflanzenin Mitteleuropa. Svensk Natur, Stockholm, pp. 139-149. - Scr. Geobot. 18: 1-248. Gatsuk, L.E., Smirnova, O.V., Vorontzova, L.l., Zaugolnova, Ericson, L. & Wallentinus, H.-G. 1979. Sea-shore vegetation L.B. & Zhukova, L.A. 1980. Age states of plants of various around the Gulf of Bothnia. - Wahlenbergia 5: 1-142. growth forms: a review. - J. Ecol. 68: 675-696. Eriksson, B. 1982. Data rorande Sveriges temperaturklimat. - Geiger, R. 1961. Das Klima der bodennahen Luftschicht. 4. SMHI Reports Meteor. Climat. RMK 39, Norrkoping, 34 Aufl. - Vieweg & Sohn, Braunschweig, 646 pp. pp. Glenn-Lewin, D.C. & van der Maarel, E. 1992. Patterns and Eriksson, B. 1983. Data rorande Sveriges nederbordsklimat. processes of vegetation dynamics. In: Glenn-Lewin, D.C., Normalviirden for perioden 1951-80. - SMHI Rapport Peet, R.K. & Veblen, T.T. (eds.) Plant succession. Theory 1983: 28, Norrkoping, 92 pp. and prediction. - Chapman & Hall, London, pp. 11-59. Ernst, W.H.O. 1979. Population biology of Allium ursinum in Goff,F.G. 1968. Use of size stratificationand differentialweight­ northern Germany. - J. Ecol. 67: 347-362. ing to measure forest trends. - Am. Midi. Nat. 79: 402-418. Evans, G.C. 1956. An area survey method of investigating the Grolle, R. 1983. Hepatics of Europe including the Azores; an distribution of light intensity in woodlands, with particular annotated list of species, with synonyms from the recent reference to sunflecks. - J. Ecol. 44: 391-428. literature. - J. Bryol. 12: 403-459. Evans, G.C. 1966. Model and measurement in the study of Gustafsson, C. 1992. Small-scale variation in the field layer of woodland light climates.-Brit. Ecol. Soc. Symp. 6: 53-76. deciduous forests in relation to light and soil acidity. Report Falinski, J.B. 1986. Vegetation dynamics in temperate lowland Dept. of Ecological Botany, Uppsala Univ. - Unpubl., primeval forests. - Geobotany 8: 1-537. Uppsala, 27 pp. Falkengren-Grerup, U. 1986. Soil acidification and vegetation Gustafsson, L. 1994. A comparison of biological characteristics changes in deciduous forests in southern Sweden. - and distribution between Swedish threatened and non-threat­ Oecologia (Berl.) 70: 339-347. ened forest vascular plants. - Ecography 17: 39-49. Falkengren-Grerup, U. 1987. Long-term changes in pH of forest Hregstrom, C.-A. 1983. Vegetation and soil of the wooded soils in southernSweden. - Environ. Pollut. 43: 79-90. meadows in Nato, Aland. - Acta Bot. Fenn. 120: 1-66. Falkengren-Grerup, U. 1992. Mark- och florafodindringar i Halden, B.E. 1950. Korpbergetvid Viksberg i Salem, Soderman­ sydsvensk adellovskog. - Statens naturvardsverk, Solna, land, jamte nagra ord om biotoperna for Asplenium ruta­ 97 pp. muraria och hassel. - Sven. Bot. Tidskr. 44: 535-550. Fekete, G. 1974. Tolgyesek relatfv megvibigftasa es gyepszint­ Hallbacken, L. & Tamm, C.O. 1986. Changes in soil acidity fajainak eloszlasa (Relative light intensity and distribution from 1927 to 1982-1984 in a forest area of South-West of herb layer species in oakwoods). - Stud. Bot. Hung. 9: Sweden. - Scand. J. For. Res. 1: 219-232. 87-96. Hallingback, T. 1991. Mossor som indikerar skyddsvard skog. Forster, M. 1981. Waldgesellschaften der Biickeberge. - - Sven. Bot. Tidskr. 85: 321-332. Tuexenia 1: 213-231. Havas, P.J. 1967. Zur Okologie der Laubwalder, insbesondere Fottland, H. 1980. Rik lauvskog i Midre Hardanger. - Nor. der Grauerlenwalder, an der Kiiste der Bottenwiek. - Aquilo. Vidensk. Selsk. Mus. Rapp. Bot. Ser. 1980 (5): 127-133. Ser. Bot. 6: 314-346. Fralish, J.S. 1988. Predicting potential stand composition and Heckert, L. 1959. Die klimatischen Verhaltnisse in Laubwaldern. site characteristics in the Shawnee Hills forest of Illinois. - - Z. Meteorol. 13: 211-223. Am. Midi. Nat. 120: 79-101. Heikkinen, R.K. 1991. Multivariate analysis of esker vegetation Fransson, S. 1965. The Borderland. - Acta Phytogeogr. Suec. in southern Hame, S Finland. - Ann. Bot. Fenn. 28: 20 1- 50: 167-175. 224. Fremstad, E. 1979. Phytosociological and ecological investiga­ Heinken, T. 1993. Phytosociological and historical investiga­ tions of rich deciduous forests in Orkladalen, Central Nor­ tions in beech forests and birch-oak forests on pleistocene way. - Norw. J. Bot. 26: 111-140. sandy soils without ground water influencein Lower Saxony Fremstad, E. 1983. Role of Black Alder (Alnus glutinosa) in (NWGerman y). - Scr. Geobot. 21: 61-66. vegetation dynamics in West Norway. - Nord. J. Bot. 3: Helliwell, D.R. 1978. Floristic diversity in some central Swed­ 393-410. ish forests. - Forestry 51: 151-161. Fremstad, E. & Normann, 0. 1982. Inventering av rik li21vskog i Hermy, M. & Stieperaere, H. 1981. An indirect gradient analysis Troms. - Tromura: 1-97. of the ecological relationships between ancient and recent Fremstad, E. & 0vstedal, D.O. 1978. The phytosociology and riverine woodlands to the south of Bruges (Flanders, Bel­ ecology of Grey Alder (Alnus incana) forests in central gium). - Vegetatio 44: 43-49. Troms, North Norway. - Astarte 11: 93-112. Hill, M.O. 1979. TWINSPAN. A Fortran program for arranging Fries, M. 1948. Limes Norrlandicus-studier. En vaxtgeografisk multivariate data in an ordered two-way table by classifica­ gransfraga historiskt belyst och exemplifierad. - Sven. tion of the individuals and attributes. - Cornell Univ., Bot. Tidskr. 42: 51-69. Ithaca, New York, 49 pp. Fries. M. 1958. Skogslandskapet pa Sotenas och Stiingenas i Hinneri, S. 1972. An ecological monograph on eutrophic de­ Bohuslan under historisk tid. - Geographica 35: 1-105. ciduous woods in the SW archipelago of Finland. - Ann. Fries, M. 1965. The Late-Quaternary vegetation of Sweden. - Univ. Turku. Ser. A 11. Biol.-Geogr.-Geol. 50: 1-131.

Acta Phytogeogr. Suec. 80 108 M. Diekmann

Hintikka, V. 1963. Uber das GroBklima einiger Pflanzenareale omradet bvetorp-Back-Vanserum pa mellersta bland. - in zwei Klimakoordinatensystemen dargestellt. - Ann. Bot. Lansstyrelsen i Kalmar lan informerar 1985 (12): 1-82. Soc. Zool. Bot. Fenn. 'Vanamo' 34 (5): 1-64. J ohansson, S. 1955. Borga Hage med Solliden. In: Sterner, R. & Hofgaard, A. 1993a. Structure and regeneration patterns in a Curry-Lindahl, K. (eds.) Natur pa bland. - Bokforlaget virgin Picea abies forest in northern Sweden. - J. Veg. Sci. Svensk Natur, Stockholm, 214-226. 4: 601-608. Jones, E.W. 1945 . The structure and reproduction of the virgin Hofgaard, A. 1993b. 50 years of change in a Swedish boreal old forest ofthe North Temperate zone. - New Phytol. 44: 1 30- growth Picea abies forest. - J. Veg. Sci. 4: 773-782. 148. Hofmann, G. 1958. Die eibenreichen Waldgesellschaften Jongman, R.H.G., ter Braak, C.F.R. & van Tongeren, O.F.R. Mitteldeutschlands. - Arch. Forstwes. 7: 502-558. (eds.) 1987. Data analysis in community and landscape Hofmeister, H. 1990. Die Waldgesellschaften des Hildesheimer ecology. - PUDOC, Wageningen, 299 pp. Waldes. - Tuexenia 10: 443-473. Julin, E. 1948. Vessers udde. Mark och vegetation i en Holten, J.I. 1993. Potential effects of climate change on distribu­ igenvaxande 16vang vid Bjarka-Saby. - Acta Phytogeogr. tion of plant species, with emphasis on Norway. In: Holten Suec. 23: 1-186. J.I., Paulsen, G. & Oechel, W.C. (eds.) Impacts of climatic Julin, E. 1965. The north-east corner. Terrestrial vegetation and change on natural ecosystems, with emphasis on boreal and flora. - Acta Phytogeogr. Suec. 50: 205-209. arctic/alpine areas. - NINA, Trondheim, pp. 84- 104. Kalda, A. 1960. Eesti NSV laialehiste lehtmetsade taimkate Hulden, E. 1941. Studien iiber Fraxinus excelsior L. - Acta (Die Pflanzendecke der Edellaubwalder der Estnischen SSR). Bot. Fenn. 28: 1-250. - Tartu Riikliku Ulik. Toim. 93: 123-155. Hulten, E. 1971. Atlas over vaxternas utbredning i Norden. 2. Kers, L.E. 1978. Rido-arkipelagen i Malaren. Naturinventering. uppl. - Generalstabens Litogr. Anstalts Forlag, Stockholm, Del 1: Vegetation och flora. - Statens naturvardsverk. 531 pp. Rapport. SNV PM 1071, Solna, 142 pp. Huntley, B. 1988. Glacial and Holocene vegetation history - Kielland-Lund, J. 1967. Zur Systematik der KiefernwalderFenno­ 20 ky to present. Europe. - Handb. Veg. Sci. 7: 341-383. scandiens. - Mitt. Florist.-Soziol. Arbeitsgem. N.F. 1 1112: Hustich, I. 1960. Plant geographical regions. In: Somme, A. 127-141. (ed.) A geography of Norden. - J.W. Cappelens Forlag, Kielland-Lund, J. 1971. A classification of Scandinavian forest Oslo, pp. 54-62. vegetation for mapping purposes. In: IBP i Norden 7. - Hytteborn, H. 1975. Deciduous woodland at Andersby, Eastern Universitetsforlaget, Oslo, pp. 13-43. Sweden. Above-ground tree and shrub production. -Acta Kielland-Lund, J. 1973. Komentarer til "A classification of Phytogeogr. Suec. 61: 1-96. Scandinavian forest vegetation for mapping purposes - Hytteborn, H. 1986. Methods of forest dynamics research. In: IBP i Norden 7: 13-43". In: IBP i Norden 11.-Universitets­ Fanta, J. (ed.) Forest dynamics research in Western and forlaget, Oslo, pp 33-38. Central Europe. - PUDOC, Wageningen, pp. 17-31. Kielland-Lund, J. 1981. Die Waldgesellschaften SO-N orwegens. Hyttebom, H., Packham, J.R. & Verwijst, T. 1987. Tree popula­ - Phytocoenologia 9: 53-250. tion dynamics, stand structure and species composition in Kielland-Lund,J. 1994. Syntaxonomy of Norwegian forest veg­ the montane virgin forest of Vallibacken, northern Sweden. etation 1993. - Phytocoenologia 24: 299-310. - Vegetatio 72: 3-19. Klokk, T. 1980. River bank vegetation along lower parts of the Ingel6g, T. 1981. Floravard i skogsbruket. Del 1 - Allman del. River Gaula, Orkla and Stj0rdalselva, Central Norway. ­ - Skogsstyrelsen, Jonkoping, 153 pp. Nor. Vidensk. Selsk. Skr. 1980 (4): 1-71. Ingelog, T., Thor, G. & Gustafsson, L. 1984. Floravard i Klokk, T. 1982. Mire and forest vegetation from Klaebu, Central skogsbruket. Del 2-Artdel. - Skogsstyrelsen, Jonkoping, Norway. - Gunneria 40: 1-71. 408 pp. Klotzli, F. 1970. Eichen-, Edellaub- und Bruchwalder der Inghe, 0. & Tamm, C.O. 1985. Survival and flowering of Britischen Inseln. - Schweiz. Z. Forstwes. 121: 329-366. perennial herbs. IV. The behaviour of Hepatica nobilis and Klotzli, F. 1975a. Edellaubwalder im Bereich der stidlichen Sanicula europaea on permanent plots during 1943-1981. Nadelwalder Schwedens. - Ber. Geobot. Inst. Eidg. Techn. - Oikos 45: 400-420. Hochsch. Stift. Rtibel 43: 23-53. Ivarsson, R. 1962. Lovvegetationen i Mollosunds socken. - Klotzli, F. 1975b. Zum Standort von Edellaubwaldernim Bereich Acta Phytogeogr. Suec. 46: 1-197. des stidlichen borealen Nadelwaldes. - Mitt. Eidg. Anst. lversen, J. 1944. Viscum, Hedera and /lex as climate indicators. Forstl. Versuchswes. 51: 49-64. - Geol. For. Stockh. Forh. 66: 463-483. Koponen, T. 1967. On the dynamics of vegetation and flora in Iversen, J. 1958. Pollenanalytischer Nachweis des Relikten­ Karkali Nature Reserve, southern Finland. - Ann. Bot. charakters eines jtitischen Linden-Mischwaldes. - Veroff. Fenn. 4: 121-218. Geobot. Inst. Eidg. Techn. Hochsch. Stift. Rtibel 33: 137- Korsmo, H. 1974. 0stfold, Akershus, Hedmark og Oppland. 144. N aturvernradetslands plan for edellauvskogreservater iNorge Jalas, J. 1957. Die geobotanische Nordostgrenze der sogenannten 1. - As-NLH, mimeogr., As, 111 pp. Eichenzone Stidwestfinnlands.-A nn. Bot. Soc. Zool. Bot. Kowarik, I. & Seidling, W. 1989. Zeigerwertberechnungen nach Fenn. 'Vanamo' 29 (5): 1-32. Ellenberg - Zu Problemen und Einschrankungen einer Johansson, 0. 1982. Vegetationen i Hagelstads naturreservat. ­ sinnvollen Methode. - Landschaft Stadt 21: 132- 143. Calidris 11: 263-275. Kuusipalo, J. 1985. An ecological study of upland forest site Johansson, 0. 1985. Vegetation och markanvandning inom classification in southernFinland. - Acta For. Fenn 192: 1-

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 109

72. geologi. 4. uppl. - Svenska Bokforlaget/Norstedts, Stock­ Landolt, E. 1977. Okologische Zeigerwerte zur Schweizer Flora. holm, 698 pp. - Veroff. Geobot. Inst. Eidg. Techn. Hochsch. Stift. Ri.ibel Makirinta, U. 1968. Haintypenuntersuchungen im mittleren Stid­ 64: 1-208. Hame, Stidfinnland. - Ann. Bot. Fenn. 5: 34-64. Larcher, W. 1980. Physiological plant ecology. 2nd ed. - Makirinta, U. 1982. Zur Struktur der stidfinnischenHainwalder. Springer-Verlag, Berlin, 303 pp. In: Dierschke, H. (ed.) Struktur und Dynamik von Waldern. Larsson, A. 1974. Ottenby Naturreservat. Vegetation och Ber. Intern. Symp. IVV 1981. -J. Cramer, Vaduz, pp. 37 1- markanvandning. - Medd. Avd. Ekol. Bot. Lunds Univ. 2 382. (6): 1-70. Makirinta, U. 1990. Validation of the Finnish forest types by Larsson, K. 1982. Den hallandska adellovskogen. - Halland. numerical methods illustrated by means oftwo sets of releve Nat. 46 (2): 4-1 1. data. - Vegetatio 88: 143-1 50. Leemans, R. 1992. Simulation and future projection of succes­ Malmer, N. 1974. Scandinavian approach to the vegetation sion in a Swedish broad-leaved forest. - For. Ecol. Man­ science. - Medd. Avd. Ekol. Bot. Lunds Univ. 2 (1): 1-32. age. 48: 305-3 19. Malmer, N., Lindgren, L. & Persson, S. 1978. Vegetation suc­ Leibundgut, H. 1982. Europaische Urwalder der Bergstufe. - cession in a South Swedish deciduous forest. - V egetatio Verlag Paul Haupt, Bern, 308 pp. 36: 17-29. Leuschner, C., Rode, M. & Heinken, T. 1993. Gibt es eine Malmstrom, C. 1937. Tonnersjohedens fOrsokspark i Halland. Nahrstoffmangel-Grenze der Buche im nordwestdeutschen - Medd. Statens SkogsfOrsoksanst. 30: 323-528. Flachland? - Flora 188: 239-249. Malmstrom, C. 1939. Hallands skogar under de senaste 300 Lindgren, L. 1970. Beech forest vegetation in Sweden - a aren. - Medd. Stat. Skogsforsoksanst. 31: 171-300. survey. - Bot. Not. 123: 401 -424. Mascher, J.W. 1977. Spontana fOrekomster av adla lOvtrad i Lindgren, L. 1975. Beech forest vegetation and soil in Sweden. Angermanland. - Sven. Bot. Tidskr. 71: 315-325. In: Dierschke, H. (ed.) Vegetation und Substrat. Ber. Intern. Matuszkiewicz, W. 1962. Zur Systematik der nattirlichen Symp. IVV 1969. -1. Cramer, Vaduz, pp. 401 -421. Kiefernwalderdes mittel- und osteuropaischen Flachlandes. Lindquist, B. 1931. Den skandinaviska bokskogens biologi. ­ -Mitt. Florist.-Soziol. Arbeitsgem. N.F. 9: 145-186. Sven. Skogsvardsfor. Tidskr. 29: 179-532. Matuszkiewicz, W. 1977. Spat- und Frtihfroste als standorts­ Lindquist, B. 1932. Om den vildvaxande skogsalmens raser och okologischer Faktor in den Waldgesellschaften des Bialo­ deras utbredning i Nordvasteuropa. - Acta Phytogeogr. wieza-Nationalparkes (Polen). In: Dierschke, H. (ed.) Veg­ Suec. 4: 1-56. etation und Klima. Ber. Intern. Symp. IVV 1975. - J. Lindquist, B. 1934. Den svenska lOvskogen. - Sver. Nat. 1934: Cramer, Vaduz, pp. 195-233. 13-36. McCook, M.J. 1994. Understanding ecological community suc­ Lindquist, B. 1938. Dalby Soderskog. En sk{msk lovskog i cession: Causal models and theories, a review. - Vegetatio forntid och nutid. - Acta Phytogeogr. Suec. 10: 1-273. 110: 115-147. Linkola, K. 1916. Studien tiberden Einflussder Kultur auf die Moen, A. 1987. The regional vegetation of Norway; that of Flora in den Gegenden nordlich vom Ladogasee. I. Central Norway in particular. - Nors. Geogr. Tidskr .. 41: Allgemeiner Teil. - Acta Soc. Fauna Flora Fenn. 45 ( 1): 1- 179-225. 432. Moller, H. 1979. Das Chrysosplenio oppositifolii-Alnetum Linkola, K. 1929. Zur Kenntnis der Waldtypen Eestis. - Acta glutinosae (Meij. Drees 1936), eine neue Alno-Padion­ For. Fenn. 34: 1-73. Assoziation. - Mitt. Florist.-Soziol. Arbeitsgem. N.F. 21: Linkola, K. 1930. Uber die Halbhainwalder in Eesti. - Acta 167-188. For. Fenn. 36: 1-30. Mueller-Dombois, D. & Ellenberg, H. 1974. Aims and methods Linnaeus, C. 1745. Carl Linnaei - Olandska och gothlandska of vegetation ecology. - John Wiley & Sons, New York, resa - fOrrattad Ahr 1741. Stockholm & Uppsala [Transl. 547 pp. by Asberg, M. & Steam, W.T. 1973. Linnaeus's bland and Nageli, W. 1940. Lichtmessungen im Freiland und in geschlos­ Gotland journey 1741.-Biol. 1. Linn. Soc. 5: 1-107.]. senen Altholzbestanden. - Mitt. Schweiz. Anst. Forstl. Lippmaa, T. 1935. Une Analyse des Forets de L'ile Estonienne Versuchswes. 21: 250-306. D'Abruka (Abro) sur la base des associations unistrates. ­ Neuhausl, R. 1969. Systematisch-soziologische Stellung der Acta Comm. Univ. Tart. A 28(1): 1-97. baumreichen Hochmoorgesellschaften Europas. - Vegetatio Lippmaa, T. 1940. A contribution to the ecology of the Estonian 17: 104-121. deciduous forest. - Ann. Acad. Sci. Est. 1: 30-85. Nihlgard, B. 1969. The microclimate in a beech and a spruce Loucks, O.L. 1962. Ordinating forest communities by means of forest (a comparative study from Kongalund, Scania, Swe­ environmental scalars and phytosociological indices. - den). - Bot. Not. 122: 333-352. Ecol. Monogr. 32: 137-166. Norden, U. 1992. Soil acidificationand element fluxes as related Lundqvist, G. 1959. Description to accompany the map of the to tree species in deciduous forests of south Sweden. - Quaternarydeposits of Sweden. - Sver. Geol. Unders. Ser. Doctoral thesis, Lund. Ba 17: 1-116. Nordseth, K. 1987. Climate and hydrology ofNorden. In: Varjo, Lundqvist, J. 1968. Plant cover and environment of steep hill­ U. & Tietze, W. (eds.) Norden. Man and environment. ­ sides in Pite Lappmark. - Acta Phytogeogr. Suec. 53: 1- Gebr. Borntrager, Berlin, pp. 120-128. 153. N y lander, C.-E. 197 5. V egetationshistoria och vegetation i sodra Magnusson, N.H., Lundqvist, G. & Regnell, G. 1963. Sveriges Brakne-Hoby, Blekinge. - Medd. Avd. Ekol. Bot. Lunds

Acta Phytogeogr. Suec. 80 110

Univ. 3 (1): 1-213. Peet, R.K., van der Maarel, E., Rosen, E., Willems, J., Norquist, Oberdorfer, E. 1983. Pflanzensoziologische Exkursionsflora. 5. C. & Walker, J. 1990. Mechanisms of coexistence in spe­ Aufl. - Verlag Eugen Ulmer, Stuttgart, 1051 pp. cies-rich grasslands. - Bull. Ecol. Soc. Am. 71 (2). Supple­ Oberdorfer, E. 1992. Si.iddeutsche Pflanzengesellschaften. Teil ment: p. 283. 4: Walder und Gebiische.2. Aufl. - Gustav Fischer Verlag, Persson, H. 1975. Deciduous woodland at Andersby, Eastern Jena, 282 pp. (Textband) 580 pp. (Tabellenband) Sweden: Field-layer and below-ground production. - Acta + Odland, A. 1981. Pre- and subalpine tall herb and fern vegeta­ Phytogeogr. Suec. 62: 1-71. tion in Ri��ldal, W Norway. - Nord. J. Bot. 1: 67 1-690. Persson, S. 1980. Succession in a south Swedish deciduous Odland, A. 1991. On the ecology of Th elypteris limbosperma ­ wood: a numerical approach. - Vegetatio 43: 103-122. a synecological investigation of T. limbosperma-dominated Persson, S. 1981. Ecological indicator values as an aid in the stands in W Norway. - Nord. J. Bot. 10: 637-659. interpretation of ordination diagrams. - J. Ecol. 69: 71-84. Odland, A. 1992. A synecological investigation of Matteuccia Persson, S., Malmer, N. & Wallen, B. 1987. Leaf litter fall and struthiopteris-dominated stands in Western Norway. - soil acidity during half a century of secondary succession in Vegetatio 102: 69-95. a temperate deciduous forest. - Vegetatio 73: 31-45. Odland, A., Birks, H.J.B. & Line, J.M. 1990. Quantitative Perttula, U. 1950. Kasvillisuudesta ylisella Syviirillaseka siihin vegetation-environment relationships in west Norwegian etelassa rajoittuvalla Juksovon seudulla (Uber die Vegeta­ tall-fernvegetation. - Nord. J. Bot. 10: 511-533. tion am oberen Lauf des Flusses Swir nebst der im Si.iden 0kland, R.H. 1990. Vegetation ecology: theory, methods and anschliessenden Gegend von Juksowo). - Ann. Bot. Soc. applications with reference to Fennoscandia. - Sommerfeltia Zool. Bot. Fenn. 'Vanamo' 23 (6): 1-204. Suppl. 1: 1-233. Peterken, G.F. 1981. Woodland conservation and management. 0kland, R.H. & Eilertsen, 0. 1993. Vegetation-environment - Chapman & Hall, London, 328 pp. relationships of boreal coniferous forests in the Solhomfjell Peterken, G.F. & Game, M. 1984. Historical factors affecting area, Gj erstad, S Norway. - Sommerfeltia 16: 1-254. the number and distribution of vascular plant species in the 0kland, T. 1988. An ecological approach to the investigation of woodlands of Central Lincolnshire. - J. Ecol. 72: 155-1 82. a beech forest in Vestfold, SE Norway. - Nord. J. Bot. 8: Pettersson, B. 1946. Gotlandskt ange. In: Pettersson, B. & 375-407. Curry-Lindahl, K. (eds.) Natur pa Gotland. - Bokforlaget 0kland, T. 1990. Vegetational and ecological monitoring of Svensk Natur, Stockholm, pp. 165-177. boreal forests in Norway. I. Rausjlllmarka in Akershus county, Pettersson, B. 1958. Dynamik och konstans i Gotlands flora och SE Norway. - Sommerfeltia 10: 1-52. vegetation. - Acta Phytogeogr. Suec. 40: 1-288. Oksanen, J. 1983. Vegetation of forested inland dunes in North Pigott, C.D. 1975. Natural regeneration of Tilia cordata in Karelia, eastern Finland. - Ann. Bot. Fenn. 20: 28 1 -295. relation to forest-structure in the forest of Bialowieza, Po­ Olsson, H. 1974. Studies on South Swedish sand vegetation. ­ land. - Trans. R. Soc. Lond. B. Biol. Sci. 270: 151-179. Acta Phytogeogr. Suec. 60: 1-170. Pons, T.L. 1983. An ecophysiological study in the field layer of Olsson, H. 1975. Acidophilous oak forests in South Sweden. ­ ash coppice. - Thesis, Utrecht Univ., 128 pp. Coli. Phytosociol. 3: 261 -27 1. Pott, R. 1992. Die Pflanzengesellschaften Deutschlands. - Ottosson, I. 1965. Woods on the isle of Jungfrun. - Acta Verlag Eugen Ulmer, Stuttgart, 427 pp. Phytogeogr. Suec. 50: 141-143. Prentice, I.C. & Helmisaari, H. 1991. Silvics of north European 0vstedal, D.O. 1985. The vegetation of Lindas and Austrheim, trees: compilation, comparisons and implications for forest Western Norway. - Phytocoenologia 13: 323-449. succession modelling. - For. Ecol. Manage. 42: 79-93. Pahlsson, L. 1969. Vegetation, microclimate and soil moisture Raunkiaer, C. 1934. The life form of plants and statistical plant of beechwood and open pasture land on the esker Knivsas, geography. - Oxford University Press, Oxford, 632 pp. Central Scania. - Oikos Suppl. 12: 87-103. Rodwell, J.S. (ed.) 1991. British plant communities. Vol. 1. Palmer, M.W. 1990. Spatial scale and patterns of vegetation, Woodlands and scrub. - Cambridge University Press, Cam­ flora and species richness in hardwood forests of the North bridge, 395 pp. Carolina piedmont. - Coenoses 5: 89-96. Ri��sberg, I. & 0vstedal, D.O. 1987. Phytosociology and soil Palmgren, A. 1915-1917. Studier Ofver Lofangsomradena pa properties of Corylus avellana coppices on the coast of Aland. 1-111. - Acta Soc. Fauna Flora Fenn. 42 (1): 1-633. western Norway. - Nord. J. Bot. 7: 169- 185. Parker, J. 1952. Environment and forest distribution of the Rudberg, S. 1987. Geology and geomorphology of Sweden. In: Palouse Range in northern Idaho. - Ecology 33: 45 1-461. Varjo, U. & Tietze, W. (eds.) Norden. Man and environ­ Parker, K.C. 1988. Environmental relationships and vegetation ment. - Gebr. Borntraeger, Berlin, pp. 104-119. associates of columnar cacti in the northern Sonoran Desert. Riihl, A. 1960. Ober die Waldgesellschaften Estlands. - Ann. - Vegetatio 78: 125-140. Soc. Lit. Est. Svecia Ill: 44-55. Passarge, H. 1958. Vergleichende Betrachtungen i.iber das Riihling, A. & Tyler, G. 1986. Vegetationen i sydsvenska sozio1ogische Verhalten einiger Waldpflanzen. - Arch. ekskogar - en regional jiimforelse. - Sven. Bot. Tidskr. Forstwes. 7: 302-3 15. 80: 133- 143. Peet, R.K. & Christensen, N.L. 1988. Changes in species diver­ Runemark, H. 1950. Forest with a field layer dominated by high­ sity during secondary forest succession on the North Caro­ grown ferns. - Bot. Not. 1950: 230-233. lina piedmont. In: During, H.J., Werger, M.J.A. & Willems, Ryberg, M. 1956. Kolso, ett sormlandskt lundomrade och dess J.H. (eds.) Diversity and pattern in plant communities. ­ utveckling. - Sven. Bot. Tidskr. 50: 163-185. SPB Academic Publishing, The Hague, pp. 233-245. Ryberg, M. 1971. The deciduous woods on the Naset Peninsula

Acta Phytogeogr. Suec. 80 Deciduous fo rest vegetation in Boreo-nemoral Scandinavia 111

at Tullgarn, province of Soderrnanland. - Kungl. Sv. Skogen, A. 1971. 0kologiske og plantegeografiskeunders��Skelser Vetenskapsakad. Handl. 4 Ser. 14 (1): 1-80. i verdens nordligste ekelund. - Blyttia 29: 235-250. Ryberg, M. 1986. Parksmultronet, Fragaria moschata, en Skoglund, J. 1989. Regeneration, establishment and distribution tradgardsflykting med anpassningsforrnaga. - Sven. Bot. of Quereus robur in relation to a floodingand light gradient. Tidskr. 80: 63-80. -Stud. Plant Ecol. 18: 238-239. Rydgren, K. 1993. Herb-rich spruce forests in W Nordland, N Skre, 0. 1979. The regional distribution of vascular plants in Norway: an ecological and methodological study. - Nord. Scandinavia with requirements for high summer tempera­ J. Bot. 13: 667-690. tures. - Norw. J. Bot. 26: 295-318. Salisbury, E.J. 1936. The light climate of woodlands. - Ber. Skult, H. 1956. Skogsbotaniska studier i skargardshavet. - Schweiz. Bot. Ges. 46: 1-11. Acta Bot. Fenn. 57: 1-244. Schroder, F.G. 1983. Die therrnischen Vegetationszonen der Skye, E. 1965. Glimpses of the Bothnian coast. - Acta Erde. Ein Beitrag zur Prazisierung der geobotanischen Phytogeogr. Suec. 50: 176- 179. Terminologie. - Tuexenia 3: 31-46. Steen, E. 1958. Betesinflytelser i svensk vegetation. - Statens Schwabe, A. 1985. Monographie Alnus incana-reicher Jordbruksforsok. Medd. 89: 1-82. W aldgesellschaftenin Europa. V ariabilitat und Ahnlichkeiten Stemer, R. 1926. blands vaxtvarld. Sodra Kalmar lan Ill. ­ einer azonal verbreiteten Gesellschaftsgruppe. - Phyto­ Hj almar Appeltoffts Bokhandel, Kalmar, 237 pp. coenologia 13: 197-302. Stemer, R. 1955. Sydolandska lundar. In: Stemer, R. & Curry­ Seibert, P. 1969. Ober das Aceri-Fraxinetum als vikariierende Lindahl, K. (eds.) Natur pa bland. - Bokforlaget Svensk Gesellschaft des Galio-Carpinetum am Rande der Bayer­ Natur, Stockholm, pp. 311-325. ischen Alpen. - Vegetatio 17: 165-175. Stemer, R. 1986. blands karlvaxtflora (ed. by Lundqvist, A.). Selander, S. 1955. Det levande landskapet i Sverige. - Albert 2nd rev. ed. of "Flora der Insel bland. - Acta Phytogeogr. Bonniers Forlag, Stockholm, 492 pp. Suec. 9: 1-169". - SBT-forlaget, Lund, 400 pp. Siegel, S. & Castellan, N.J. 1988. Nonparametric statistics for St!ISrmer, P. 1938. Vegetationsstudien auf der Insel Ha��Sya im the behavioral sciences. 2nd ed. - McGraw-Hill Book Co., Oslofjord unter besonderer Berticksichtigung der GefaB­ New York, 399 pp. pflanzen und Moose. - Skr. Nor. Vidensk.-Akad. Oslo. I. Sjobeck, M. 1931. Det aldre kulturlandskapet i Sydsverige. ­ Mat.-Naturvidensk. Kl. 1938 (9): 1-155. Sven. Skogsvardsfor. Tidskr. 29: 45-73. Stoutjesdijk, Ph. & Barkman, J.J. 1992. Microclimate, vegeta­ Sjobeck, M. 193 3. Lovangskulturen i Sydsverige. Dess uppkomst, tion and fauna. - OPULUS Press, Uppsala, 216 pp. utveckling och tillbakagang. - Ymer 53: 33-66. Sykes, M.T. & Prentice, I.C. 1993. Modelling the response of Sjogren, E. 1961. Epiphytische Moosvegetation in Laubwaldem Nordic forests to climatic change. In: Holten, J.I. & Paulsen, der lnsel bland. - Acta Phytogeogr. Suec. 44: 1-149. G. & Oechel, W.C. (eds.) Impacts of climatic change on Sjogren, E. 1964. Epilithische und epigaische Moosvegetation natural ecosystems, with emphasis on boreal and arctic/ in Laubwaldem der Insel bland. - Acta Phytogeogr. Suec. alpine areas. - NINA, Trondheim, pp. 75-80. 48: 1-184. Tamm, C.O. 1972. Survival and flowering of perennial herbs. Sjogren, E. 1991. Mossfloraoch mossvegetation pa Kullaberg. m. The behaviour of Primula veris on permanent plots. - - Kullabergs Natur, Molle, 112 pp. Oikos 23: 159-166. Sjogren, E. in press. Changes in the epilithic and epiphytic moss Tamm, C.O. & Hallbacken, L. 1988. Changes in soil acidity cover in two deciduous forest areas on the island of bland from the 1920s to the 1980s in two forest areas with different (Sweden) - a comparison between 1958-1962 and 1988- acid deposition. - Ambio 17: 56-61. 1990. - Stud. Plant Ecol. 19. Tansley, A.G. 1920. The classification of vegetation and the Sjogren, E., Tidigs, A., Akerlind, L.-0. & Ekstam, U. 1974. concept of development. - J. Ecol. 8: 118-149. Lovskogsproj ektet. Slutredogorelse 7-45172. Metodforskning Tapio, S. 1953. Untersuchungen tiber die Hainvegetation und for bevarande av olandska lOvskogstyper. - Vaxtbiol. Inst. die okologische Verteilung der Hainpflanzen im rnittleren Uppsala, 35 pp. + 29 fig. (part 1) +maps (part 2). Mimeogr. Teil des Hainzentrums von Pirkkala in Stidfinnland. - Ann. Sjors, H. 1954. Siatterangar i Grangarde Finnmark. - Acta Bot. Soc. Zool. Bot. Fenn. Vanamo 25 (3): 1-57. Phytogeogr. Suec. 34: 1-135. Tapper, P.-G. 1992. On ash (Fraxinus excelsior). - Doctoral Sjors, H. 1960. Karlvaxtflora och vegetationstyper vid Anger­ thesis, Stockholm. manalven mellan Namforsen och Moforsen. - Sven. Bot. ter Braak, C.J.F. 1987. CANOCO - a Fortran program for Tidskr. 54: 121-175. canonical community ordination by (partial) (detrended) Sjors, H. 1961. Some chemical properties of the humus layer in (canonical) correspondence analysis, principal component Swedish natural soils. - K. Skogshogsk. Skr. 37: 1-51. analysis and redundancy analysis (version 2. 1). - Agricul­ Sjors, H. 1963. Arnphi-Atlanticzonation, Nemoral to Arctic. In: tural Mathematics Group, Wageningen, 95 pp. Love, A. & Love, D. (eds.) North Atlantic biota and their ter Braak, C.J.F. 1990. Update notes: CANOCO version 3.10. history. - Pergamon Press, Oxford, pp. 109-125. - Agricultural Mathematics Group, Wageningen, 35 pp. Sjors, H. 1965a. Features ofland and climate. - Acta Phytogeogr. ter Braak, C.J.F. & Barendregt, L.G. 1986. Weighted averaging Suec. 50: 1-12. of species indicator values: its efficiency in environmental Sjors, H. 1965b. Forest regions. - Acta Phytogeogr. Suec. 50: calibration. - Math. Biosci. 78: 57-72. 48-63. Thompson, K., Hodgson, J.G., Grime, J.P., Rorison, I.H., Band, Sjors, H. 1967. Nordisk vaxtgeografi. 2. uppl. - Svenska S.R. & Spencer, R.E. 1993. Ellenberg numbers revisited. ­ Bokforlaget Bonniers, Stockholm, 240 pp. Phytocoenologia 23: 277-289.

Acta Phytogeogr. Suec. 80 112 M. Diekmann

Tilman, D. 1988. Plant strategies and the dynamics and structure nutrient availability, shoot biomass and species richness in of plant communities. - Princeton University Press, grassland and wetland communities. - Vegetatio 53: 121- Princeton, 360 pp. 126. Tonteri, T., Hotanen, J.-P. & Kuusipalo, J. 1990. The Finnish Vevle, 0. 1983. Norwegian vegetation types. A preliminary forest site type approach: ordination and classificationstud­ survey of higher syntaxa. - Tuexenia 3: 169- 178. ies of mesic forest sites in southern Finland. - Vegetatio Vevle, 0. & Aase, K. 1980. Om bruk av pkologiske faktortall i 87: 85-98. norske plantesamfunn. - Nor. Vidensk. Selsk. Mus. Rapp. Tranquillini, W. 1960. Das Lichtklima wichtiger Pflanzen­ Bot. Ser. 1980 (5): 178-201. gesellschaften. - Handbuch der Pflanzenphysiologie 5 (2): Wallden, B. 1961. Misteln vid dess nordgriins. - Sven. Bot. 304-338. Tidskr. 55: 427-549. Trass, H. & Malmer, N. 1973. North European approaches to Wallin, G. 1973. Lovskogsvegetation i Sjuhiiradsbygden. - classification. - Handb. Veg. Sci. 5: 529-574. Acta Phytogeogr. Suec. 58: 1-114. Troedsson, T. & Nykvist, N. 1973. Markliira och markvard. ­ Waiter, H. 1979. Vegetation of the earth and ecological systems Almqvist & Wiksell Liiromedel AB, Stockholm, 403 pp. of the Gee-biosphere. 2nd ed. - Springer-Verlag, New Tuhkanen, S. 1980. Climatic parameters and indices in plant York, 274 pp. geography.-Acta Phytogeogr. Suec. 67: 1-110. Waiter, H. & Breckle, S.-W. 1983. Okologie der Erde. Band 1. Tuhkanen, S. 1984. A circumboreal system of climatic-phyto­ Okologische Grundlagen in g1obaler Sicht. - Gustav Fischer geographical regions. - Acta Bot. Fenn. 127: 1-50. Verlag, Stuttgart, 238 pp. Tutin, T.G., Heywood, V.H., Burges, N.A., Valentine, D.H., Waiter, H. & Lieth, H. 1960- 1967 . Klimadiagramm-Weltatlas. Waiters, S.M. & Webb, D.A. 1964- 1980. Flora Europaea. 1- - VEB Gustav Fischer Verlag, Jena. 5. - Cambridge University Press, Cambridge. Weber, H.E. 1988. Zur praktischen Anwendung des Codes der Ttixen,R. 1951.Eindrticke wahrend der pflanzengeographischen pflanzensoziologischen Nomenklatur und VorschHige zur Exkursion durch Stid-Schweden. - Vegetatio 3: 149-171. Erganzung der Regeln. - Tuexenia 8: 383-392. Tyler, G. 1989. Interacting effects of soil acidity and canopy Weimarck, H. 1947. Bidrag till Ski'mes Flora. 37 . Distribution cover on the species composition of field-layer vegetation in and ecology of Quercus petraea. - Bot. Not. 1947: 189- oaklhombeam forests. - For. Ecol. Manage. 28: 101-114. 206. van der Maarel, E. 1969. On the use of ordination models in Westhoff, V. & van der Maarel, E. 1973. The Braun-Blanquet­ phytosociology. - Vegetatio 19: 21-46. approach. - Handb. Veg. Sci. 5: 617-726. van der Maarel, E. 1979. Transformation of cover-abundance Winterhoff, W. 1965. Die Vegetation der Muschelkalkfelshange values in phytosociology and its effects on community simi­ im hessischen Werrabergland. - VerOff. N aturschutz Land­ larity. - Vegetatio 39: 97- 114. schaftspflege Baden-Wtirttemb. 33: 146-197. van der Maarel, E. 1993 . Relations between sociological-eco­ Woodward, F.I. 1987. Climate and plant distribution. - Cam­ logical species groups and Ellenberg indicator values. - bridge University Press, Cambridge, 174 pp. Phytocoenologia 23: 343-362. Zedler, P.H. & Goff, F.G. 1973. Size association analysis of van der Poel, A.J. & Stoutjesdijk, Ph. 1959. Some rnicro­ forest successional trends in Wisconsin. - Ecol. Monogr. climatological differences between an oak wood and a 43: 79-94. Calluna heath. - Meded. Landbouwhogesch. Wageningen Zoller, H. 1956. Die nattirliche GroBgliederung der fenno­ 59 (2): 1-8. skandischen Vegetation und Flora. - Ber. Geobot. Inst. Vermeer, J.G. & Berendse, F. 1983. The relationship between Eidg. Techn. Hochsch. Stift. Rtibel 1955: 74-98.

Acta Phytogeogr. Suec. 80 Svenska Viixtgeografiska Siillskapet 113

SVENSKA V Ax:TGEOGRAFISKA SALLSKAPET SOCIETAS PHYTOGEOGRAPHICA SUECANA

Adress: Viixtbiologiska Institutionen, Villaviigen 14, S-752 36 Uppsala, Sweden

Sallskapet har till andamal att vacka och underhalla intresse for The object of the Society is to promote investigations in flora vaxtgeografieni vidstrackaste mening, att framj a utforskande av and vegetation, their history and their ecological background. flora och vegetation i Sverige och andra lander och att havda Through publication of monographs, and other activities, the geobotanikens praktiska och vetenskapliga betydelse. society tries to stimulate geobotanical research and its applica­ Sallskapet anordnar sammankomster och exkursioner samt tion to practical and scientific problems. utger tva publikationsserier. Medlemskap kan erhilllas efter Individual members and subscribers (societies, institutes, li­ anmalan hos sekreteraren. Foreningar, bibliotek, laroanstalter braries, etc.) receive Acta Phytogeographica Suecica on pay­ och andra institutioner kan inga som abonnenter. ment of the annual membership fee. There are additional fees in Sallskapet utger arligen Acta Phytogeographica Suecica. years when more than one volume is issued. For membership Medlemmar och abonnenter erhaller arets Acta mot postfOrskott please, apply to the Secretary. pa arsavgiften jamte porto och expeditionskostnader. The Society also issues Studies in Plant Ecology (vols. 1-18 Vissa ar utges extra band av Acta, som erhalls mot en tillaggs­ Vaxtekologiska studier), which appears irregularly and can be avgift. ordered through OPULUS PRESS AB, Box 25 137, 750 25 Sallskapet utger ocksa den ickeperiodiska serien Studies in Uppsala, Sweden or by standing order. Plant Ecology (vols. 1-18 Vaxtekologiska studier). Den kan forvarvas efter bestallning ell er genom staende abonnemang hos Sallskapet.

ACTA PHYTOGEOGRAPHICA SUECICA

1. E. Almquist. 1929. Upplands vegetation och flora. (Veg­ 11. N. Stalberg. 1939. Lake Vattem. Outlines of its natural etation and flora of Uppland.) Out of print. history, especially its vegetation. ISBN 91-7210-011-7. 2. S. Th unmark. 1931. Der See Fiolen und seine Vegetation. Price: 160 SEK. ISBN 91-7210-002-8. Price: 240 SEK. 12. G. E. Du Rietz, A. G. Hannerz, G. Lohammar, R. Santesson 3. G. E. Du Rietz. 1931. Life-forms of terrestrial flowering & M. W� rn. 1939. Zur Kenntnis der Vegetation des Sees plants. I. ISBN 91-721 0-003-6. Price: 160 SEK. Tilkern. ISBN 91-7210-0 12-5. Price: 160 SEK. 4. B. Lindquist. 1932. Om den vildvaxande skogsalmens 13. Vaxtgeografiska studiertillagnade Carl Skottsberg pasextio­ raser och deras utbredning i Nordvasteuropa. (Summary: arsdagen 1/12 1940. (Geobotanical studies dedicated to C. The races of spontaneous Ulmus glabra Huds. and their Skottsberg.) 1940. ISBN 91-7210-0 13-3. Price: 290SEK. distribution in NW Europe.) Out of print. 14. N. Hy lander. 1941. De svenska formema av Mentha gentilis 5. H. Osvald. 1933. Vegetation of the Pacific coast bogs of L. coil. (Zusammenfassung: Die schwedischen Formen der North America. ISBN 91-7210-005-2. Price: 160 SEK. Mentha gentilis L. sensu coli.) ISBN 91-7210-014-1. 6. G. Samuelsson. 1934. Die Verbreitung der hoheren Wasser­ Price: 160 SEK. pflanzen in Nordeuropa. 1934. Out of print. 15. T. E. Hasselrot. 1941. Till kannedom om nagra nordiska 7. G. Degelius. 1935. Das ozeanische Element der Strauch umbilicariaceers utbredning. (Zusammenfassung: Zur und Laubflechtenflora von Skandinavien. Out of print. Kenntnis der V erbreitung einiger U mbilicariaceen in 8. R. Sernander. 1936. Granskar och Fiby urskog. En studie Fennoscandia.) ISBN 91-7210-01 5-X. Price: 240 SEK. over stormluckomas och marbuskamas betydelse i den 16. G. Samuelsson. 1943. Die Verbreitung der Alchemilla­ svenska granskogens regeneration. (Summary: The primi­ Arten aus der Vulgaris-Gruppe in Nordeuropa. ISBN 91- tive forests of Granskar and Fiby. A study of the part 7210-016-8. Price: 160 SEK. played by storm-gaps and dwarf trees in the regeneration of 17. Th. Arwidsson. 1943. Studien iiber die Gefasspflanzen in the Swedish spruce forest.) ISBN 91-7210-008-7. Price: den Hochgebirgen der Pite Lappmark. ISBN 91-7210-017- 240 SEK. 6. Price: 240 SEK. 9. R. Stemer. 1938. Flora der Insel bland. Die Areale der 18. N. Dahlbeck. 1945. Strandwiesen siidostlichen Ore­ am Gefasspflanzen Glands nebst Bemerkungen zu ihrer sund. (Summary: Salt marshes on the S. E. coast of Ore­ Oekologie und Soziologie. ISBN 91-7210-009-5. Out of sund.) ISBN 91-7210-01 8-4. Price: 160 SEK. print. 19. E. von Krusenstjerna. 1945. Bladmossvegetation och blad­ 10. B. Lindquist. 1938. Dalby Soderskog. En skansk lovskog i mossflora i Uppsalatrakten. (Summary: Moss flora and fomtid och nutid. (Zusammenfassung: Ein Laubwald in moss vegetation in the neighbourhood of Uppsala.) ISBN Schonen in der V ergangenhei t und Gegenwart.) ISBN 91- 91-7210-019-2. Price: 290 SEK. 7210-010-9. Price: 240 SEK.

Acta Phytogeogr. Suec. 80 114 Svenska Viixtgeografiska Sdllskapet

20. N. Albertson. 1946. bsterplana hed. Ett alvaromrade pa the Langan drainage area, Jamtland, Sweden.) ISBN 91- Kinnekulle. (Zusammenfassung: bsterplana hed. Ein 7210-036-2. Price: 240 SEK. Alvargebiet auf dem Kinnekulle.) ISBN 91-721 0-020-6. 37. M.-B. Florin. 1957. Plankton of fresh and brackish waters Price: 240 SEK. in the Soderta.Ije area. ISBN 91-7210-037-0. Price: 160 21. H. Sj Ors. 1948. Myrvegetation i Bergslagen. (Summary: SEK. Mire vegetation in Bergslagen, Sweden.) ISBN 91-7210- 38. M.-B. Florin. 1957. lnsjostudier i Mellansverige. Mikro­ 021-4. Price: 290 SEK. vegetation och pollenregn i vikar av bstersjobackenet och 22. S. Ahlner. 1948. Utbredningstyper bland nordiska barrtrads­ insjoar fran preboreal tid till nutid. (Summary: Lake stud­ lavar. (Zusammenfassung: Verbreitungstypen unter fenno­ ies in Central Sweden. Microvegetation and pollen rain in skandischen Nadelbaumflechten.) ISBN 91-7210-022-2. inlets of the Baltic basin and in lakes from Preboreal time Price: 240 SEK. to the present day.) 1957. ISBN 91-7210-038-9. Price: 23 . E. Julin. 1948. Vessers udde, mark och vegetation i en 160 SEK. igenvaxande lOvang vid Bjarka-Saby. (Zusammenfassung: 39. M. Fries. 1958. Vegetationsutveckling och odlingshistoria Vessers udde. Boden und Vegetation in einer verwachsen­ i V amhemstrakten. En pollenanalytisk undersokning i Vas­ den Laubwiese bei Bjarka-Saby in bstergotland, tergotland. (Zusammenfassung: Vegetationsentwicklung Siidschweden.) ISBN 91-7210-023-0. Price: 240 SEK. und Siedlungsgeschichte im Gebiet von Varnhem. Eine 24. M. Fries. 1949. Den nordiska utbredningen av Lactuca pollenanalytische Untersuchung aus Vastergotland (Stid­ alpina, Aconitum septentrionale, Ranunculus platanifolius schweden).) ISBN 91-7210-039-7. Price: 160 SEK. och Polygonatum verticillatum. (Zusammenfassung: Die 40. B. Pettersson. 1958. Dynamik och konstans i Gotlands nordische Verbreitung von Lactuca alpina.) ISBN 91- flora och vegetation. (Zusammenfassung: Dynamik und 7210-24-9. Price: 160 SEK. Konstanz in der Flora und Vegetation von Gotland, 25. 0. Gjcerevoll. 1949. Sn(ISleievegetasjonen i Oviksfjellene. Schweden.) ISBN 91-7210-040-0. Price: 400 SEK. (Summary: The snow-bed vegetation of Mts Oviksfjallen, 41. E. Uggla. 1958. Skogsbrandfalt i Muddus nationalpark. Jamtland, Sweden.) ISBN 91-7210-025-7. Price: 160 (Summary: Forest fire areas in Muddus National Park, SEK. northernSweden.) ISBN 91-7210-041-9. Price: 160 SEK. 26. H. Osvald. 1949. Notes on the vegetation of British and 42. K. Thomasson. 1959. Nahuel Huapi. Plankton of some Irish mosses. ISBN 91-7210-026-5. Price: 160 SEK. lakes in an Argentina National Park, with notes on terres­ 27. S. Selander. 1950. Floristic phytogeography of southwest­ trial vegetation. ISBN 91-7210-042-7. Price: 160 SEK. em Lule Lappmark (Swedish Lapland). I. 1950. ISBN 91- 43. V. Gillner. 1960. Vegetations- und Standortsuntersuch­ 7210-027-3. Price: 240 SEK. ungen in den Strandwiesen der schwedischen Westktiste. 28. S. Selander. 1950. Floristic phytogeography of southwest­ ISBN 91-7210-043-5. Price: 240 SEK. em Lule Lappmark (Swedish Lapland). 11. Karlvaxtfloran 44. E. Sj ogren. 1961. Epiphytische Moosvegetation in Laub­ i sydvastra Lule Lappmark. (Summary: Vascular flora.) waldem der Insel bland, Schweden. (Summary: Epiphytic ISBN 91-7210-028- 1. Price: 160 SEK. moss communities in deciduous woods on the island of 29. M. Fries. 195 1. Pollenanalytiska vittnesbord om senkvartar bland, Sweden.) ISBN 917210-044-3 (ISBN 91-7210- vegetationsutveckling, sarskilt skogshistoria, i nordvastra 444-9). Price: 160 SEK. Gotaland. (Zusammenfassung: Pollenanalytische Zeugnisse 45. G. Wistrand. 1962. Studier i Pite Lappmarks karlvaxtflora, der spatquartaren Vegetationsentwicklung, hauptsachlich med sarskild hansyn till skogslandet och de isolerade der Waldgeschichte, im nordwestlichen Gotaland, fj allen. (Zusammenfassung: Studien tiber die Gefasspflan­ Stidschweden.) ISBN 91-7210-029-X. Price: 240 SEK. zenflora der Pite Lappmark mit besonderer Bertick­ 30. M. Wcern. 1952. Rocky-shore algae in the bregrund Archi­ sichtigung des Waldlandes und der isolierten niederen pelago. ISBN 91-7210-030-3. Price: 290 SEK. Fj elde.) ISBN 91-7210-045-1 (ISBN 91-7210-445 -7). 31. 0. Rune. 1953. Plant life on serpentines and related rocks Price: 240 SEK. in the North of Sweden. 1953. ISBN 91-7210-03 1-1. Price: 46. R. lvarsson. 1962. Lovvegetation i Mollosunds socken. 240 SEK. (Zusammenfassung: Die Laubvegetation im Kirchspiel 32. P. Kaaret. 1953. Wasservegetation der Seen Orlangen und Mollosund, Bohuslan, Schweden.) ISBN 91-7210-046-X Trehorningen. ISBN 91-7210-032-X. Price: 160 SEK. (ISBN 91-7210-446-5). Price: 160 SEK. 33. T. E. Hasselrot. 1953. Nordliga lavar i Syd- och Mellan­ 47. K. Thomasson. 1963. Araucanian Lakes. Plankton studies sverige. (Nordliche Flechten in Slid- und Mittelschweden.) in North Patagonia, with notes on terrestrial vegetation. ISBN 91-7210-033-8. Price: 240 SEK. ISBN 91-7210-047-8. Price: 240 SEK. 34. H. Sj ors. 1954. Slatterangar i Grangarde Finnmark. (Sum­ 48. E. Sj ogren. 1964. Epilitische und epigilische Moosvege­ mary: Meadows in Grangarde Finnmark, SW Dalama, tation in Laubwa.Idem der Insel bland, Schweden. (Sum­ Sweden.) ISBN 91-7210-034-6. Price: 160 SEK. mary: Epilithic and epigeic moss vegetation in deciduous 35. S. Kilander. 1955. Karlvaxtemas ovre granser pa fj all i woods on the island of bland, Sweden.) ISBN 91-7210- sydvastra Jamtland samt angransande delar av Harjedalen 048-6 (ISBN 91-7210-448- 1). Price: 240 SEK. och Norge. (Summary: Upper limits of vascular plants on 49. 0. Hedberg. 1964. Features of afroalpine plant ecology. mountains in southwesternJamtland and adj acent parts of (Resume fran�ais.) ISBN 91-7210-049-4 (ISBN 91-7210- Harjedalen (Sweden) and Norway.) ISBN 91-7210-035-4. 449-X). Price: 240 SEK. Price: 240 SEK. 50. The Plant Cover of Sweden. A study dedicated to G. Einar 36. N. Quennerstedt. 1955. Diatomeema i Langans sj o­ Du Rietz on his 70th birthday by his pupils. 1965. ISBN vegetation. (Summary: Diatoms in the lake vegetation of 91-7210-050-8. Price: 41 12 SEK.

Acta Phytogeogr. Suec. 80 Svenska Viixtgeografiska Siillskapet 115

51. T. Flensburg. 1967. Desmids and other benthic algae of den. ISBN 91-7210-066-4 (ISBN 91-7210-466-X). Price: Lake Kavsjon and Store Mosse. SW Sweden. ISBN 91- 240 SEK. 7210-05 1-6 (ISBN 91-7210-451-1). Price: 240 SEK. 67. S. Tuhkanen. 1980. Climatic parameters and indices in 52. E. Skye. 1968. Lichens and air pollution. A study of crypto­ plant geography. ISBN 91-7210-067-2 (ISBN 91-7210- gamic epiphytes and environment in the Stockholm region. 467 -8). Price: 240 SEK. ISBN 91-721 0-052-4 (ISBN 91-72 10-452-X). Price: 240 68. Studies in plant ecology dedicated to Hugo Sjors. E. Sjogren SEK. (ed.) 1980. ISBN 91-7210-068-0 (ISBN 91-721 0-468-6). 53. J. Lundqvist. 1968. Plant cover and environment of steep Price: 290 SEK. hillsides in Pite Lappmark. (Resume: La couverture vege­ 69. C. Nilsson. 1981. Dynamics of the shore vegetation of a tale et l'habitat des flancs escarpes des collines de Pite North Swedish hydro-electric reservoir during a 5-year Lappmark.) ISBN 91-7210-053-2 (ISBN 91-7210-453-8). period. ISBN 91-7210-069-9 (ISBN 91-7210-469-4). Price: Price: 240 SEK. 240 SEK. 54. Conservation of Vegetation in Africa South of the Sahara. 70. K. Warenberg. 1982. Reindeer forage plants in the early Proceedings of a symposium held at the 6th Plenary meet­ grazing season. Growth and nutritional content in relation ing of the AETFAT, Uppsala Sept. 12-16, 1966. 1968. to climatic conditions. ISBN 91-72 10-070-2 (ISBN 91- ISBN 91-7210-054-0 (ISBN 91-721 0-454-6). Price: 290 721 0-470-8). Price: 240 SEK. SEK. 71. C. Johansson. 1982. Attached algal vegetation in running 55. L. -K. Konigsson. 1968. The Holocene history of the Great waters of Jamtland, Sweden. ISBN 917210-07 1-0 (ISBN Alvar of bland. ISBN 91-7210-055-9 (ISBN 91-7210- 91-7210-47 1-6). Price: 240 SEK. 455-4). Price: 290 SEK. 72. E. Rosen. 1982. Vegetation development and sheep graz­ 56. H. P. Hallberg. 1971. Vegetation auf den Schalenablager­ ing in limestone grasslands of South bland, Sweden. ISBN ungen in Bohuslan, Schweden. (Summary: Vegetation on 91-7210-072-9 (ISBN 91-721 0-472-4). Price: 290 SEK. shell deposits in Bohuslan, Sweden.) ISBN 91-7210-056- 73. Zhang Liquan. 1983. Vegetation ecology and population 7 (ISBN 91-7210-456-2). Price: 240 SEK. biology of Fritillaria meleagris L. at the Kungsangen Na­ 57. S. Fransson. 1972. Myrvegetation i sydvastra Varmland. ture Reserve, eastern Sweden. ISBN 91-7210-073-7 (ISBN (Summary: Mire vegetation in southwestern Varmland, 91-721 0-473-2). Price: 240 SEK. Sweden.) ISBN 91-7210-057-5 (ISBN 91-7210-457-0). 74. I. Backeus. 1985. Aboveground production and growth Price: 240 SEK. dynamics of vascular bog plants in central Sweden. ISBN 58. G. Wallin. 1973. Lovskogsvegetation i Sjuharadsbygden. 91-7210-074-5 (ISBN 91-7210-474-0). Price: 240 SEK. (Summary: Deciduous woodlands in Sjuharadsbygden, 75. E. Gunnlaugsd6ttir. 1985. Composition and dynamical Vastergotland, southwestern Sweden.) ISBN 91-7210- status of heathland communities in Iceland in relation to 058-3 (ISBN 91-72 10-458-9). Price: 240 SEK. recovery measures. ISBN 91-7210-07 5-3 (ISBN 91-7210- 59. D. Johansson. 1974. Ecology of vascular epiphytes in 475-9). Price: 240 SEK. West African rain forest. (Resume: Ecologie des epiphytes 76. Plant cover on the limestone Alvar on bland. Ecology­ vasculaires dans la foret dense humide d' Afrique Sociology-Taxonomy. E. Sjogren (ed.) 1988. ISBN 91- occidentale.) ISBN 91-7210059-1 (ISBN 91-7210-459- 7210-076- 1 (ISBN 91-72 1 0-476-7). Price: 320 SEK. 7). Price: 290 SEK. 77. A. H. Bj arnason. 1991. Vegetation on lava fields in the 60. H. Olsson. 1974. Studies on South Swedish sand vegeta­ Hekla area, Iceland. ISBN 91-7210-077-X (ISBN 91- tion. ISBN 91-7210-060-5 (ISBN 91-7210-460-0). Price: 721 0-477-6). Price: 290 SEK. 240 SEK. 78. /. Wallentinus & P. Snoeijs (eds.). 1992. Algological 61. H. Hytteborn. 1975. Deciduous woodland at Andersby, studies of Nordic coastal waters - A festschrift to Prof. eastern Sweden. Above-ground tree and shrub production. Mats Wrem on his 80th birthday-. ISBN 91-7210-078-8 ISBN 91-7210-061-3 (ISBN 91-7210-461-9). Price: 240 (ISBN 91-7210-478-3 ). Price: 290 SEK. SEK. 79. TamratBekele. 1993. Vegetation ecology ofrernnant Afro­ 62. H. Pe rsson. 197 5. Deciduous woodland at Andersby, east­ montane forests on the Central Plateau of Shewa, Ethiopia. em Sweden. Field-layer and below-ground production. ISBN 91-721 0-079-6 (ISBN 91-721 0-479-1). Price: 290 ISBN 91-7210-062-1 (ISBN 91-721 0-462-7). Price: 160 SEK. SEK. 80. M. Diekmann. 1994. Deciduous forest vegetation in Boreo­ 63. S. Brakenhielm. 1977. Vegetation dynamics of afforested nemoral Scandinavia. ISBN 91-7210-080-X (ISBN 91- farmland in a district of south-eastern Sweden. ISBN 91- 7210-480-5). Price: 290 SEK. 721 0-063-X (ISBN 91-7210-463-5). Price: 240 SEK. 64. M. Y. Ammar. 1978. Vegetation and local environment on A limited number of cloth-bound copies of Acta 44, 45, 46, 48, shore ridges at Vickleby, bland, Sweden. An analysis. 49, 51, 52, 53, 56, 57, 61, 63, 66, 67, 68, 69, 70, 71, 72, 73, 74, ISBN 91-721 0-064-8 (ISBN 91-72 10-464-3). Price: 240 75, 76, 77, 78, 79 and 80 is available at an additional cost of 75 SEK. SEK per volume. (Use ISBN n°s. within brackets to order.) 65. L. Kullman. 1979. Change and stability in the altitude of the birch tree-limit in the southern Swedish Scandes 1915- 1975. ISBN 91-7210-065-6 (ISBN91-7210-465-1). Price: 240 SEK. 66. E. Wa ldemarson Jens en. 1979. Successions in relationship to lagoon development in the Laitaure delta, North Swe-

Acta Phytogeogr. Suec. 80 116 Svenska Viixtgeografiska Siillskapet

STUDIES IN PLANT ECOLOGY (VOL. 1-18)

I. S. Brakenhielm & T. IngelOg. 1972. Vegetationen i Kungs­ 808-8. Price: 160 SEK. hamn-Morga naturreservat med forslag till skotselplan. 9. J. Lundqvist & G. Wistrand. 1976. Strandflora inom ovre (Summary: Vegetation and proposed management in the och mellersta Skelleftealvens vattensystem. Med en sam­ Kungshamn-Morga Nature Reserve south of Uppsala.) manfattning betdiffande botaniska skyddsvarden. (Sum­ ISBN 91-7210-801 -0. Price: 112 SEK. mary: Riverside vascular flora in the upper and middle 2. T. IngelOg & M. Risling. 1973. Kronparken vid Uppsala, catchment area of the River Skelleftealven, northern Swe­ historik och bestandsanalys av en 300-arig tallskog. (Sum­ den.) ISBN 91-721 0-809-6. Price: 112 SEK. mary: Kronparken, history and analysis of a 300-year old 10. A. Miiller-Haeckel. 1976. Migrationsperiodik einzelliger pinewood near Uppsala, Sweden.) ISBN 91-7210-802-9. Algen in Fliessgewassern. ISBN 91-7210-810-X. Price: Price: 112 SEK. 112 SEK. 3. H. Sjors et al. 1973. Skyddsvarda myrar i Kopparbergs lan. 11. A. Sj odin. 1980. Index to distribution maps of bryophytes [Summary: Mires considered for protection in Kopparberg 1887- 1975. I. Musci. (hard-bound). ISBN 91-7210-811-8. County (Prov. Dalama, Central Sweden.)] ISBN 91-7210- Price: 160 SEK. 803-7. Price: 112 SEK. 12. A. Sj odin. 1980. Index to distribution maps of bryophytes 4. L. Karlsson. 1973. Autecology of cliff and scree plants in 1887-1975. 11. Hepaticae. (hard-bound). ISBN 91-72 10- Sarek National Park, northern Sweden 1973. ISBN 91- 812-6. Price: 112 SEK. 7210-804-5. Price: 160 SEK. 13. 0. Eriksson, T. Palo & L. Soderstrom. 1981. Renbetning 5. B. Klasvik. 1974. Computerized analysis of stream algae. vintertid. Undersokningar rorande svensk tamrens narings­ ISBN 91-7210-805-3. Price: 112 SEK. ekologi under snoperioden. ISBN 91-7210-81 3-4. Price: 6. Y. Dahlstrom-Ekbohm. 1975. Svensk miljovards- och 112 SEK. omgivningshygienlitteratur 1952-1 972. Bibliografi och 14. G. Wistrand. 1981. Bidrag till Pite lappmarks vaxtgeo­ analys. ISBN 91-7210-806-1. Price: 112 SEK. grafi. ISBN 91-7210-814-2. Price: 112 SEK. 7. L. Rodenborg. 1976. Bodennutzung, Pflanzenwelt und 15. T. Karlsson. 1982. Euphrasia rostkoviana i Sverige. ISBN ihre Veranderungen in einem alten Weidegebiet auf Mit­ 91-7210-81 5-0. Price: 160 SEK. tel-Oland, Schweden. ISBN 91-7210-807-X. Price: 112 16. Theory and Models in Vegetation Science: Abstracts. ISBN SEK. 91-7210-816-9. 1985. Price: 160 SEK. 8. H. Sj ors & Ch. Nilsson. 1976. Vattenutbyggnadens effek­ 17. I. Backeus. 1988. Mires in the Thaba-Putsoa Range of the ter pa levande natur. En faktaredovisning overvagande Maloti, Lesotho. ISBN 91-7210-817-7. Price: 160 SEK. fran Umealven. (Summary: Bioeffects of hydroelectric 18. E. Sj ogren ( ed. ). 1989. Forests of the world - diversity and development. A case study based mainly on observations dynamics (Abstracts) 1989. ISBN 91-7210-818-5. Price: along the Ume River, northern Sweden.) ISBN 91-7210- 290 SEK.

Distributor: OPULUS PRESS AB Box 25 137, S-750 25 Uppsala, Sweden Tel. + 46 18 320662 Fax + 46 18 321 368

Acta Phytogeogr. Suec. 80

Distributor: OPULUS PRESS AB, Uppsala, Sweden

ISBN 91-7210-080-X (ISBN 91-7210-480-5) ISSN 0084-5914